“IIIIIII'IIIIIIIIIIIIIHIIII' .....-

 

lllllllllllllllllllllllllllllllllllllll

3 1293 0177

LIBRARY
Michigan State
. University

 

 

 

 

This is to certify that the
dissertation entitled

THE EFFECTS OF SOCIOECONOMIC STATUS, ETHNICITY AND
NUTRITIONAL STATUS ON THE GROWTH AND PHYSICAL
FITNESS OF 10 YEAR OLD SOUTH AFRICAN BOYS

presented by

Reshma Babra Naidoo

has been accepted towards fulﬁllment
of the requirements for

Ph.D. degreein Kinesioiogy

 

Wm

Major professor

Date 7/11/9 9

MSU it an Afﬁrmative Action /Equal Opportuniry Institution 0- 12771

 

 

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE I DATE DUE I

 

DATE DUE
. m “181‘"?

 

Jilli in 821193

 

WWW;

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1!” WWW“

THE EFFECTS OF SOCIOECONOMIC STATUS, ETHNICITY, AND
NUTRITIONAL STATUS ON THE GROWTH AND PHYSICAL FITNESS
OF 10 YEAR OLD SOUTH AFRICAN BOYS

by

Reshma Babra Naidoo

A DISSERTATION
Submitted to
Michigan State University
inlpartial fulﬁllment of the requirements

for the degree of
DOCTOR OF PHILOSOPHY

Department of Kinesiology
1999

ABSTRACT
THE EFFECTS OF SOCIOECONOMIC STATUS, ETHNICITY, AND
NUTRITIONAL STATUS ON THE GROWTH AND PHYSICAL FITNESS 0F 10
YEAR OLD SOUTH AFRICAN BOYS
by

Reshma Babra Naidoo

The interrelationships among growth, physical ﬁtness, habitual physical activity,
SES, race and nutritional status of 389 ten year old South African boys attending 18
schools in the greater Johannesburg area were considered in this study. Boys were
classiﬁed by race using anthroscopic appraisal, and all boys of mixed ancestry (so-called
“Colored”) were excluded from the sample. Boys of African (n = 208) and Indian (n =
137) ancestry were stratiﬁed into higher (n = 107, n = 99, respectively) and lower (n =
101, n = 38, respectively) 8158 groups. All boys of European ancestry (n=42) were
classiﬁed as higher 8158. Data for each subject included anthropometry, physical ﬁtness,
3-day physical activity records, habitual physical activity questionnaires, and
sociodemographic data. Nutritional status was based on body mass index and was
classiﬁed as undernourished, nutritionally normal or overweight.

Race accounts for a small percentage of the variation in body proportions (sitting
height/stature, 2%; and bicristal/biacromial, 4%) and fatness (sum of ﬁve skinfolds, 2%).
Body mass, the sitting height/stature ratio, fatness (sum of ﬁve skinfolds) and relative fat
distribution (trunk/extremity skinfold ratio) differ signiﬁcantly by nutritional status when
race and SES are statistically controlled. This difference may be confounded by racial

variation since there are proportionally more overweight and undernourished Indian boys

compared to African or European boys. SES differences in the bicristal/biacromial ratio,
mass and stature are signiﬁcant when the effects of race and nutritional status are
controlled, but may be confounded by race. All European boys are of higher SES.

Race accounts for a small percentage of the variation (1% to 5%) in the push-ups,
9 minute run, sit and reach, and grip strength when the effects of race and nutritional
status are controlled. Nutritional status accounts for a minimal amount of the variation in
the sit-ups and dash (1%), but for a larger proportion of the variance in absolute grip
strength (11%) and grip strength per unit body mass (22%). SES contributes an
additional 2% to the variation in absolute grip strength.

Race accounts for little of the variance in estimated energy expenditure
(kcal/kg/day) in sedentary (4%) and moderate to vigorous (2%) physical activity
(MVPA). There are signiﬁcant nutritional status differences in MVPA and total
estimated energy expenditure (kcal/kg/day) among Indian boys when the effects of SES
are controlled, and in light activity for European boys.

The results of this study indicate that a relatively small proportion of variation in
growth and physical ﬁtness is related to race, whereas a signiﬁcant percentage of
variation in energy expenditure is associated with nutritional status. SES does not
account for a signiﬁcant amount of the variation in growth, physical ﬁtness or estimated
energy expenditure of 10 year old South African boys of African, Indian and European
ancestry. Thus, in this sample of South African boys, most of the variation in measures

of growth and physical ﬁtness is related to factors other than race and SES.

Acknowledgments

I would like to express my gratitude to the many individuals who assisted me
through the long and arduous process of writing my dissertation. Writing this
dissertation was an interesting experience and was completed in three places - Austin,
Johannesburg (South Africa) and East Lansing.

First, I would like to thank the students and staff of the eighteen schools in
Gauteng, South Africa, at which I gathered data - for their hospitality and
accommodating me through my data collection. The 428 boys who graciously subjected
themselves to hours of poking, prodding, pinching and prying were wonderfully helpful
and cooperative. Samantha Helmey, Hillary Helmey, Margaret Sibiya, and Thara Rama,
my research assistants, worked through rain and hail to complete data collection, and
without whom this would have been a much, much longer project. My sister, Valerie
Ramsingh, and her family who provided me with a place to stay, a car and childcare and
support for the three months that I spent in South Africa collecting data, thank you.

Dr Robert Malina, who stood by me through all those years, especially when I
needed a little help just to get through the day, for providing me with an immense
education and always challenging me to go that extra yard, thank you. I would like to
thank my dissertation committee members, Crystal Branta, Jim Pivarnik and David

Wiley for their support, understanding and helpful comments in, writing this dissertation.

iv

Rosane Rosendo Da Silva provided me with my home away from home, and this is
greatly appreciated. .

Finally, I would like to thank my incredible family for their selﬂess and enduring
support and conﬁdence in me. My husband, Suren Naidoo, has been extremely
supporting, patient and understanding throughout this long process. My children,
Udbhav and Udipta, who had to learn the word dissertation almost as soon as they could
speak, waited patiently for me to complete my dissertation. To my mom, Vidya
Ramsingh, and mother in law, Sheila Naidoo, who took turns traveling to Austin from
South Africa to be surrogate moms to my family while I went off to Michigan, thank

you.

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... viii

List of Figures ................................................................................................................... xii

Chapter 1

INTRODUCTION .............................................................................................................. 1
Purposes .......................................................................................................................... 3
Research Question ........................................................................................................... 4
Research Questions ......................................................................................................... 4
Deﬁnition of terms .......................................................................................................... 5
Limitations ...................................................................................................................... 7
Scope of Study ................................................................................................................ 9
Signiﬁcance of Study .................................................................................................... 10

Chapter 2

REVIEW OF LITERATURE ........................................................................................... 12
South Africa: geographic, demographic and sociopolitical background ...................... 12
Growth and physique .................................................................................................... 20
Physical ﬁtness .............................................................................................................. 47
Habitual Physical Activity ............................................................................................ 66

Chapter 3

METHODS ....................................................................................................................... 78
South African schools ................................................................................................... 78
Subjects ......................................................................................................................... 79
Procedures ..................................................................................................................... 84
Statistical Analyses ....................................................................................................... 99

Chapter 4

RESULTS ...................................................................... . ............................................. 103
Descriptive Statistics ................................................................................................... 103
Research Question One ............................................................................................... 138
Research Question Two .............................................................................................. 143
Research Question Three .............................................. . ........................................... 152
Research Q estion F our ............................................................................................... 156
Research Question Five .............................................................................................. 161

Chapter 5 ......................................................................................................................... 170

DISCUSSION ................................................................................................................. 170
Growth ........................................................................................................................ 170
Race and SES .............................................................................................................. 171
Nutritional status: BMI-for-age .................................................................................. 177
Physical Fitness ........................................................................................................... 188

vi

Chapter 6 ......................................................................................................................... 215

SUMMARY AND CONCLUSIONS ............................. . ............................................... 215
Conclusions ................................................................................................................. 220
Recommendations. ...................................................................................................... 222

Appendixes
Appendix A: University committee on research involving human subjects ............... 227
Appendix B: Parent consent form ............................................................................... 230
Appendix C: Assent form for subjects ........................................................................ 232
Appendix D: Sociodemographic information questionnaire ...................................... 234
Appendix E: Physical activity record .......................................................................... 237
Appendix F: Habitual physical activity questionnaire ................................................ 242

Bibliography ................................................................................................................... 244

vii

Table 1.
Table 2.

Table 3.

Table 4.
Table 5.
Table 6.

Table 7.
Table 8.
Table 9.

Table 10.

Table 11.

Table 12.

Table 13.

Table 14.
Table 15.

Table 16.

Table 17.
Table 18.

Table 19.
Table 20.

Table 21.
Table 22.

Table 23.

Table 24.

LIST OF TABLES

Basic indicators of health status in South Africa. ............................................ 15
Relative changes in the economic status of South Africa over the last

decade ............................................................................................................... 17
Prevalence of stunting, underweight and wasting in children by

ethnicity. ........................................................................................................... 23
Recommended cut-off values for adolescents .................................................. 26
BMI percentiles for 10 year old US. and French boys .................................... 28
Mean heights, weights and estimated BMI from selected samples of 10

year old South African boys ............................................................................. 36
Parasitic infestation rates of hospitalized children in Cape Town .................... 42
Prevalence of infestations in school boys in Kinshasa ..................................... 44
Comparison of the mean body size and performance-related physical

ﬁtness of adolescent South African males 12 to 18 years of age by age

and race ............................................................................................................ 57
Comparison of the physical ﬁtness (means) of adolescent South

Aﬁican boys by race ........................................................................................ 59
Comparison of age, body size and physical performances (means) of
adolescent European, ‘Colored’ and African Sonth African males

between 1966 and 1985 .................................................................................... 61
Absolute and relative 11702 M of normally nourished, low weight-for-

age and low-weight-for-height 6 to 16 year old Colombian boys .................... 64
Oxygen consumption of migrant and well-to-do 12 year old Brazilian

boys .................................................................................................................. 65
Absolute and relative 1{"02 max of boys with a mean age of 10.5 years. ........... 66
Estimated activity level (SD) of South African European, “Colored”

and African adolescent males .......................................................................... 78
Income of the head of household per month and per year in South

African Rands and US. Dollars ....................................................................... 82
Total sample by race and SES. ........................................................................ 83
Inna-observer technical errors of measurement in the present study

(n=42) compared to those for several other studies. ........................................ 89
Within day correlations of ﬁtness tests of trials one and three for 42

boys with comparative data for three other samples. ....................................... 93
Between day correlations of ﬁtness tests taken four days apart with
comparative data for Brazilian children. .......................................................... 94
Activity Code Sheet (after Bouchard et al., 1983). .......................................... 97
Means, SD, correlations and AN OVA for observed and recalled

estimated energy expenditure (kcal/day) in physical activity of 57 boys. ....... 98
Sample Sizes, means, standard deviations, ranges and medians (where
appropriate) for all variables in the total sample. ........................................... 104
Descriptive statistics by race .......................................................................... 106

vii1

1' 3:”
I.
'-

,(37.

1"

Table 25.
Table 26.
Table 27.
Table 28.
Table 29.
Table 30.
Table 31.
Table 32.
Table 33.
Table 34.
Table 35.

Table 36.

Table 37.
Table 38.
Table 39.

Table 40.

Table 41.
Table 42.

Table 43.

Table 44.

Table 45.

Results of AN OVA (F ratios) and Scheffé post hoc tests for mean

differences (MD) between races ..................................................................... 108
Descriptive statistics for Aﬁican boys by SES. ............................................. 113
Descriptive statistics for Indian boys by SES. ............................................... 115
Descriptive statistics for higher SES Aﬁican, Indian and European

boys. ............................................................................................................... 1 18
Scheffé post hoc tests of mean differences (MD) between races for

variables showing a signiﬁcant F ratio among higher SES boys. .................. 120
Descriptive statistics of Aﬁican boys by nutritional status based on the

BMI. ............................................................................................................... 122
Scheffé post hoc tests of mean differences (MD) between nutritional

status of Aﬁican boys for variables showing a Signiﬁcant F ratio. ................ 125
Descriptive statistics of Indian boys by nutritional status based on the

BMI. ............................................................................................................... 128
Scheffé post hoc tests of mean differences (MD) between nutritional

status of Indian boys for variables showing a signiﬁcant F ratio. .................. 130
Participation in speciﬁc physical activities by frequency of

participation in the past seven days for the total sample (n=350). ................. 133
Race-speciﬁc medians and ﬁequencies of participation in the past 7

days in speciﬁc physical activities, and F values. .......................................... 135

SES-speciﬁc medians and frequencies (races combined) of
participation in the past 7 days in speciﬁc physical activities, and F

values .............................................................................................................. 136
Nutritional status-Speciﬁc medians and frequencies (races combined) of
participation in the past 7 days in speciﬁc physical activities ....................... 137
Regression analysis of selected growth status variables by SES, race ,

and nutritional status for total sample (n=3 80). ............................................. 140
Summary table of MANOVA and MANCOVA F-values for SES, race

and nutritional status effects on growth status. .............................................. 140

Pairwise comparisons of estimated marginal means for selected grth
status variables with signiﬁcant F ratios of nutritional status, race and

SES groups with nutritional status, race and SES as covariates. ................... 141
Regression analysis of selected physical ﬁtness variables by SES, race
and nutritional status for total sample (n=380). ............................................. 146
Summary table of MANOVA and MANCOVA F-values for SES, race
and nutritional status effects on physical ﬁtness. ........................................... 146

Pairwise comparisons of estimated marginal means for selected health-
related ﬁtness variables with signiﬁcant F ratios of nutritional status,

race and SES groups with nutritional status, race and SES as covariates. ..... 147
Regression analysis of components of the energy expenditure by SES,

race and nutritional status for total sample (n=228). ...................................... 153
MANOVA of estimated energy expenditure variables by SES, race and
nutritional status for total sample (n=228). .................................................... 154

ix

O
‘l

,. .’.'

Table 46.

Table 47.

Table 48.

Table 49.

Pairwise comparisons of estimated marginal means for selected

physical activity variables with signiﬁcant F ratios of nutritional status,

race and SES groups with nutritional status, race and SES as covariates. ..... 155
Summary table of race speciﬁc AN COVA F values, controlling for

SES, of nutritional status differences in estimated energy expenditure

(kcal/day and kcal/kg/day). ............................................................................ 158
Pairwise comparisons of estimated marginal means for selected

estimated energy expenditure (kcal/day) and physical activity variables

with signiﬁcant F ratios of nutritional status for African, Indian and

European boys. ............................................................................................... 159
Pairwise comparisons of estimated marginal means for selected

estimated energy expenditure (kcal/kg/day) and physical activity

variables with signiﬁcant F ratios of nutritional status for Indian and

European boys. ............................................................................................... 160

Table 50. Regression analyses of estimated energy expenditure variables by

Table 51.

Table 52.

Table 53.

Table 54.

Table 55.

Table 56.

Table 57.

Table 58.

Table 59.

Table 60.

nutritional status for the total sample. ............................................................ 160
Racial distribution, mean and standard deviations of estimated energy
expenditure of boys in the highest and lowest categories of MV PA and

mean daily total estimated energy expenditure (kcal/kg/day) for the

total sample by race ........................................................................................ 162
Chi square statistics for race differences for boys in the highest and

lowest categories of MVPA and mean daily total estimated energy

expenditure (kcal/kg/day) ............................................................................... 162
Means, standard deviations and F ratios for the growth status and

physical ﬁtness of African boys grouped into the highest and lowest

categories of mean total daily estimated energy expenditure ......................... 164
Means, standard deviations and F ratios for growth status and physical

ﬁtness of Indian boys grouped into the highest and lowest categories of

mean total daily estimated energy expenditure. ............................................. 165
Means, standard deviations and F ratios for growth status and physical

ﬁtness of European boys grouped into the highest and lowest categories

of mean total daily estimated energy expenditure .......................................... 166
Means, standard deviations and F ratios for growth status and physical

ﬁtness of African boys grouped into the highest and lowest categories

of estimated energy expenditure in MVPA (kcal/kg/day). ............................ 167
Means, standard deviations and F ratios for growth status and physical

ﬁtness of Indian boys grouped into the highest and lowest categories of
estimated energy expenditure in MVPA (kcal/kg/day). ................................. 168
Means, standard deviations and F ratios for growth status and physical

ﬁtness of European boys grouped into the highest and lowest categories

of estimated energy expenditure in MVPA (kcal/kg/day). ............................ 169
Size and fatness of 10 year old South African boys compared to US.

reference values. ............................................................................................. 172
Mass and Stature of Indian boys from India and South Aﬁica ...................... 174

Table 61. Indicators of SES (mean i SD) by race for upper and lower SES .................. 176

X

Table 62.

Regression analyses of stature by sociodemographic characteristics for

the total sample (n=330) ................................................................................. 177
Table 63. Means, standard deviations, medians and ranges of the BMI by

nutritional status of African, Indian and European boys ................................ 178
Table 64. Means and standard deviations for the BMI, skinfolds, limb

circumferences and skeletal breadths of 10 year old boys from selected

South Aﬁican studies, by race ........................................................................ 181
Table 65. Changes in indicators of the quality of life of South Africans from 1970

to 1995 ............................................................................................................ 187
Table 66. Summary of means (SD, where available) for physical ﬁtness of 10 year

old boys from selected studies. ...................................................................... 189
Table 67. Comparison of fatness and selected performance items of present study

relative to AAHPERD (1980) percentiles by race for higher SES boys

and the total sampleabc. ................................................................................. 193
Table 68. Coefﬁcients of the regression of the BMI and fatness on physical

ﬁtness variables for the total sample of South African boys (n=378) ............ 197
Table 69. Coefﬁcients of the regression of the BMI and fatness on physical ﬁtness

variables for African boys (n=188). ............................................................... 197
Table 70. Coefficients of the regression of BMI and fatness on selecteda physical

ﬁtness items for Indian boys (n=130). ........................................................... 198
Table 71. Coefficients of regression of the BMI and fatness on selecteda physical

ﬁtness items for European boys (n=40). ........................................................ 198
Table 72 Summary of studies of physical activity and estimated energy

expenditure of children ................................................................................... 201
Table 73. Regression analyses of reported 7-day habitual physical activity (PAQ-

C score) on selected physical ﬁtness variables by race. ................................. 209
Table 74. Regression analyses of reported habitual physical activity during

speciﬁc times of the day on selected physical ﬁtness variables by race. ....... 210
Table 75. Correlations and partial correlations of health—related and perfonnance-

related ﬁtness and indicators of activity ......................................................... 213
Table 76. Regression analyses of estimated energy expenditure (kcal/kg/day) on

selected physical ﬁtness variables for the total sample, and by race. ............ 213

xi

LIST OF FIGURES

Figure 1. Stature of 10 year old rural Africans, 1968-1992. ............................................ 39
Figure 2. SES adjusted means for nutritional status - race interactions by SES for
estimated arm muscle area. ............................................................................ 144

Figure 3. Adjusted means and interaction effects for the shuttle run (Race X
SES), 9 minute run (Nutritional status X Race), and standing long jump

(Nutritional status X Race) ............................................................................. 150
Figure 4. Adjusted means for the interaction of SES and nutritional status on

daily total estimated energy expenditure (kcal/kg/day). ................................ 157
Figure 5. BMI (mean :I: SD) of African, Indian and European boys compared to

the U. S. reference median. .......................................................................... 173

Figure 6. Stature, mass and the BMI of African, Indian and European boys In
1962-1963 (Smit et al., 1967), 1978 (Richardson, 1978), 1981 (Kotze
et al., 1987) and 1997 (present study). ........................................................... 185

xii

Chapter 1

INTRODUCTION

The change of the South African government from apartheid capitalism to a
democracy in 1992, resulted in mass migration into the urban periphery, changing both
the character of the urban population and the circumstances of the majority of South
Africans. South African boys who turned ten years of age in 1997 have lived half of their
lives under a system of apartheid capitalism and the other half under a democracy, and
are an ideal example of children in a transitional society. Under apartheid capitalism,
economic and social privileges were granted ﬁrst on the basis of race and then by SES,
creating the socioeconomic differences among the four major racial classiﬁcations in
South Africa - African, Indian, the so-called “Colored” and White. In the post-apartheid
South Africa, the artiﬁcial racial divides were removed, theoretically providing greater
equity among the peoples of South Africa.

Growth is a plastic process and can be inﬂuenced by many environmental factors,
including nutrition, socioeconomic status, geographic location, childhood diseases, and
perhaps physical activity, among others. These factors may also influence physical
performance and physical ﬁtness directly or indirectly. In addition, body size,
composition, proportions and physique affect strength and motor performance (Malina,

1975). Absolute and relative fat-free mass are positively correlated with strength and

motor performance, while absolute and relative fat mass are negatively correlated with
performance, of especially in items that require projection of the body as in runs and
jumps (Clarke, 1971; Beunen et al., 1983; Malina etal., 1995).

Body size is affected by nutritional status. Negative environmental conditions
result in stunted linear growth and possibly growth failure (Zeitlin, 1991; Shetty, 1993).
Undemourished children have absolutely less body fat and muscle tissue compared to the
well-nourished (Desai etal., 1984; Malina and Buschang, 1985; Parizkova, 1987; Spurr
and Reina, 1990). There is also reasonable consensus that undernourished children score
absolume lower on tests of motor perormance, including speed, jumping, strength,
distance runs, and throwing (Ghesquiere and Eeckels, 1984; Malina and Buschang, 1985;
Goslin and Burden, 1986; Malina et al., 1987; Ferreira et al., 1991). Lower physical
working capacity and maximal aerobic power have also been observed in undernourished
children and adolescents (Satyanarayana etal., 1979; Desai et al., 1984; Spurr et al.,
1991). In mild-to-marginally undernourished populations, taller and heavier boys have
higher absolute physical ﬁtness scores compared to smaller age mates, especially for
static strength and physical working capacity. Most of the differences, however,
disappear when scores are adjusted for body size (Ghesquiere and Eeckels, 1984; Desai
et al., 1984; Malina and Buschang, 1985; Spurr et al., 1991).

Populations of the developing world have a greater reliance on child labor, and
individuals often enter into the waged labor market at relatively young ages. These
practices are more prevalent among children from low socioeconomic status (SES)

families compared to those from higher SES families. The added burden of

responsibility results in a signiﬁcantly higher energy output in physical activity for boys
from lower SES compared to high SES families (Post et al., 1992; Slooten et al., 1994).

The relationship between nutritional status and physical ﬁtness may be influenced
by habitual physical activity. In well-nourished populations, there is evidence to indicate
that children with higher general habitual activity have a higher levels of ﬁtness
(Schmucker, 1984; Dotson and Ross, 1985; Mirwald and Bailey, 1986; Sunnegardh and
Bratteby, 1987; Pate et al., 1990). In undernourished populations, habitual physical
activity also appears to be related to physical ﬁtness (Sloan and Hansen, 1969; Goslin
and Burden, 1986) and working capacity (Satyanarayana et al., 1979; Desai et al., 1984),
but the intensity of the habitual physical activity is probably a greater determining factor
of physical ﬁtness among undernourished populations (Wyndham, 1973; Goslin and
Burden, 1986) than frequency and duration of habitual physical activity.

The factors affecting growth, physical ﬁtness and physical activity of children are
multiple and interacting. This study addresses the effects of SES, race and nutritional
status on the growth, physical ﬁtness and physical activity of 10 year old South African

boys.

mg
The purpose of this study is to investigate the effects of SES, ethnicity (race), and
nutritional status on the growth, physical ﬁtness and physical activity of 10 year old

South African boys of European, African and Indian ancestry.

Research Question

The study examines the following question:
What is the relationship of socioeconomic status, race, and nutritional status to
growth status, physical ﬁtness and habitual physical activity of 10 year old South

African boys?

Research Questions

The following research questions were tested:

1. Are there signiﬁcant SES, race and nutritional status differences in the growth status
of 10 year old South African boys?

2. Are there signiﬁcant SES, race and nutritional status differences in the health- and
performance-related physical ﬁtness of 10 year old South African boys?

3. Are there signiﬁcant SES, race and nutritional status differences in the habitual
physical activity of 10 year old South African boys?

4. Are there signiﬁcant nutritional status differences in race-speciﬁc habitual physical
activity of 10 year Old South African boys?

5. Are there signiﬁcant race-speciﬁc differences in the growth status and physical
ﬁtness of 10 year old South African boys of African, Indian and European ancesz

with different levels of estimated energy expenditure?

Deﬁnition of terms

Adolescent encompasses all individuals between 10 to 19 years of age (World
Health Organization, 1995).

Family is a term that is used loosely. There has been a considerable eroding in
the nuclear family in South Africa. The term family is used to describe the group of
people with whom the child lives. This group could be a collection of older relatives,
ﬁiends, grandparents and/or other unrelated individuals. The continued low-scale civil
war over the last decade has eroded conventional family life in many African families
and has made it necessary to redeﬁne the traditional family in the South African context.

Growth is an increase in size of the body as a whole or of speciﬁc parts, due to
changes in cellular processes that include hyperplasia, hypertrophy, and accretion
(Malina and Bouchard, 1991).

Habitual physical activity is the amount of physical activity that the child
normally performs in his regular daily schedule. Physical activity is “any body
movement produced by the skeletal muscles that results in a substantial increase over the
resting energy expenditure” (Bouchard and Shephard, 1994, p. 77). This deﬁnition of
physical activity includes physical leisure, exercise, sport, occupational work and chores,
and other factors that modify the total daily energy expenditure.

Health-related physical ﬁtness is a state characterized by an ability to perform
daily activities with vigor and demonstrate traits and capacities that are associated with
low risk of premature development of the hypokinetic diseases (Pate, 1988). It is

operationalized as body composition (subcutaneous fatness), cardiorespiratory endurance

(9 minute run), and musculoskeletal function (sit and reach, push-ups and sit-ups)
(AAPHERD, I984).

Malnutrition is deﬁned as a body mass index (BMI)-for-age below the 5th
percentile (undemutrition) or at the 85th percentile and above (overweight) of the
NCHS/WHO reference data (World Health Organization, 1995).

Performance-related physical ﬁtness is an indication of the attained levels of
motor performance. Static strength (grip strength), agility (50 meter shuttle run),
running speed (50 meter dash) and explosive power (standing long jump) are components
of performance-related ﬁtness.

Race is a variably deﬁned concept. In South Africa, there are four major racial
classiﬁcations by the Nationalist Government: African (of African ancestry), Indian (of
Indian Ancestry), White (of European ancestry), and so-called “Colored” for people of
mixed ancestry (International Defense and Aid Fund, 1991). Race is a biological referent
while ethnicity is a cultural referent (Smaje, 1996). In this study, race is used
interchangeably with ethnicity since there is little distinction between race and ethnicity
in South Africa. The Group Areas Act and Color Bar Acts in South Africa have
minimized the differences between these concepts.

Sociodemographic status (SDS) is the combination of socioeconomic status
(SES) and race. SES is a reﬂection of the social and economic circumstances of the
individual and his family. The education, income, and occupation of the head of the
household, and the occupancy ratio (number of people per room in the home) are

indicators that affect the SES of the individual. The individual is placed in an SES

category based on the cumulative score of education, income and occupational prestige

of the head of the household.

Limitations

1. The most accurate measures of the energy intake and expenditure of an
individual in ﬁeld studies use weighing and recording of all food consumed, and
recording of all physical activities of the individual. These are time consuming and labor
intensive procedures that require well trained personnel to weigh, calculate and record
the amount of food consumed and to estimate its composition, as well as to observe and
record all physical activity. These methods were not available for this study.

Waterlow et a1. (1977) recommend that weight-for-height be used as the primary
indicator of present nutritional status and height-for-age as an indicator of past nutrition
for the assessment of the nutritional status in cross-sectional studies of children under 10
years of age. While the relationship of weight and height is reasonably stable during
childhood, there are maturational status differences in body weight and stature during
adolescence. Thus, at a given height, the weight corresponding to a particular percentile
is not the same for all ages. The meaning of weight-for-height differs with each age.

The recent guidelines on anthropometry issued by the World Health Organization,
(WHO, 1995) advise the use of the body mass index (BMI) for age as a proxy for
nutritional status. The BMI has high speciﬁcity (99%) and efﬁciency (89%) in overall
correctness in classiﬁcation of obesity. However, it has low sensitivity (29%) as an
indicator of obesity in adolescent boys (Himes and Bouchard, 1989). On the other hand,

the speciﬁcity, efﬁciency and sensitivity of the BMI as an indicator of thinness or

7

undernutrition has not been estimated, but low BMI values are reported for
undernourished adults (Desai et al., 1984; Spurr, 1987; Satyanarayana et al., 1989; Giay
and Khoi, 1991; Naidu and Rao, 1994).

The WHO Expert Committee recommended the provisional use of the US.
reference data of Must et al. (1991) for the BMI until better reference data for
adolescence are available. The recommendations are based on the following strengths:
(1) sampling weights were appropriately used in calculating the national estimates, (2)
percentiles recommended for cut-offs were presented, and (3) percentiles were smoothed
mathematically across ages in an acceptable manner (WHO, 1995).

2. The use of heart rate monitors would increase the accuracy of the
measurement of the physical activity. Monitors were not available for this ﬁeld study.
The estimate of physical activity was based on 3-day activity records maintained by the
subjects. The activity records provided an indication of involvement in light, moderate
and heavy physical activity. Daily energy expenditure and detailed accounts of physical
activity were not attained using the 24-hour recall records. The recall records provided
an indication of the trends in the habitual physical activity of subjects.

Recall surveys are subject to memory decay, social desirability, interpretation of
the questions, and the ability of the interviewer in providing memory enhancing triggers.
The ability of 10 year old South African boys to complete recall records may be affected
by their concept of time and awareness of the intensity and length of an activity. These

potential problems are major limitations. The use of memory triggers and estimates of

time were provided to assist in the recall. However, these limitations are inherent in the
methodology employed and have to be accounted for in the interpretation of data.

3. The health- and performance-related physical ﬁtness of subjects were
estimated from scores on a battery of tests. Culture, nutritional status, health, and
motivation inﬂuence performance on such tests. While efforts were made to increase the
arousal and motivation of the subjects to perform at their best, understanding Of the word
“maximum” may have been a limiting factor.

4. No indicators of maturity status were used. It is assumed that most boys in the
sample were prepubescent. However, some boys from the upper SES samples may have
been in the early stage of puberty.

5. A convenience sample of volunteers from Government, Model-C and private
schools was used in this study. The potential school effect biases in the sample were not
considered.

6. While growth status is a combination of past and current nutritional history,
there was no control for the nutritional history of subjects.

7. The ongoing low scale civil war in South Africa introduces stressors that
inﬂuence the development and maturation of children. These factors, however, fall

outside the scope of this study.

W

This study looks at the complexity of factors that affect the growth status and
physical ﬁtness of 10 year old South African school boys from. the greater Johannesburg

area (Guateng). The relationship of SES, race and nutritional status to growth and

9

physical ﬁtness are considered. SES and demographic factors play an important role in
identifying and explaining components that impact the growth and physical ﬁtness of an
individual. The relationship of SES, race, and nutritional status to growth, physical
ﬁtness and physical activity contribute to the understanding of factors that affect the

growth and physical ﬁtness of children.

Signiﬁcance of Study

This study addresses the relationship of SES, race and nutritional status to the
growth status, health-related and performance-related physical ﬁtness, and habitual
physical activity of 10 year old South African boys of African, Indian and European
ancestry. The effects of SES, race and nutrition and their interactions have direct and
indirect impacts on growth status and physical ﬁtness. The complexity of factors that
impact on growth is well documented (Ijsselmuiden, 1984; Richardson and Sinwell,
1984; Chaning-Pearce and Solomon, 1986; Cameron et al., 1988a; Cameron, 1991).
Several studies have addressed aspects of SES, race, nutritional status, growth and
physical ﬁtness relationships of South African children. These include, for example,
nutritional status and physical ﬁtness (Sloan and Hansen, 1969), ethnic variation and
physical ﬁtness (Sloan, 1966a, 1966b), and performance (Kelder, 1984), physical ﬁtness
and growth status (Smit, 1969; Henneberg et al., 1987), and the efﬁcacy of enriched
physical activity programs on the growth and physical ﬁtness of low SES children (Van
Rooyen, 1994).

South Africa is undergoing signiﬁcant transitions in the 1990s (IDAF, 1991).
There has been a major urban-rural shift in recent years (Hindson et al., 1994). The

10

desegregation of population groups and restoration of rights and privileges to previously
disenfranchised populations has created a complexity of environmental dynamics. Thus,
examination of the growth and physical ﬁtness of children in a transitional society may
provide a unique perspective 0 the transition.

Race and SES are closely aligned in South Africa (IDAF, 1991). Physical
activity and nutritional status are inﬂuenced by cultural and religious practices.
Therefore, focusing on the individual and combined effects of the SES-race couplet is
necessary to tease out individual contributions to the growth status and physical ﬁtness of
10 year old South African boys. Boys who turned ten years of age in 1997 would have
spent the ﬁrst half of their lives under a system of apartheid capitalism and the latter half

under a system of democracy, and provide a good example of children in a transitional

society.

11

Chapter 2

REVIEW OF LITERATURE

The literature review is divided into three parts. The ﬁrst presents an overview of
South Africa to contextualize the sociopolitical environment and to provide insight into
the dynamics that determine and deﬁne socioeconomic status (SES), race or ethnicity,
and nutritional status. The second part focuses on SES, race or ethnicity and nutrition in
studies of human growth. In the third part, the relationship of SES, race or ethnicity and
nutrition with physical ﬁtness is examined; the relationship of habitual physical activity,

energy balance, growth and physical ﬁtness is also considered.

'Mmmmmmmmmmm

South Aﬁica spans the southern tip of the African continent. It covers an area of
1,221,040 krn2 (472,359 miles). It is mostly semi-arid, but is subtropical along the east
coast. The sociopolitical and economic climate of South Africa is a mixture of the
apartheid capitalism of previous years and its newly won democracy.

South Aﬁica formally endorsed the policies of apartheid from 1910 with the
enactment of the Land Act of 1910 until the democratic elections of 1994. Enforcing

apartheid in South Africa was made possible by the enactment of laws that clearly

deﬁned the boundaries of racial integration. The Prohibition of Mixed Marriages Act

12

(1949) and the Irnmorality Act (1950) made mixed marriages and sexual intercourse
between Europeans and Blacks illegal. The Population Registration Act (1950)
classiﬁed the population into four major racial groups.

White was used to deﬁne anyone who obviously looked European but was not
generally accepted as a “Colored” person, and included all people of European ancestry
(Riley, 1991). The term Black was used to group people of any aboriginal tribe of
Africa, hereafter referred to as African (Riley, 1991). The term “Colored” was a broad
classiﬁcation that included only those of mixed descent in terms of the racial
classiﬁcations of the country and people of Malaysian, Khoi, and San descent. They are
hereafter, referred to as so-called “Coloreds” (IDAF, 1991). People of Indian and
Middle Eastern ancestry were classiﬁed as Indian (IDAF, 1991). The proportion of
people of Middle Eastern ancesny classiﬁed as Indian is unknown.

The Pass Laws of August 1966 made it compulsory for all people over the age of
16 to possess, carry and produce identity cards at the request of an authorized person
(UNESCO, 1974). The Pass Laws were part of the inﬂux control measures and were
used to minimize the number of Africans in White South Aﬁica. The Bantu Self-
Govemment Bill (1959) provided for the establishment of eight national units and was
used to transfonn former African reserves into self-goveming homelands with limited
legislative authority (Riley, 1991). Bantustan Homelands were the only areas in which
Africans were legally permitted to own land. Residence in these homelands was based
on tribal affiliations and served to increase inter-tribal conﬂict. Township residence was

restricted to those actively engaged in waged labor (IDAF, 1991).

13

The political economy of South Africa was one of apartheid capitalism where
race and class deﬁned access to, and ownership of, power in South Africa. In apartheid
capitalism, the wealth and resources of the country were distributed on a sliding scale
where the interaction of race and class determined access to wealth and power.

Apartheid capitalism of South Africa created a society in which minimal distinction
between class and race existed (UNESCO, 1974). The economy of South Africa was
made to ﬁt the imperatives of the racial political system and was done by controlling the
labor market, and exercising monopoly control and racial exclusiveness over skilled jobs
(Wolpe, 1987).

Education was stratiﬁed along the lines of race (Bantu Education Act, 1953;
“Colored” People’s Education Act, 1963; Indian’s Education Act, 1965) to provide for
the different sectors of the labor market (UNESCO, 1974). This stratiﬁcation was the
basis of apartheid capitalism. European education was designed to produce the leaders in
the community and African education to provide unskilled manual laborers. This
segregation was further enforced by forbidding Blacks, mainly Africans, from ﬁlling
certain skilled jobs. Aﬁicans thus provided the pool of cheap unskilled and semi-skilled
labor which formed the basis of the economy (UNESCO, 1974). Urban Africans were
predominantly employed in the mines and industries, while the majority of rural Africans
were employed by European farmers (UNESCO, 1974). Indian education, on the other
hand, was designed to produce skilled, semi-skilled, clerical and professional labor, and

lower to middle management (UNESCO, 1974).

14

Under the apartheid laws, Europeans were legal or ﬁrst class citizens of the
country. Africans, “Coloreds” and Indians were second class citizens of the country.
Suffrage, compulsory and free education, access to public health care, and choosing
where to live were the rights of ﬁrst class citizens, which were denied to second class
citizens. The Group Areas Act (1950), which restricted residence to racially designated
areas, was pivotal in maintaining the status quo in South Africa (Riley, 1991).

Similar inequalities were expressed in all aspects of civil life. The majority of the
resources in the country were spent on provisions for Europeans (Table 1). Thus, while a
relatively small proportion (2.2%) of the GNP was spent on health care compared to
other countries in the region, a 63-fold greater per capita health care expenditure was
made for Europeans ($146) than for Africans ($23.75) (Henry J. Kaiser Family

Foundation, 1991).

Table 1. Basic indicators of health status in South Africa.‘I

 

 

Indicator and year African European “Colored” Indians
Population (million), 1996 " 322 52 35 1.1
Percentage of population 76.7 124 8.4 25
Infant mortality, per 1000 Iivebom, 1988 80.0 119 463 19.0
Percentage of population < under 5, 1988 159 79 11.9 103
Per capita health expenditure, 1987c $15.57 $97.70 $5557 $5836
(R95) (1596) (R339) (R356)
Per capita educational expenditure, 198 8° $45.5 $504.92 $37.05 $364.92
(11276) (R1080) (R1385) (R1226)
Life expectancy, 1985 62 71 61 67
Life expectancy, 1996-2001 645 73.6 64.4 702
Percentage of all deaths < 1 year, 1987 15.7 1.3 10.8 5.4
Percengge of all deaths < 5 years, 1987 22.5 21 163 7.3

 

'Adapted from Henry J. Kaiser Family Foundation (1991, p. 10).
”From South Africa Survey 1996/7 (Sidiropoulos et al., 1997, p.8).
°Based on the rand-dollar exchange rate of six rand and ten cents (R610) to one $1.

15

The abolition of the Inﬂux Control Act (1986) (IDAF, 1991), discontentment
with the political and economic conditions, an ongoing drought from 1984 to 1994
(Hindson et al., 1994), and the removal of the Group Areas Act (1991) led to
considerable resettlement in South Africa. The removal of the Group Areas Act made it
possible for the majority of South Africans to choose where they lived. Large numbers
of people moved out of the impoverished homelands into urban areas in search of
employment (IDAF, 1991). There was considerable relocation between residential
suburbs. Additionally, peri-urban squatter settlements grew rapidly (Hindson et al.,
1994).

Major changes in the political economy and social conditions, subject to SES, of
South African populations occurred in the past 10 years (Table 2). In the ten years from
1985 to 1994, there was a 12.8% increase in personal disposable income, with race
speciﬁc increases of more than 35% for Africans and 24% for Indians, and a decrease of
2.5% for Europeans (Sidiropoulos et al., 1997). The Human Development Index (HDI)
of the United Nations is a measure of socioeconomic development and is a composite of
life expectancy, per capita income and the level of education. Between 1980 and 1991,
Africans in South Africa moved from a low HDI of 0.394 to a medium level of 0.500.
Both Europeans and Indians moved from a medium (0.739 and 0.655, respectively) to a
high (0.901 and 0.836, respectively) level of human development, indicating an
improved quality of life for all groups.

Unemployment, however, is high. It is currently estimated that 29% of the

economically active sector is unemployed (Sidiropoulos ct al.,.1997). In Gauteng, 20.9%

16

of the economically active population was unemployed in 1995 (Sidiropoulos et al.,
1997). Despite the change in the hegemony, there are large race differences in
unemployment rates. African women (46.3%) and men (28.9%) have the highest, and
European males (3.7%) and females (8.3%) have the lowest unemployment rates. In

1995, 87% of all unemployed people were African.

Table 2. Relative changes in the economic status of South Africa over the last
decade.‘I

 

African Indian European “Colored” Total

 

Personal disposable income“ 35% 24% -2.5% 24% 12.8%
I-IDI° 26.9% 27.6% 21.9% 24.6% 21.5%
Unemployment rates 36.9% 13.4% 5.5% 22.3% 29.3%

 

' Extrapolated from Sidiropoulos et a1. (1997).
" Change between 1985 to 1994 ° Change between 1980 to 1991

There has been a rapid increase in the urbanization of South Africa. The Gauteng
urban conurbation (greater Johannesburg to Pretoria) has a current population of more
than 11 million. It is now the second largest metropolitan spread in Africa after Cairo
and is “possibly the fastest growing emerging city in the world” (Weekly Mail, May 16,
1997). In 1995, Johannesburg was the ninth largest African city (Sidiropoulos et al.,
1997). The region is characterized by mushrooming zones of informal settlements that
are populated by people who are ﬂeeing rural poverty and those who are relocating from
former townships. These informal settlements are characterized by a lack of sewage

facilities or rubbish removal, and do not have access to piped water.

17

The change in the governing policies of the country has been coupled by an
unprecedented increase in crime. Statistics for 1995 indicate a dramatic increase in
violent crimes over the last 21 years, with increases in individual categories being 24%
for assault, 119% for murder, rape by 149%, and robbery by 171% (Sidiropoulos et al.,
1997). There has been a 275% increase in theft of motor vehicles and a 223% increase in
housebreaking between 1974 and 1995 (Sidiropoulos et al., 1997). Statistics for the
percentage change in crimes in South Africa since the shift in government between 1994
and 1995 are not available. The number of vehicle hijackings has escalated in reCent
years with the 1996 (from January to August) average being 36 cars a day. Gauteng is
one of the most crime ridden provinces of South Africa, with 22% of the murders, 80.7%
of all car hijackings, and 26% of all rapes (Sidiropoulos et al., 1997).

Any analysis of health and development in South Africa must take into account
the ﬂuidity of its population. The majority of African children are newly arrived from
homelands, townships, rural and peri-urban areas (Hindson et al., 1994), and bring with
them a social, nutritional and cultural history. Thus, a contextual analysis which
incorporates the niral, peri-urban and urban populations is necessary to understand the
factors that affect the status of the population.

A sample of boys who turned 10 years of age in 1997 are survivors of the
apartheid era in South Africa. The Group Areas Act (legal enforcement of racially
segregated residential areas) and the Color Bar Act (legal prohibition of interracial
marriages or cohabitation) enforced by the apartheid government presented a barrier to

racial admixture. These boys experienced the privileges (for Europeans) and

18

disenfrancement (for people of color but especially Africans) that the apartheid policies
of South Africa enforced for the ﬁrst 6 years of their lives. Despite the removal of the
Color Bar Act in 1982, the Group Areas Act (abolished in 1991) served to control
interracial marriages and cohabitation.

The diversity of the South African population is large and includes the
geographical divide which banished Africans to barren and harsh climes. Additionally,
the Color Bar and Group Areas Acts limited racial admixture. Thus, the South African
population presents an opportunity to look at the effects of race, culture and SES on
growth and development of a diverse group of children within the same country. Boys of
African, European and Indian ancestry who turned 10 years of age in 1997 present a

sample with clearly different SES histories and circumstances.

Summary

'South Africa has a population that has been traditionally classiﬁed into three
racial groups of African, Indian and European, with people of mixed ancestry classiﬁed
as the so-called “Colored” by the Colonialists and the former government. Racial
classiﬁcation by the former government was limited to four groups that were arbitrarily
assigned using broad visual criteria. The political economy of the country created a
society that is stratiﬁed in terms of SES (largely deﬁned by race) with a large working
class and a small middle class. The working class and lower SES groups have

maintained cultural homogeneity, while the middle class were products of English

19

mission education and assimilated the culture of the English (Archer and Bouillon,
1982)

The current status of South Africans is the legacy of apartheid, rapid
urbanization, proliferation of peri-urban squatter settlements, high prevalence rates of
HIV infection and AIDS (McItyre, 1996), high unemployment rates, high rates of
illiteracy, single parent families and orphaned children. Unequal education, inequity in
the power sharing and decision making processes, and the oppression of the majority of
the people in South Africa created a society with large differences in living standards.
Despite the change in the political climate of South Africa, the majority of South

Africans have not beneﬁted from these changes.

Growth and physique

It is well established that children in the developing world are shorter and lighter,
and mature later, on average, than children in developed world populations (Eveleth and
Tanner, 1976, 1990; Cameron, 1991). Poverty (Martorell, 1982), culture (Gopalan,
1992), malnutrition and infectious disease (Mata, 1972; Stephenson et al., 1991, 1993)
are perhaps among the major environmental factors that affect the growth of children in
the developing world.

Stunting, or linear growth retardation, depends on SES, maternal nutrition and
health status, and the educational level of the family (W aterlow, 1994). The main causes
of stunting are insufﬁcient or inadequate prenatal and/or postnatal energy and nutrient

supply (Martorell, 1993), increased prevalence of infection (Mata, 1972; Stephenson et

20

al., 1991), altered fetal growth, (Allen and Uauy, 1994) and mother-child interaction
(Chavez and Martinez, 1984; Waterlow, 1994). Genetic transmission, congenital or
environmental factors, maternal disease or nutritional deﬁcits, and maternal malnutrition

affect linear growth in utero (Allen and Uauy, 1994).

Nutrition and poverty in South Africa

The nutritional status of people reﬂects, to a large extent, the poverty level of a
country. In South Africa, the majority of the African population lives in conditions
where basic survival needs are often unmet or only marginally met (Henry J. Kaiser
Family Foundation, 1991). Preventable communicable diseases and malnutrition affect
most populations in rural areas and informal settlements in the peri-urban areas (Henry J.
Kaiser Family Foundation, 1991; Hindson et al., 1994). Poor health situations are due to
the lack of sanitation, clean water supplies, and waste removal services; crowding; and
the public health infrastructure. The increased incidence and prevalence of respiratory
infections are attributable in part to the use of kerosene, coal and anthracite as fuel
(Y ach, 1988).

African and “Colored” children are 14 to 15 times more likely to die before the
age of 5 than European children, and over 200,000 African children were wasted, with
low weight-for-height in the 1980’s (UNICEF, 1987). Differences in poverty levels by
race, age, and urban-rural status affect the nutritional status of the total population (Table

3). More recent birth to 5 data has not been reported.

21

Table 3. Prevalence of stunting, underweight and wasting in children by ethnicity.

 

African
Urban Rural “Colored” Indian European

 

ltoSyearsofage'

Stunting (90% height-for-age) 11.7 40.9 18.6 6.4 3.7

Underweight (80% weight-for-age) 28.2 42.7 48.6 35.0 15.6

Wasting (80% weight-for-height) 6.9 3.0 11.3 14.5 7.5
6 to 12 years of ageb

Weight-for-age 22.2 24.7 17.8 3.8

 

' Adapted from Wyndham (1986, p.281), based on the NCHS reference data.
b Adapted from The Department of National Health and Population Development (1990);
criterion for reference data was not indicated.

It is currently estimated that 2.3 million (17.4%) people of the total population in
South Africa are malnourished (Ministry for Welfare and Population Development,
1995). However, the criterion for malnutrition was not speciﬁed.

SES is a strong determinant of growth status in most populations. Cemerud
(1993) found associations between the father’s occupation level and the height of 10 year
old children from Stockholm and the number of inhabitants per room. Other data (Desai
et al., 1984; Chaning-Pearce and Solomon, 1986; Satyanarayana et al., 1989; Bogin et
al., 1992; Cameron and Kgarnphe, 1993) indicate differences in height and weight, and
body composition between children of upper and lower SES in several areas. In
countries where ethnicity (race) is closely tied with SES, the distinction between SES
and ethnicity is unclear. The interrelationship between SES and race has been noted in

England (Rona and Chinn, 1986), Guatemala (Bogin et al., 1992), the United States (Joe,

22

1987), Zaire (Ghesquiere and Eeckels, 1984), and South Africa (Sloan and Hansen,
1969; Richardson, 1978; Chaning-Pearce and Solomon, 1986). People of European
ancestry in both Europe and developing world populations are generally of higher SES
compared to non-European populations. Furthermore, in developing world countries,
aboriginal populations are generally of lower SES compared to immigrant populations.

The relationship between SES and growth is not always linear. In environments
with a high prevalence of malnutrition, stunting and illness among children, some
children demonstrate a ‘positive deviance’ by growing and developing adequately despite
the poor environmental conditions (Zeitlin, 1991). Personal and household hygiene, the
preparation and storage of food, and handling of water are amongst the factors that are
responsible for the “positive deviance. (Martorell, 1982; Zeitlin, 1991).

Cultural practice associated with food and eating inﬂuence the energy and
nutrient content of the meal. Preferential feeding patterns in families that favor certain
individuals within the family can alter the SES-nutrition relationship and produce
nutritional status differences between individuals within the same family. Preferential
feeding of males is a common practice in cultures that favor male offspring. Birth order
and spacing, gender, and age are factors which determine sharing and portioning of food
(Dugdale, 1985). Cultural food taboos also affect nutritional status. Religious customs,
such as vegetarianism as practiced in some forms of Hinduism and the prolonged periods
of food abstinence (mandated for all healthy adults and most school-aged children)
during Rarnadaan as a discipline of Islam, may affect the nutritional status of children

and adults.

23

Nutrition and growth

The assessment of the nutritional status of an individual or population includes
dietary history, physical examination, anthropometry, body composition, and
biochemical analyses (Spurr, 1983a). Undemutrition is deﬁned as “measurable changes
in nutritional status that result from a chronic marginal deﬁcit of food quality and or
quantity” (Allen, 1994, p. 924). It is generally accepted that malnutrition in developing
countries is largely due to caloric or energy insufﬁciency (F A0, 1987).

The severity of the physiological effects of malnutrition can range from mild to
severe and are due to the interaction of intensity, duration and timing of the stress
(Malina, 1985). Social, cultural, economic and political factors; and parasitic and
infectious diseases are several interrelated contributors to undemutrition (F erreira, et al.,
1991). Mild-to-moderate manifestations are a result of chronic early childhood exposure
to the stress of undernutrition. Characteristic features of undemutrition are stunted
growth, delayed biological maturation, reduced muscle mass, and decreased physical
working capacity (Spurr, 1983; Malina, 1984, 1985; Malina and Buschang, 1985;
Johnston et al., 1985), retardation of functional and cognitive development (Chase and
Martin, 1970; Chavez and Martinez, 1979; Allen, 1990; Pollitt et al., 1993, 1995; Wachs,
1995), reduced levels of physical activity (Adams et al., 1994), impaired resistance to
infection (Frank and Zeisel, 1988), increased morbidity and disabilities, and eventually
death (Food Nutrition and Agriculture, 1992). Marginal undemutrition is more prevalent
than severe undemutrition in developing countries; severe undemutrition presents itself

as marasmus, kwashiorkor or an undifferentiated form (Allen, 1993).

24

Anthropometric assessment of nutritional status includes weight-for—age
(underweight), height-for-age (stunting), weight-for-height (wasting), BMI-for-age
(thinness and overweight), and the triceps and subscapular skinfolds (WHO, 1995).
Chronic and persistent undemutrition is characterized by low weight-for-height and low

height-for-age (WHO, 1995) based on reference data for American children and youth

 

 

(Table 4).
Table 4. Recommended cut-off values for adolescents.“
Indicator Anthropometric variable cutoﬁ‘ values
Stunting Height-for-age < 3rd percentile or
< -2 Z-score
Thinness BMI-for-age < 5th percentile
Obesity BMI-for-age 2 85th percentile for BMI,
Tri s skinf ld S 2 90th percentile for S,
cep o ( 1) 2 90th percentile for S,

Subscapular skinfold (8,)

 

'Adapted from WHO (1995, p. 271)
bAdolescents are deﬁned as youth between 10 and 19 years of age.

Weight-for-age and height-for-age were the previously recommended
anthropometric indicators for the nutritional status of adolescents, deﬁned as 10 to 19
years of age (WHO, 1995). However, potential maturity-associated variation in body
size make the reference data awkward to use without adequate control. The BMI-for-age
is currently recommended as the indicator for assessing nutritional status in adolescents
(WHO, 1995). The correlation between BMI and total body fat is 0.90, and between
BMI and percentage body fat is 0.68 in boys 6 to 12 years of age, using underwater

weighing (Siri Equation) to estimate density and fatness (Roche et al., 1981). The BMI

25

was the best indicator of total body fat, but a poorer estimate of percentage body fat
compared to relative weight, weight /stature3, and the triceps, subscapular and suprailiac
skinfolds (Roche et al., 1981).

BMI percentiles based on NHANES I have been estimated by at least three sets of
independent researchers (Hammer etal., 1991; Must et al., 1991; Cronk and Roche,
1982). Hammer et al. (1991) limited their analysis to Whites only. Cronk and Roche
(1982) analyzed the data for the 6 to 50 year old population and estimated the BMI for
Blacks and Whites separately. Must et al. (1991) analyzed the data for Blacks and
Whites separately, but also presented BMI values for the total population.

Different weighting techniques were used by Cronk and Roche (1982) and Must
et al. (1991a). Cronk and Roche (1982) smoothed all percentile values using low term
Fourier transforms. The terms chosen for smoothing were based on the goodness of ﬁt to
the original percentile plots and the effectiveness of smoothing. The abruptness and
spacing of inﬂection of the unsmoothed data were used to choose the number of terms.

In some cases, the smoothing resulted in the 5th and 10th percentiles meeting at some
ages, partly because the smoothing was done independently for each percentile. When
this asymmetry occurred, the percentiles were separated by the smaller unit of the
variable. Additionally, BMI percentiles for Whites and Blacks were presented
separately.

Must et a1. (1991) used locally weighted regression scatter-plot smoothing across

percentiles for age groups. Race-speciﬁc and population-based curves for the BMI were

26

smoothed separately, but were also presented as percentiles for the total population.
These produced different BM] reference values (Table 5).

The distributions of the BMI-for-age in U.S. children (Must et al., 1991) are
skewed and elevated compared to French children (Rolland-Cachera, 1991; Rolland-
Cachera et al., 1993), indicating that the U.S. sample was heavier for a given height than
the French sample (Table 5). Thus, U.S. percentiles are not necessarily representative of
other populations. While the BMI has not been fully validated as an indicator of
thinness or undemutrition in adolescents, WHO (1995) posits that it provides a single

index of body mass that is applicable at both extremes.

Table 5. BMI percentiles for 10 year old U.S. and French boys.

 

 

 

mass
51h 10m 25m 50th 7Sth 90th 95th

Crank and Roche (1982)'

White males 14.5 14.9 15.9 17.1 18.9 21.3 23.4

Black males 14.4 15.1 15.7 16.6 17.9 21.1 23.2
Must et a1. (1991)A - Total population 14.4 (15.2)” 16.8 (19.7)c 22.7
Rolland-CacheraJl991J 14.0 14.8 15.4 16.3 17.8 18.8 21.0
' NHANES I data.
b 15th percentile in parentheses ° 851h percentile in parentheses

Deﬁning 10 year old boys as adolescent may be problematic. The mean age for
entry into the early stage of puberty, genital stage 2 (G2), is 11.20 i 1.50 years for Swiss
(Large and Prader, 1983), and 11.61 i 1.07 years for English (Marshall and Tanner,
1970) boys. At 10 years of age approximately 16% of English and Swiss boys would be
in the early stages of puberty. African South African boys from higher SES families, on

the other hand, appear to be advanced in relation to poor rural African boys. The mean

27

age of entry into GZ is 10.5 i 2.06 years for a sample (n=152) of higher SES African
boys attending fee paying private schools in Johannesburg (Cameron et al., 1993). The
variation in the sample is considerable with a large standard deviation. Approximately
84% of the sample of higher SES African boys are in the early stage of puberty. Lower
SES rural South African boys, however, have a mean age of 13.35 i 2.01 years upon
entry into G2 (Cameron et al., 1993). At age 10, only 2.4% of lower SES rural African
boys are pubertal.

At 10 years of age, height-for-age and weight-for-age may be valid indicators of
nutritional status since most 10 year old boys are probably still prepubertal. Despite
limitations, the WHO (1995) advocates the BMI-for-age as an indicator of undemutrition
in a population with a high prevalence of undernutrition. Additionally, population based
reference data are available (Must et al., 1991a, 1991b). The BMI-for-age, however, has

not been fully validated as an indicator of thinness or undemutrition in adolescents.

Indicators of health status in South African populations

Infant and Under 5 mortality

Infant and preschool mortality (1.00 to 4.99 years of age - U5) rates are accepted
indicators of the health and nutritional status of a population (Martorell, 1982). There is
a great disparity in the health and living conditions between urban and rural populations
in South Africa, which serves to illustrate the regional differences in the health and

nutritional status in South Aﬁica.

28

Rural areas in South Africa have a scarcity of facilities and sanitation, lack proper
food Storage, and have limited access to a varied diet. These factors favor growth
stunting (Little et al., 1989). The majority of the rural population in South Africa live in
conditions of poverty and undemutrition, which is reﬂected in the infant and US (1.00 to
4.99 years of age) mortality rates. Infant mortality (IM) rates ranged from 48 to 151 per
1000 liveborn in mm] villages in northern South Africa in the 19705 and 19805
(Richardson and Sinwell, 1984), and remained high (88 per 100011vebom ) in the late
19805 (Koumans, 1992). Irwig and Ingle (1984) reported an US mortality rate of 190 per
1000 population in rural Transkei (a homeland), whereas IM for African South Africans
was 80 per 1000 liveborn in the same period (Henry J. Kaiser Family Foundation, 1991).
More recent data with a breakdown of urban and rural infant mortality rates are not
available. The combined urban and rural IM rates were 52/1000 liveborn with an US
mortality rate of 69/1000 population (World Bank, 1995). In Northern Transvaal, 15%
of annual admissions and 28% of all deaths at the Elim hospital, which serves Venda and
Gazankulu (homelands), are primarily due to undemutrition (Ijsselmuiden, 1984).

High rates of stunting and wasting are reported in infants and children in the rural
homeland populations of South Africa (Shuenyane et al., 1977; Ijsselmuiden, 1984;
Richardson and Sinwell, 1984; de Villiers, 1987). Issjelmuiden (1984) and Richardson
(1986) report that one-third or more of the rural US population are underweight (32.3% <
3rd percentile for weight-for-age, and 42.7% < 80% weight-for-age, respectively).

However, only 28.2% of urban U5 African boys are underweight (<80% weight-for-age)

29

(Richardson, 1986). A higher proportion of rural (40.9% ) compared to urban (11.7% )

U5 boys are stunted (< 90% height-for-age) (Richardson, 1986).

Children and Adolescents

Studies of growth status of children and adolescents in South Africa have several
limitations. Data are generally presented only graphically. Although the majority of
studies use the NCHS (U .S.) reference data, some use British reference data. Criteria for
deﬁning stunting and wasting range between 80% to 90% of reference values, and < 5th
or < 3rd percentiles. Thus, direct comparisons are often limited. Many of the studies
included in this review are based on extrapolations from graphs. Mean values are plotted
and in most cases, the range, standard deviations and/or standard errors are not indicated.
Some researchers (Leary, 1968; Walker and Walker, 1977; Cameron and Kgamphe,
1993) do not indicate the method used to specify age groups or to determine age. Ten
years of age may include children between 9.5 to 10.4 years, with a mean of about 10.00
years, or between 10.00 and 10.99 years of age, with a mean of about 10.5 years.
Additionally, the accuracy of reported birth dates and thus ages may further confound
comparisons. Data intrapolated from graphs are for 10.5 years of age.

This review includes 31 samples: 11 rural African, 7 urban African (one sample
of African school children did not distinguish between urban and rural status, Kotze et
al., 1986), 6 urban European, 3 urban Indian, and 4 urban “Colored”. Populations of
“Colored”, Indian, and European ancestry are underrepresented. Additionally, 7 of the
samples are from the same study. The two samples of Tswana boys are from Botswana,
but are included in this review since the Tswana are split across the South Africa-

30

Botswana border. Two studies (Richardson, 1978; Kotze et al., 1986) present data on a
cross-section of South African children. Richardson (1977, 1978) compiled grth
charts based on data from other studies. Kotze et a1. (1986) present data from a survey of
24,426 South African school children randomly selected in 1981. Additionally, the
majority of the studies were conducted in the 19705, and considerable sociopolitical

changes have taken place since.

Growth status of South African children and adolescents

On average, rural African populations (both male and female) in South Africa are
shorter and lighter, age for age than urban African children (Walker and Walker, 1977;
Richardson, 1978; Wagstaff et al., 1987). Similar urban-rural differences in stature and
mass occur in Namibia (Jooste et al., 1992). However, urban-nlral differences are not
evident across all South African groups, with greater variability among means with urban
and rural samples (Table 6). Some studies (Leary, 1968; Walker and Walker, 1977;
Badenhorst et al., 1992; Cameron and Kgamphe, 1993) indicate that rural samples are
taller and heavier than urban samples (W agstaff et al., 1987). Sampling and cohort
variation are confounding factors. Corlett and Woollard (1988), on the other hand,
report taller stature in Tswana children 7 to 14 years of age from rural (Kgalagadi)
compared to urban (Gabarone) Botswana. However, urban Tswana children are shorter
than the NCHS reference data (Corlett, 1986). The children from Kgalagadi (rural) have
signiﬁcantly lower body weights than the urban sample.

Approximately 76.5% of all rural Africans (South Africa) lived in former

homelands in 1990 (Henry J. Kaiser Family Foundation, 1991). These areas are

31

characterized by forced relocation, overcrowding and lack of economic opportunity. The
inhabitants of homelands are primarily women, children and the elderly (Henry J. Kaiser
Family Foundation, 1991).

Urban-rural differences in growth are confounded by SES, the economic status of
an individual within the society in which he/she resides. A major limitation in
comparing SES across studies is deﬁnition. SES does not have a universal singular
indicator, and commonly used indicators such as occupational prestige, area of residence,
education of parents and household income do not necessarily indicate the SES of
individuals within their society or the same thing in different societies. The use of a
composite SES score that includes several of the indicators of SES may increase the
accuracy of the classiﬁcation. A singular indicator of SES, such as the possession of
electronic equipment or the ownership of a drivers license, is used but their meanings
may differ across urban and rural, or developed compared to developing world
populations. Additionally, the deﬁnition of urban and rural may differ. The comparison
of SES across studies must account for the deﬁnition of SES.

Most rural African populations are of lower SES compared to urban African
populations. However, the relationship between SES and growth is not always positive
and linear. For example, Cameron and Kgarnphe (1993) did not observe a relationship
between SES and growth in two South African rural populations. These authors
compared the growth of a sample of male and female Zulu from Ubombo and North
Sotho/Pedi from Vaalwater, 6 to 18 years of age. Despite the instability and seasonality

of food supply of the Ubombo compared to the Vaalwater sample, the men were

32

consistently taller, heavier and fatter than the Vaalwater sample. The Ubombo sample
were also the tallest, heaviest and fattest African children compared to samples from
rural Botswana (Corlett and Woollard, 1988), pastoralists and rural Kenyans, and average
urban children from Soweto (W agstaff et al., 1987). Ubombo youth have, on average,
larger subscapular and triceps skinfolds, and arm circumferences, and are heavier and
taller compared to the Vaalwater youth. Cameron et al. (1992) argue that these
differences in the morphology between the Northern Sotho/Pedi (Vaalwater) and Zulu
(Ubombo) may reﬂect major dietary differences, and posit that the Northern Sotho/Pedi
children exhibit a marasmic ‘line of development’ which is characterized by low fat and
low muscle. On the other hand, they suggest that Zulu children show a kwashiorkor ‘line
of development’ which is exempliﬁed by high fat and low muscle. Cameron et a1.
(1992) postulate that the observed differences in the morphology of the Ubombo and
Vaalwater samples could reﬂect genetic variation in subcutaneous fat distribution.
Samples from 1967 and 1993 were compared to evaluate the growth status of
South African populations over the past 25 years (Table 6). This raised the question if
some Of the differences in stature and mass are related to secular changes in urban
populations. There is no evidence of secular change in the rural African population over
the last 25 years. With the exception of the sample of Badenhorst et a1. (1992), there is a
general negative trend in the heights of 10 year old rural African populations (Figure 1),
while there is a slight positive trend in the heights of urban populations from 1967 to

1986.

33

Table 6. Mean heights, weights and estimated BMI from selected samples of 10

year old South African boys.

 

 

Height (cm) Weight (kg) BMI (kg.nr:)‘
Rural African
Pedi (Leary, 1968) 137.8 29.4 15.5
Bochern - 1976 (Walker and Walker, 1977) 131.5 27.2 15.7
Kornatipoort - 1974 (Walker and Walker, 1977) 137.9 30.6 16.1
Letaba - 1975 (Walker and Walker. 1977) 132.7 27.5 15.6
Rustenburg - 1975 (Walker and Walker. 1977) 129.3 25.6 15.3
Saulspoort - 1975 (Walker and Walker, 1977) 131.0 28.3 16.5
Richardson (1978)” 128.0 23.0 14.0
Tswana (Corlett and Woollard, 1988) 129.2 25.3 15.2
Johannesburg (Badenhorst et al., 1992) 140.7 31.4 15.9
Ubombo - 1985-1986 (Cameron and Kgamphe, 1993) 132.5 28.0 15.9
Vaalwater - 1985-1986 (Cameron and Kgarnphe. 1993) 130.5 26.5 15.6
Urban African
Pretoria - 1963 (Smit, 1969) 132.3 26.3 15.0
Soweto - 1976 (Walker and Walker, 1977) 131.8 28.8 16.6
Richardson (1978)” 134.0 28.5 15.9
Tswana (Corlett. 1986) 136.1 28.0 15.1
Johannesburg - 1976-1977 (Chaning-Pearce and Solomon. 1986) 131.0 29.2 17.0
Kotze et a1. - 1981 (1986)° 131.6 27.7 15.7
Soweto (Wagstaﬂ'et al., 1987) 133.8 28.5 15.9
European
Cape Town (Ianie and Ford. 1958) 133.1 30.4 17.2
Pretoria - 1962, 1965 (Smit, 1969) 141.3 33.2 16.6
Johannesburg - 1974 (Walker and Walker, 1977) 139.9 33.9 17.3
Richardson (1978)” 138.0 31.0 16.3
Johannesburg - 1976-1977 (Chaning-Pearee and Solomon, 1986) 139.0 33.3 17.2
Kotze et a1. - 1981 (1986 )c 139.9 33.4 17.1
Indians
Pretoria - 1964 (Smit. 1969) 134.0 26.6 14.8
Richardson (1978)b 137.0 27.5 14.7
Kotze et a1. - 1981 (1986 )‘ 135.1 28.5 16.6
“Colored”
Cape Town (Lurie and Ford, 1958) 125.6 24.1 15.3
Pretoria - 1964 (Smit, 1969) 133.2 27.6 15.2
Richardson (1978)” 137.0 28.5 15.2
Kotze et al.- 1981 (1986)° 131.6 27.8 16.1
By eumic mop“
African rural 132.8 27.5 15.6
African urban 132.9 28.1 15.9
“Colored” 131.9 27.0 15.5
Indian - urban 135.4 27.5 15.0
European urban 138.5. 32.5 17.0

 

' Calculated from mean heights and weights.

”" VduesforbodrurbmmdnnalpopulniomdnwnﬁomacmssSomhAﬁieawmmchrdedinthesesamples

‘ Theedmicyoupmeanswerecalculstedﬁ'omtbereportedms-s.
Dates when data were collected are indicated. where available

34

63?...»
2.. he 9......» a... 8.3%.: 8855...... 5 Sue Nee—race— .aeaut..< 1...... 2e .3.» 3 he 9......5 A charm

._u .o .58st .3 .o .88qu

$36: as. 6.38 ms... EBEoﬁooom 8.3.3. has? Ea 85.35 ens.
3.38% .8828 So. .8. @838. .8.

 

- 8.

. Nu.

. a.

- 8.

. was

. cam

. N2.»
3

- 42m

- on.

r a:

4 o:

- N:

 

 

35

The differences in the status of the 10 year old males in Soweto reported by
Walker and Walker (1977) and Wagstaff et a1. (1987) could indicate positive secular
changes as a result of improved living conditions and infrastructure in African urban
areas over the intervening 10 years. However, Tobias (1986, 1990) did not ﬁnd any
positive secular changes in the stature of urban adult African populations in
Johannesburg. Differences in sample composition may have affected the comparisons.
The data of Wagstaff et al. (1987) are for students from a lower elementary (grades 1 to
4) school in Soweto, while the data of Walker and Walker (1977) are for students of two
lower elementary schools in Soweto. The wide variation in SES among peOple living in
Soweto may contribute to the different observations made by Wagstaff et al. (1987) and
Walker and Walker (1977). SES variation, in contrast to improvements in the urban
infrastructure, may possibly have contributed to the difference in growth status observed
in the two samples taken 10 years apart.

None of the samples of 10 year old South African boys have a mean height or
weight above the 50th centile of the NCHS reference. Additionally, six samples (Smit,
1969; Walker and Walker, 1977; Richardson, 1977; Corlett and Woollard, 1988;
Cameron and Kgamphe, 1993) have a mean weight that is <5th percentile of the
reference. When the BMI is calculated from the mean heights and weights, one sample
of South African European boys (Walker and Walker, 1977) has a BMI > 50th percentile
of the NCHS reference and one sample (Richardson, 1978) has a BM] <5th percentile of

the reference.

36

Urban African boys are slightly heavier and have a higher BMI than niral African
boys. The range of variation between mean heights and weights of urban and rural
African samples is large. Seasonal variation in food supply, geographic and climatic
differences, and ethnicity may affect the short term food supply and thus inﬂuence body
weight; sampling variation is an additional factor.

In a comparison of mean heights, weights and estimated BMI of 10 year old
South African boys by ethnicity, all means are < 50th percentile of the NCHS reference.
Boys of European ancestry are the tallest and heaviest, and have the highest BMI.

“Colored” boys have the shortest mean height and lowest weight, and smallest BMI.

Nutrition and Infection

People in developing countries are exposed to greater infectious disease loads
than people in developed nations, and there is signiﬁcant interaction between
undemutrition and infectious disease. Both undemutrition and infection are related to
poverty, ignorance, and absence of public services (Martorell, 1982). Poverty and low
income determine the quantity and quality of food available and the preventative
measures adopted by families. Martorell (1982) also lists the lack of shoes, inadequate
shelter and housing, and limited facilities and supplies for personal hygiene as important
corollaries of poverty. Landlessness, poverty, food insecurity, high birth rates, lack of
political commitment, migrant labor, limited breastfeeding, poor maternal nutrition
education, absence of a national nutrition policy, and absence of trained health personnel
are additional contributing factors to the high prevalence of PEM in South Africa

(F incham et al., 1996).

37

Some microorganisms, such as malaria and intestinal worms, utilize signiﬁcant
amounts of nutrients for their own growth, reproduction and maintenance, reducing the
availability of nutrients to the host (Martorell, 1982; Stephenson et al., 1989, 1991).
Poor nutritional status may possibly impair the body’s defense against disease, and
children who fail to thrive often suffer from recurring infections (Suskind, 1977). The
impaired resistance may increase susceptibility to infection and create a cycle of repeated
infection and undemutrition (Mata, 1972). The interruption of the infection-
undemutrition cycle requires full immunization and aggressive work-up and treatment of
the suspected infection in addition to improved nutrition (Frank and Zeisel, 1988).
Estimates of the effects of infection and parasitic infestation on growth stunting, based on
height-for-age, suggest that one-third of total linear growth failure can be ascribed to
infection and diarrheal diseases (Allen, 1994).

Children in developing countries are exposed to large parasitic infestation loads,
which are often endemic and asymptotic. An estimated 900 to 1,300 million people in
the world are infected with at least one of the geohelminths, which includes hookworrns,
whipworrns and roundworms (Adams et al., 1994).

Monitoring infections and infestations is largely a clinical process and requires
sterile refrigeration of samples, specialized personnel, equipment, supplies and time, but
incidence and prevalence data for infection and infestation, provide an indication of the
relative impact of health problems in the area (Martorell, 1982). Clinical data, however,
may reﬂect a sampling bias, and may not reﬂect the incidence and prevalence in the

community as a whole. Individuals who go to traditional healers, or use home remedies,

38

or over the counter drugs are not included. Furthermore, the incidence of infestations
and infections are so high in some communities that they are no longer identiﬁed as
problematic and go unnoticed, and certain diseases may be classiﬁed as less of a problem
in some communities despite the endemic infection rates (Schutte, 1985; Martorell,
1982)

The prevalence of intestinal parasites in South Africa is high, and infection rates
remain high despite municipal services to minimize the risk of infections by geohelmints
(F incham, 1996). A large part of South Africa has high rates of endemic waterborne
diseases, which are particularly high in the Umgeni and Umsinduzi river catchment areas
(Appleton and Bailey, 1990). Estimates of the prevalence of Cryptosporidium infection
range between 9% (Moodley et al., 1991) and 18.4% (Berkowitz et al., 1989), while
estimates of the prevalence of helminth infestation in South African children vary
between 1.5% and 96% (Burger, 1968; Van Niekerk et al., 1979; Freeman and
Grunewald, 1980; Schutte et al., 1981) depending on the age, ethnic group and
geographic area studied. Gear and Pitchford (1985) estimate that approximately 2
million (5.3%) of all South Africans are infected with Schistosoma. However, the
majority of the infected people are asymptotic, since they come from areas where
schistosomiasis is endemic. It is estimated that 8% of rural Africans are infected with the
tapeworm Taenia solium (Schutte, 1985), with prevalence as high as 20% in some areas
(Heinz and McNab, 1965). Further, arnoebiasis accounts for thousands of deaths
annually, while malaria is endemic in many populations of South Africa (Schutte, 1985;

Sharp et al., 1988).

39

Millar et a1. (1989) identiﬁed Ascaris lumbricoides, Trichuris trichiura, Giardia
lamblia, Enterobius vermicularis and Salmonella in a random sample of patients at a
hospital in Cape Town. Of the 101 children sampled, 45.5% (46) were infested with
parasites (Table 7). The authors estimated that at least one-half of the children in Cape

Town may be at risk of complications from intestinal parasitic infestations.

Table 7. Parasitic infestation rates of hospitalized children in Cape Town.‘

 

 

Parasite Number Percent of total (%)
Ascaris lumbricoides 8 17.4
Trichuris trichiura 8 17.4
Ascaris lumbricoides + Trichuris trichiura 20 43.5
Giardia lamblia 5 10.9
Giardia lamblia + Ttrichuris trichiura 3 6.5
Enterobius vermicularis l 2.2
Salmonella 1 2.2

 

' Adapted from Millar et a1. (1989)

Rates of parasitic infections in South African populations are high. A prevalence
of 98% for Trichuris and 40% for Ascaris was noted in a West Coast South African
community, which had puriﬁed drinking water supplied to all homes, ﬂush toilets in 98%
of the homes, a modern sewage disposal plant and regular garbage collection, free
antihelmintic treatment at the community clinic, and electricity in 85% of the homes
(F incham et al., 1996). The high rates of infection were also evident in the prevalence of
Schistosomiasis eggs in urine samples of South African canoeists (Appleton and Bailey,
1990). During 1988 and 1989, 70.7 % and 48.7% of canoeists, respectively, were

diagnosed positive for schistosomiasis.

40

Despite the high prevalence of Schistosoma haemotobium, schistosomiasis is not
considered as a signiﬁcant cause of death in South Africa and is given low priority in
terms of public health. Malarial infection, on the other hand, is not as high as
schistosomiasis or geohelminth infections, but is the parasitic disease that is given the
highest priority in South Africa (Schutte, 1985). Few studies (Connolly and Kvalsig,
1993; Fincham et al., 1996) have addressed the relationship between parasitic infection
and the nutritional status in South Africa.

Connolly and Kvalsig (1993) found that 55.5% of children with trichuris and/or
lumbricoides infections had height-for-age z-scores s -1 compared to 19.7% of children
without parasitic infection. Only 33% of children with parasitic infections had height-
for-age z-scores 2 +0.5 compared to 52.5% of children who were not infected. The
results suggest that a greater percentage of children with parasitic infections were stunted
compared to those without infections. On the other hand, Stephenson et a1. (1991)
reported signiﬁcant gains in both mass and stature in severely infected Kenyan boys four
months after receiving a single dose antihelminthic treatment. However, in a subsequent
study, treatment for intestinal parasites increased weight gain but did not improve linear
growth of Kenyan school children (Stephenson et al., 1993).

Ghesquiere and Eeckels (1984) noted SES and ethnic variation in the prevalence
of parasitic infestations in elementary school boys (6-13 years of age) from Kinshasa,
Zaire (Table 8). Lower SES African boys had a higher prevalence of infestation, with
97% of the sarnple. In a series of studies on the effects of treatment of helminth

infections on growth, nutrition and physical activity of Kenyan elementary school

41

children, Stephenson et a1. (1989, 1991, 1993) found that physical ﬁtness, food intake
and appetite of children improved following treatment for helminth infections compared
to a control sample. The physical ﬁtness of elementary school boys with hookworm,
Trichuris trichiura and Ascaris lumbricoides who were treated with albendazole was
assessed. The Harvard step test, heart rate, and heart rates at 1, 2, 3 and 4 minutes after
the step test was measured at baseline and four months after a single dose of albendazole
(Stephenson et al., 1991). Boys who received the antihelmentic treatment showed

signiﬁcantly better performances on the Step test, while the placebo group showed no

 

 

changes.
Table 8. Prevalence of infestations in school boys in Kinshasa.’I
African: Lower SES African: Upper SES Ern'opean: Upper SES
(Shanty Town) (Private School)
Abdominal distases 72.0% 0.0% 0.0%
Malaria 45.0% 1.5% 0.4%
Intestinal parasites 97.0% 63.0% 20.0%

 

' Adapted from Ghesquire and Eeckels (1984).

Data on physical activity and parasitic infections are mixed. Kvalsvig (1981)
found an inverse relationship between the severity of parasitic infection and physical
activity, with lower levels of physical activity in severely infested children. On the other
hand, Walker et a1. (1972) found that the running performance of 14 year old African
boys (South Africa) with Schistosoma haemotobium infections was similar to that of
uninfected African boys from the same areas. Kvalsvig et a1. (1991) caution that some of

the discrepancies in the literature on physical activity and parasitic infestation may be

42

 

due to a priori differences in the groups. More active children have greater exposure to
parasitic agents since they engage in play for longer periods of time increasing the risk of
exposure to the parasite. On the other hand, severely malnourished or severely infected

children are less active.

Catch-up growth

SES, ethnicity, and nutritional status affect the growth of children. This raises the
question: Does the trend towards growth stunting follow an invariant negative path?

Martorell and coworkers (1990) posit that there is a limited ability for the
chronically malnourished child to catch-up, since the stunting in height appears to be
much greater than the delay in the appearance of the number of ossiﬁcation centers in
rural Guatemalan children. Observing a relationship between height at age 3 and ﬁnal
adult height, the authors concluded that stunting is a condition that results from events in
early childhood and persisted with no catch-up growth in childhood or adolescence
(Martorell et al., 1990). These ﬁndings are consistent with studies in South Africa
(Chaning-Pearce and Solomon, 1986) and India (Satyanarayana et al., 1989) .

There is evidence to indicate that the efﬁcacy of nutritional supplementation may
decrease after the ﬁrst 3 years of life (Satyanarayana et al., 1989; Martorell et al., 1990;
Martorell, 1993; Perez-Escamilla and Pollitt, 1995). It has also been argued that
providing nutritional interventions after the ﬁrst three years of life has little or no
potential to enhance physical growth (Y arbrough et al., 1975; Gillespie and Mason,
1991). On the other hand, some evidence indicates that enhancing nutritional status after

three years of age may induce positive changes in stature (Martorell, 1993; Lampl et al.,

43

 

 

1978), stature and mass (McKay et al., 1978), and weight-for-age and height-for-age
(Stephenson et al., 1993). Data from the INCAP study in Gautemala indicated that the
greatest growth stunting occurred during the ﬁrst 3 years of life. This was also the
period during which the greatest responses to the nutritional interventions were most
likely to occur (Rivera et al., 1995). Thus, the size of the growth response to
supplementation introduced after, compared to before, 3 years of age, was smaller.

A major reason for the disparity may be the period of time over which the follow
up was conducted. In the initial Guatemalan study, subjects were followed for the ﬁrst 7
years of life. Although children from the high protein-energy (Atole) villages were taller
and heavier at three years of age compared to the energy only (Fresco) villages, there was
little or no observable difference in the growth between the groups from 3 to 7 years of
age (Martorell, 1990). When the same subjects were followed in late adolescence and
adulthood (11 to 26 years of age), signiﬁcantly greater stature, body mass and FFM were
observed in the both the Atole and Fresco groups compared to nonsupplemented
individuals from the same population (Rivera et al., 1995). At the follow-up, subjects
from the Atole villages were taller and had greater estimated FFM than subjects from the
energy only villages. The statistically signiﬁcant differences in height and estimated
FFM, equivalent to a positive shift of 0.5 SD, were not statistically signiﬁcant when
adjusted for height at age three.

Interventions introduced prenatally and during the ﬁrst three years of life in rural

Guatemala continued to produce beneﬁts in growth years after the termination of the

intervention (Rivera et al., 1995). Additionally, a 0.7 SD positive shift in 90, m , both

 

 

absolute and per unit mass, was reported (Haas et al., 1995) and the differences in Y0;
m, (absolute and per unit mass) could not be explained by the differences in estimated
FFM.

There was a statistically signiﬁcant positive difference of 0.6 SD in the
intellectual performance of the high protein-energy compared to the energy group (Pollitt
et al., 1993). Subjects who received the high protein-energy supplementation during
gestation and for the ﬁrst 2 years of life also had a higher score in intellectual
performance as adolescents (13-19 years of age) than those who began supplementation
later in life. There also was a strong indication that the effects in adolescence were found
only in cohorts exposed to supplementation during pregnancy and the ﬁrst 2 years of life.

Pérez—Escamila and Pollitt (1995) noted that gains in stature and mass associated
with nutritional supplementation of Colombian children were in proportion to the
duration of supplementation, regardless of the age (38, 42, 59, or 75 months of age) at
which the nutritional intervention was introduced. Unlike the results for Guatemalan
children (Rivera et al., 1995), the sample of Colombian children (Perez-Escamila and
Pollitt, 1995) did not continue to demonstrate gains in height and weight after the
cessation of supplementation. The Colombian subjects, however, were not followed into
adolescence and adulthood.

Satyanarayana et al. (1989) found that adolescent boys who were severely
malnourished at 5 years of age reached puberty later, had a signiﬁcantly lower growth
rate, a longer period of adolescent growth, and similar adolescent gain in stature to a

British comparison group. The results suggest that undemutrition during childhood did

45

not add to height deﬁcits during puberty in this sample of Indian children. Cameron and
Kgarnphe (1993) found similar delays in the onset of and a prolongation of puberty in
rural African populations in South Africa.

It is posited that preschool growth is largely determined by environmental factors
such as nutritional status, SES and disease. Undemourished populations tend to have
depressed preadolescent growth which accumulates throughout childhood, but the
majority of the variation is established before six years of age. Although, adolescent
growth is largely controlled by separate genetic factors (Tanner, 1962), negative
environmental stresses in childhood may result in later puberty, and slower and longer
adolescent growth (Chaning-Pearce and Solomon, 1986, 1987; Cameron et al., 1990).
‘Colored’ South African males and females who were treated for kwashiokor between
the ages of 5 months, and 4 years and 4 months had delayed sexual maturation (Cameron
et al., 1988; 1990). Additionally, lower SES Africans have a longer pubertal growth in
stature and mass compared to Europeans (Chaning-Pearce and Solomon, 1986), and also
have a later onset and prolonged duration of pubertal growth (Chaning-Pearce and
Solomon, 1987). Low growth velocities during childhood result in mean heights and
weights that are < 5th centile by the beginning of the adolescent growth spurt for
approximately 50% of South African children (de Villiers, 1987; Cameron et al., 1988;

Smit, 1969).

Summary

There is a strong association among ethnicity (race), SES and growth status in

South Africa. Urban-rural status is also associated with SES and affects growth status.

46

 

 

The available literature indicates that rural African populations are lighter and have a

lower BMI than urban African populations. Both urban African and Indian populations
are shorter and lighter than European South Africans. The data for Indians, however, are
limited. Variations in growth may be affected by nutritional status and infectious
diseases, and repeated bouts of infectious diseases and may contribute to growth stunting.
Addressing the growth of any population in South Africa requires addressing the
complex interrelationships among ethnicity, SES and nutritional status. Nutritional
status differences have to be accounted for independently of the effects of ethnicity (race)

and SES to partition the effects of each of these dynamic variables.

Physical ﬁtness

The relationship between habitual physical activity and physical ﬁtness of the
child and adolescent is of relatively recent interest (Mirwald and Bailey, 1986; Pate et al.,
1990; Baranowski et al., 1992; Shephard, 1992; Malina, 1995). Most of this interest was
spurred by the Healthy People 2000 guidelines, which implied that many United States
children and youth were physically unﬁt or insufﬁciently active and were at increased
risk for cardiovascular disease later in life (Malina, 1995). This formed the impetus for
health-related physical ﬁtness which focused on developing physiological ﬁtness to
prevent premature morbidity and/or mortality due to a sedentary life style. The emphasis

was largely on developing the ﬁtness of youth for potential health beneﬁts in adulthood

(Malina, 1995).

47

Spurr (1983) estimates that only 9% of the variance in 1:702 M, is related to
physical activity. However, there is a positive association between higher ﬁtness and
higher general activity. Fourth grade American children (mean age of 10.5 years) with
higher general physical activity levels have higher physical ﬁtness (Simons—Morton et al.,
1990). Similar results are reported for a small sample (15 boys and 10 girls) of 6th grade
(mean age 12.3 years) rural German children (Schmucker et al., 1984). There were
signiﬁcant moderate to low correlations between habitual physical activity and strength
(r=0.42), aerobic capacity (r=0.67), basic Skills (r=0.55), motor coordination (1:048)
and motor communication (r=0.34) for the combined sample of males and females.

Motor performance is related to body size in both well-nourished and
undernourished children (Malina, 1975, 1994; Malina and Little, 1985; Malina et al.,
1987). Mild-to-moderate undemutrition is also characterized by reduced fat-free mass
and fat mass (Barac-Nieto et al., 1978; Satyanarayana et al., 1979; Desai et al., 1984;
Malina et al., 1991), reduced bone density and metacarpal cortical thickness (Himes,
1978), and reduced strength, motor performance and aerobic ﬁtness (Barac-Nieto et al.,
1978; Spurr et al., 1983a, 1983b, 1991; Malina and Bushang, 1985).

There is reasonable consensus that undernourished children score lower on tests
of performance- and health-related ﬁtness, including speed, power, strength, and
endurance compared to nutritionally normal children (Sloan, 1966a; Sloan and Hansen,
1969; Ghesquiere and Eeckels, 1984; Malina, 1984; Malina and Buschang, 1985; Malina
et al., 1991; Goslin and Burden, 1986; Beneﬁce, 1993). Motor skill acquisition and

performance is inﬂuenced by cultural and environmental exposure to the skill, and the

48

growth and maturational status of the individual, among other factors. There is,
however, considerable variation in the normal range of motor skill attainment and the
rate of development is inﬂuenced by a variety of environmental factors (Branta et al.,
1984)

Physical ﬁtness is separated into health-related and performance-related ﬁtness.
Performance-related ﬁtness is an indication of the attained levels of motor performance.
Static strength (grip strength), agility (shuttle runs), nrnning speed (dash) and explosive
power (standing long jump) are components of performance-related ﬁtness. Current
emphasis is more on health—related ﬁtness, which is a state characterized by an ability to
perform daily activities with vigor and demonstrate traits and capacities that are
associated with low risk of premature development of the hypokinetic diseases (Pate,
1988). It is operationalized as cardiorespiratory endurance (9 minute run),
musculoskeletal function (sit and reach, push-ups and sit-ups), and body composition
(subcutaneous fatness) (AAPHERD, 1984).

The development of behavioral competence in motor tasks is a major
developmental task of childhood and youth. Fundamental motor patterns are established
during early childhood, conditioned by the interaction of genotypic and environmental
factors, and inﬂuenced by the opportunity to practice the new skills (Malina, 1992).
Motor skill implies both economy and precision, and is related to age, body size and
composition, opportunity for practice, and motivation, among other factors (Malina et al.,

1987)

49

Eight studies of the physical performance and/or ﬁtness of South African children
and adolescents, which spanned approximately 50 years from 1942 to 1992, a period
over which considerable sociopolitical restructuring has occurred in South Africa, are
reviewed. Two of the studies from the 19405 provide a reference for changes in
performance status over the past 50 years, while three studies are from the 19605 and
another three are from the 19805. Subjects are separated according to the prevalent racial
groups of African, “Colored”, Indian and European. but Indians are included in
“Colored” samples in some studies.

Most of the South African literature focuses on the performance-related physical
ﬁtness of pubertal children, with a limited data for prepubertal children. The samples are
drawn from high school populations, which are not necessarily representative of the
general population. Until 1992 education was free and compulsory only for Europeans,
and African education received the least federal funding (Table 1). The cost of education
often limited school attendance for African children. Education also became
progressively selective with each successive year of formal schooling. Consequently,
high school attendance was limited to individuals from families of higher SES.
Additionally, most African and “Colored” school students worked to subsidize their
schooling, dropping out of school for a year or two as dictated by the economic situation
of the family.

The studies of the performance-related physical ﬁtness of youth of different racial
groups point to higher scores for European post-pubertal males compared to African,

“Colored” and Indian youth. There are also differences in the performances of

50

prepubertal, circumpubertal and pubertal males relative to body size and ethnicity. Early
studies (Culver et al., 1942; Botha et al., 1945) indicated that, despite the higher
incidence of undemutrition and parasitic infestations, African boys classiﬁed as
prepubertal had superior performances on the 100-yard dash and 600-yard run compared
to European, “Colored” and Indian prepubertal boys. On the other hand, prepubertal
European boys had superior performances on the shot put compared to prepubertal
African, “Colored” and Indian boys. Botha et a1. (1945) attributed the superior shot put
performance of European boys to their larger body size and better neuromuscular
development. The criteria for the classiﬁcation of pubertal status, however, was not
speciﬁed.

In early studies, “circumpubertal” (immediately before and aﬁer puberty)
European boys had superior running performances on the 100-yard dash and 600-yard
run compared to African boys (Culver et al., 1942). Sloan (1966a, 1966b) found a
positive association between the performances of pubescent European boys and estimated
muscle mass. This correlation probably reﬂects the larger body size and advanced
maturity status of European compared to African, “Colored” and Indian males. On the
other hand, the performances of prepubertal boys on the 10-yard dash and 600-yard run
were negatively correlated with body size (Culver et a1, 1942; Botha et al., 1945). The
results suggest that while strength appears to be a signiﬁcant determinant of performance
on the 100-yard dash and 600-yard run during puberty, other factors such as running

economy and skill level may have determined preadolescent performances.

51

Greater body size is associated with better physical performances in some tasks.
Several studies (Sloan, 1966a, 1966b; Goslin and Burden, 1986) indicated superior
performances for European pubertal adolescents, who were consistently taller and
heavier than boys from other racial groups. This raised the question: Was the superior
post-pubertal performance of European males related to their greater body size? In a
sample of 12-18 year old males, Sloan (1966a) observed that European males were taller,
heavier and demonstrated superior performances on sit-ups, the shuttle run, standing long
jump, 50-yard dash, softball throw for distance, and 600 yard run-walk (AAPHER ﬁtness
test) compared to African and “Colored” (which included Indian) males of the same age
(Table 9). “Colored” males performed better than African males on most tests, despite a
lower body weight. African boys had the poorest performances on all items except for
the softball throw for distance. Thus, differences in performance appear to be affected
by factors other than body size. Body composition, skill and practice were not
considered as possible sources of variation in performance, in these early studies

Fitness indices, calculated from age-speciﬁc T-scores for the 6 items of the
AAPHER tests of European, African and “Colored” males were compared by Sloan and
Hansen (1969). European males had a superior Fitness Index compared to African and
“Colored” boys and the performance-related ﬁtness did not appear to be related to body
size. Statistically controlling for the effects of stature and body mass had no effect on the
Fitness Index. The relationship between body size and ﬁtness, however, appeared to
differ among the racial groups. When the effect of weight was controlled, heavier 13

year old European males had a higher Fitness Index compared to lighter European males,

52

Table 9. Comparison of the mean body size and performance-related physical
ﬁtness of adolescent South African males 12 to 18 years of age by age and race.‘I

 

 

Age in Years 12 14 16 18
Stature (cm)

European 152 164 177 183

“Colored” 144 156 164 170

African 154 162 172
Weight (kg)

European 41 51 63 67

“Colored” 36 43 52 58

African 44 56 60
Sit-ups (# to exhaustion)

European 64 80 81 88

“Colored” 47 47 50.5 49

African 28 34 30
Shuttle run (sec)

European 10.8 10.2 9.9 9.5

“Colored” 10.9 10.7 10.3 10.1

African 10.9 10.3 10.2
Standing broad jump (cm)

Emopean 177 198 222 231

“Colored” 171 183 204 204

African 171 183 198
50 yard dash (sec)

European 7.8 7.1 6.6 5.9

“Colored” 7.9 7.4 7.2 7.0

African 8.0 7.5 7.8
Softball throw (m)

European 12.7 15.8 20.2 23.9

“Colored” 12.2 14.6 16.8 17.3

African 14.9 17.3 19.2
GOO-yard uni-walk (minzsec)

European 2:01 1:53 1:45 1:40

“Colored” 2:05 1:55 1:50 1:46

African 2:40 3:06 2:30

 

' Adapted from Sloan (1966a). Items are from the AAPI-IERD ﬁtness test

53

indicating an advantage of weight. On the other hand, heavier 13 year old “Colored”
boys had a lower Fitness Index than lighter “Colored” boys, indicating a disadvantage of
weight. Body composition and maturity status potentially affect the Fitness Index.

Taller and heavier European males were likely advanced in maturity status with a higher
percentage of muscle tissue compared to “Colored” and lighter European boys, which
perhaps contributed to the performance advantage of their higher body mass.
Additionally, the apparent race difference in the relationship between body size and
ﬁtness could be due to the use of an index. The Fitness Index is a composite score of test
items and does not reﬂect individual variation on speciﬁc test items. Thus, strengths and
weaknesses on individual test items are masked by the Fitness Index.

SES, training, experience and nutritional status are some of the factors that may
affect the performance-related ﬁtness of South African boys. European schools had
gymnasiums and trained physical education teachers, whereas Black schools had neither.
Additionally, European schools had compulsory participation in one summer and one
winter sport that were practiced for 3 to 4 afternoons per week (Sloan and Hansen,
1969). Black schools had no formal physical education curricula and extramural sport
was offered once a week. The consistently poorer performances of African, “Colored”
and Indian, compared to European, subjects may thus be related to differences in habitual
physical activity and access to the opportunity for instruction and practice.

While skill and training appear to be important determinants of performance in
nutritionally normal subjects, evidence indicates that strength and performance are

related to body size in undernourished populations. The absolute performances of

54

undernourished children are generally lower than those of well nourished children
(Culver et al., 1942; Botha et al., 1945; Ghesquiere and Eeckels, 1984; Malina and Little,
1985; Malina et al., 1987; Beneﬁce, 1992; Beneﬁce et al., 1996). However, many of the
observed differences are decreased or are reversed when performances are expressed per
unit body mass. This suggests that the poorer performances of undernourished youth are
more related to reduced body size rather than to differences in metabolic processes.

Absolute grip strength is lower in undernourished children, but is similar for
nutritionally normal and undernourished children per unit body mass (Malina et al.,
1987). The absolute strength scores of European males in South Africa are comparable
to American children of the same age (Goslin and Burden, 1986). African and “Colored”
children, on the other hand, have lower absolute strength compared to American children
on the same age. But, when expressed relative to body mass, the mean grip strength of
African and “Colored” boys is greater than that of European boys (Table 10).

The dash and long jump performances of moderately undernourished Zapotec
Indian (Mexico) children (6-15 years old) were proportional to smaller body size in
younger but not in older boys (Malina and Buschang, 1985). When performances of
Kinshasa (Zaire) elementary school children (6-13 years of age) were expressed per unit
body size, boys from a lower SES group performed slightly, but consistently, better on
the medicine ball throw, hand grip, vertical jump, and 50 m dash (Ghesquiere and
Eeckels, 1984).

Despite lower body mass and estimated FFM of undernourished children, they

have higher absolute scores on some tasks (Malina and Buschang, 1985; Ghesquiere and

55

Table 10. Comparison of the physical ﬁtness (means) of adolescent South African
boys by race.‘I

 

 

 

European “Colored” African
Age (years) 16.8 18.1 18.5
Activity level 10.0 9.0 9.5
Mass (kg) 68.0 59.0 61.0
Stature (cm) 177.0 169.5 172.0
Grip (kg) 89.0 81.0 90.0
Grip (kg/kg body mass) 1.3 1.4 1.5
Power mans)" 1116.0 895.0 725.0
Power (watts/kg body mass)” 16.4 15.2 11.9
Medicine ball (45kg) push (cm) 210.0 185.0 155.0
Standing long jump (cm) 216.0 210.0 190.0
Sit-ups (#lmin) 45 38 30
Half hold push-11p (sec) 49 68 52
Sit, reach and hold (cm) 31.5 35.0 35.5
Shuttle run (sec) 12.0 11.0 12.2
Estimated body fat (%) 12.0 11.0 11.0
Predicted V0, m_,,, (m1.lrg".rnin°‘ ) 61.0 52.0 49.0
; Adapted from Goslin and Burden (1986)
Heart rate not indicated

Eeckels, 1984). Performances on the standing long jump appeared to be independent of
body size in undernourished adolescents. Superior standing long jump performances
were reported for adolescent “Colored” girls despite their smaller body size (Goslin and
Burden, 1986). Slightly better throwing performances were observed in marginally
undernourished Zapotec Indian boys in Mexico (Malina et al., 1987) and African boys
(Sloan, 1966a). Malina et al. (1987) suggest that the higher performance scores may
have been related to speciﬁc cultural practices. Goslin and Burden (1986) also suggest
that ethnic variation in level of skill may be due to individual cultural experiences of
children.

Time related trends in physical ﬁtness, based on a comparison of studies in 1966
and 1985 indicated no major differences between the stature of 16 year old European and

18 year old African and “Colored” males (Table 11). European boys were heavier in

56

1986 by 5 kg, but “Colored” and African males had similar body weights in 1966 and
1985. Performance of African, “Colored” and European boys on a shuttle run and the
standing long jump were lower in 1986. Shuttle-run performances declined by 17.5%
and 16.4% for European and African males, respectively, in 1986. “Colored” males had
the best shuttle-run performances in 1986, despite the 8% decline in performance from
1966. European males had greater explosive strength on the standing long jump in both

1966 and 1986, compared to African and “Colored” boys. “Colored” males performed

Table 11. Comparison of age, body size and physical performances (means) of
adolescent European, ‘Colored’ and African South African males between 1966 and

 

 

 

l985.'I
Sloan (1966) Goslin and Burden (1986)
Age (years)
European 16.5 16.8
“Colored” 18.5 18.1
African 18.5 18.5
Stature (cm)
Elnopean 177 177
“Colored” 170 169
African 172 172
Weight (kg)
European 63 68
“Colored” 58 59
African 60 61
Shuttle nm (sec)
European 9.9 12.0
“Colored” 10.1 11.0
African 10.2 12.2
Standing long jump (cm)
European 222 216
“Colored” 204 210
African 198 190
”Time for the 50 meter dash

57

better than African males on the standing long jump, both in 1966 and 1986. However,
the difference between the performance of European and “Colored” males on the
standing long jump was smaller in 1986 compared to 1966.

The lower mean performances in 1986 could be related to lifestyle changes which
coincided with the introduction of television in 1974. The reduced physical ﬁtness and
physical activity of American children are often attributed to increased television viewing
(Dietz and Gortrnaker, 1985). This assertion assumes that activity and performance are
themselves related. While there is no conclusive evidence to support these contentions,
the initial introduction of television into a society could have had some effect on overall
activity levels, but does not necessarily translate into decrements in physical ﬁtness.
Additionally, South African European boys were heavier in the 1986 study, and the
increasing adiposity of American children and youth has been attributed to television
viewing (Dietz and Gortrnaker, 1985). However, cohort differences between 1966 and
1986 possibly contribute to the variation in performance.

Some evidence indicates a correlation between high levels of habitual physical
activity and submaximal endurance capacity. Participation in physical activity in non-
summer months, involvement in a varied physical education curriculum, weekly activity
time in physical education, and involvement in lifetime physical education were some of
the factors that had a signiﬁcant relationship with the performance on the 1 mile
walk/run of 10 to 12 year old children from the NCYFS (Dotson and Ross, 1985). No

speciﬁc correlations were given, however. Students scoring below the average range

58

also reported participation in fewer high intensity physical activities compared to
students who performed in the optimal range.

Studies on the relationship among ethnicity, performance-related ﬁtness, SES,
and nutritional status of South Africans are not conclusive (Sloan, 1966a, 1966b; Sloan
and Hansen, 1969; Goslin and Burden, 1987; Badenhorst et al., 1992). Ghesquiere and
Eeckels (1984), however, found a strong correlation among SES, ethnicity and
performance of boys from Zaire. Children of European ancestry were of higher SES,
taller and heavier, and had higher absolute performances compared to boys of African
ancestry (Ghesquiere and Eeckels, 1984). Similarly, on average, African boys (South
Africa) were of lower SES, shorter and lighter, and had lower absolute performance
scores on the sit-ups, shuttle run, standing long jump, 50-yard dash (Sloan, 1966a),
medicine ball push and tests of absolute and relative power compared to European boys
(Goslin and Burden, 1986). The difference in physical ﬁtness are confounded by the
strong race by SES interaction, which in turn, affects nutritional status and could be due

to an SES by nutritional status interaction as opposed to race.

Work capacity and {(02 m

Work capacity and maximal oxygen capacity (9'02 mu) are affected by body size,
age, maturity status, body composition, training and nutritional status, among other
factors (Krahenbuhl et al., 1985). In an examination of the 1702 m of nutritionally
normal White, Mestizo and Black Colombian boys 6-16 years of age, no signiﬁcant age,

race, or SES differences in absolute YOZM were observed (Spurr et al., 1983b).

59

However, low weight-for-age and low weight-for-height children had markedly lower

absolute YO; m, (l.min'1) compared to nutritionally normal children. When YO; m, was

adjusted for body size (ml.kg'l.min'1), marginally malnourished children had equal or
higher relative 170; m (Table 12). At the same absolute workload, undernourished
children performed at a relatively higher workload (51 to 54% of Y0; m ) compared to

normally nourished (41% of 170; mm‘) children (Spurr, 1983; Spurr et al., 1983b).

Table 12. Absolute and relative 1:702 m of normally nourished, low weight-for-age
and Iow-weight-for-height 6 to 16 year old Colombian boys.‘l

 

 

Oxygen consumption Control Low W-Ab Low W-H°
Weight for age > 95% < 95% < 95%
weight for height > 95% > 95% < 95%
V0,”, (min-1) 1.8 1.53 1.48
Yo,“ as a % of control group (1.min") 35% 82%
V02... (m1.kg".rnin") 52 53.89 55.43

.

V92 m, as a % of control W".min") 104% 107%

 

 

' Adapted from Spurr et al. (1983b)
” W-A = Weight-for-age
‘ W-H = Weight-for-height

It is generally assumed that functional capacity is impaired by marginal
undernutrition. The relationship between nutritional status and physical working
capacity (PWCm) of 12-year old migrant and well-to—do boys from Ribeirao Preto, Sao
Paulo (Brazil) was examined by Desai et a1. (1984). The migrant boys had a mean
nutrient intake that was 62.5% of the FAO/WHO recommended daily intake and 85% of
the control group of well-to-do boys, respectively. Although the well-to-do controls had
caloric intakes that were lower than recommended daily allowances, their protein intake

was within the normal range. Baseline heart rates at rest were the same for the two

60

groups. Compared to the well-to-do controls, the undernourished boys had higher mean
heart rates at each submaximal workload on the bicycle ergometer which remained
signiﬁcantly higher during the ﬁrst two minutes of recovery, higher oxygen consumption
per unit body weight, and higher levels of muscle lactate (Table 13). Overall,
undernourished boys worked at a higher relative workload for the same absolute
workload compared to well nourished age mates, and had greater muscular stress than the

well-to-do controls.

Table 13. Oxygen consumption of migrant and well-to-do 12 year old Brazilian

 

 

boys.‘

Oxygen consumption Workload (Watts) Well-to-do Migrant
{702... (m1.rnin") 0 209 190
Yo,“ (m1.rnin") %orw.t-d" ' 91%
V02 m (ml.kg".min") 4.8 5.6
V0; m (m1.kg".min'l) % of w-t-d 117%
{(02 m (ml.min") 75 1011 987
902..., (ml.rnin") %of w-t-d 98%
V0, m @111ng min") 23.1 29.0
V9 m (m1.kg".rnin") %of w-t-d 126%

 

' Adapted from Desai et a1.(1984)
” w-t-d= well- to do

Stature and mass are signiﬁcant determinants of the physical work capacity of
(PWC) and the relationship is more marked for undernourished individuals. The PWC of
marginally undernourished Brazilian boys was 71% that of a same-age control group
from the same town, but increased to 91% when expressed per unit mass (Desai et al.,
1984). Body mass alone, however, accounted for only 16% of the variation in work

capacity of marginally malnourished males, but when stature was added to mass, 57% of

61

the variance was explained. On the other hand, 64% of the variance in physical work
capacity of severely malnourished adolescent Indian boys, classiﬁed into three nutritional
groups based on stature at 5 years of age, was explained by mass at the time of testing
(Satyanarayana et al., 1979), which indicated a greater inﬂuence of mass on work
capacity in severely malnourished subjects.

Undemourished rural African boys from Johannesburg, with an estimated mean

daily nutrient consumption between 37-41% of the recommended daily allowances

(Badenhorst et al., 1992), had signiﬁcantly lower absolute and relative {’02 m (Table
14) compared to normally nourished boys from Canada (Mirwald and Bailey, 1986) and
Colombia (Spurr et al., 1991). Despite undemutrition, this sample of boys was relatively
taller than all other samples of South African boys. The disparity between growth and
nutritional status of the rural African males may have been due to under-reporting of
food intake on the 24-hour dietary recall, or seasonal variation in diets that were typical

of rural populations, or perhaps the result of a very select sample.

Table 14. Absolute and relative 1702..., of boys with a mean age of 10.5 years.

 

902.... drum") 1702.... (m1.kg".min“)

 

South Africa (Badenhorst et al., 1992) 1.55 49.7
Saskatchewan, Canada (Mirwald and Bailey, 1986) 1.80 56.2
Argentina (Perez et al., 1991)

Lower SES 1.13 36.2

Upper SES 1.50 43.0
Colombia (Spurr et al., 1991)
Urban - normally nourished 1.66 53.6
Rm'al - normally nourished 1.62 52.3

 

62

It has been argued that walking and bicycling, which are used as a primary mode
of transportation for African and “Colored” children and youth, should have a training
effect on work capacity (Wyndham, 1963). Undemourished rural African boys walked
or cycled between 5 to 15 km to and from school each day, but their 1.702 m, was
considerably lower than nutritionally normal age mates (Badenhorst et al., 1992).
Furthermore, {’02 m, was consistently higher in European than in African adolescents
(Goslin and Burden, 1986) and adults (Wyndham, 1963). The differences in 1702“,“
could be due to a training effect of higher intensity extramural activity as opposed to the
lower intensity of walking and cycling which did not provide a signiﬁcant training effect.

Physical ﬁtness (1.6 km run/walk) and habitual physical activity had a low
(R2=0.21), but signiﬁcant relationship in 8 to 9 year old children from the NCYFS II
study (Pate et al., 1990). Some of the variation in physical ﬁtness was accounted by the
global rating of the child’s activity level (11%), participation in vigorous community
activity (5%), participation in physical education at school (2%) and involvement in
private clubs/lessons and television viewing (1%). Habitual physical activity, regardless
of mode, appears to have a relatively small positive effect on the performance on the one
mile run/walk test. Schmucker (1984) reported a moderate correlation (r=0.67) between
habitual physical activity and aerobic capacity for a combined sample of rural German
boys and girls (mean 12.3 years of age). The 1.6 km run and 902,“, do not necessarily
measure the same thing. Other factors that could explain the differences in between
habitual physical activity and endurance performance possibly include lifestyle factors

and sample size. The higher correlation between habitual physical activity and

63

endurance performance in the latter study may be due to lifestyle differences. Children
from rural Germany possibly engage in higher levels of habitual physical activity
compared to children from the USA and this may possibly influence endurance
performance. The German study has a small sample size (n=25) and possibly affects the
association between habitual physical activity and endurance performance.

Satyanarayana et a1. (1979) noted a moderate correlation (r=0.53) between
habitual physical activity and physical work capacity (PWCm) of chronically
undernourished adolescent Indian males, even after controlling for the effects of body
mass. Physical activity status was the single parameter that had a signiﬁcant relationship
to work capacity per unit body mass, and 10% of the variation in PWCm could be
explained by habitual physical activity. Satyanarayana et al. (1979) postulated that the
differences in work capacity could be due to activity induced changes in the
cardiovascular and neuromuscular system. On the other hand, nutritional status history -
the duration and severity of undemutrition - could have predisposed the individual
towards lower levels of habitual physical activity which, in turn, resulted in a lower
submaximal work capacity.

Barac-Nieto et a1. (1978) found a linear relationship between the degree of

undemutrition and Y0, m of adult Colombian men. The more nutritionally

compromised the individual, the lower the ’902 m . This relationship did not change
with adjustments for body mass, height, and estimated FFM, body cell mass, or muscle

mass. This study did not take into account the potential self-selection of lower levels of

habitual physical activity by adults with lower 1’70, m. Marginally undernourished

64

individuals are able to perform the same low levels of work as normal individuals, but at
a relatively higher level of energy expenditure (Spurr et al., 1977; Barac-Nieto et al.,
1978; Satyanarayana et al., 1979; Desai et al., 1984). The energy consumption required
to sustain a normal daily workload was estimated to be approximately 40% of {(02 mu.
Spurr et al. (1983b) hypothesized that working at even 40% of Y0; ,m, undernourished
subjects would be less productive adults, on average. Energy deﬁciency related
decreases in 1702 max may result in a lower ability to sustain a given absolute work load

and in the time that a given relative effort could be sustained (Barac-Nieto et al., 1978).

Summary

Body size, SES, race and nutritional status appear to affect physical activity and
ﬁtness. SES, race and nutritional status are interrelated in most developing world
populations. Thus, while SES, race and nutritional status differences in physical ﬁtness
and habitual physical activity have been observed in several populations, race-, SES- and
nutritional status-speciﬁc differences are most likely confounded by each other. Some of
the trends observed include lower scores on tests of speed, power, strength and endurance
of undernourished compared to nutritionally normal children. Race differences in
physical ﬁtness and habitual physical activity are evident among children from South
Africa and Zaire, but these were not systematically analyzed in a SES context.

In South Africa, African boys have poorer performances on performance-related
ﬁtness tests compared to European boys, except on the dash and distance run. Physical

performance is related to body size, with greater size associated with better performances

65

on some items. The apparent race differences in ﬁtness may be due to interations of
SES, race and growth status since European boys also have a larger body size and scored
higher on ﬁtness tests. Similar race, SES, nutritional status and growth interactions were
found in Zaire. The relationships, however, are not clear. In South Africa, despite being
lighter than African boys, Colored boys have better ﬁtness scores than African boys

Skill training and the opportunity to practice are factors that inﬂuence
performance on some ﬁtness tests, and may be related to SES factors. In addition,
lifestyle factors such as mode of transportation and habitual physical activity could
contribute to differences in endurance and cardiorespiratory ﬁtness. There is some
evidence to support the contention that higher habitual physical activity results in higher
submaximal endurance capacity. But, undernourished populations have a lower physical
work capacity and work at a relatively higher workload for the same absolute load
compared to nutritionally normal populations, which suggest that their work capacity
may be compromised.

Differences in physical ﬁtness and habitual physical activity need to be examined
within the context of SES, race, current nutritional status, and nutritional history. The
independent contributions of SES, race and nutritional status to speciﬁc ﬁtness variables

also need to be explored.

Habitual Physigal Agﬁvity
Physical activity is a component of total energy expenditure, which also includes

basal and resting metabolism, the thermic effect of food, and growth (Malina and

66

Bouchard, 1991). Physical activity is a culturally bound concept and varies with ethnic,
cultural and SES conditions within the same society (Torun et al., 1996). There are
marked differences in the amount of time that children spend in structured and
spontaneous physical activities (Simons-Morton et al., 1990; Spurr and Reina, 1990). In
both developed and developing countries, there are SES and urban-rural differences in
activity participation.

SES differences in habitual physical activity are affected by interpersonal,
physical and social environments, public policy and the interaction of these variables
(McLeroy et al., 1988; Sallis et al., 1997; King et al., 1995). Environments rich in
resources such as sidewalks, parks, exercise facilities and health clubs make it easier for
people to be physically active (Sallis et al., 1997). Children in environments that have
high rates of crime, poverty and crowding engage in higher levels of sedentary activity
(Sallis et al., 1993; Goodway, 1994; Branta and Goodway, 1996). Goodway (1994)
found that Flint inner city children in an environment with high crime, unemployment,
small housing conditions, and poverty engaged in predominantly sedentary indoor
activities and had little or no exposure to physical activity outside of school. Poor
communities that have few resources are less able to provide recreation and activity
resources than more afﬂuent ones. Branta and Goodway (1996) concluded that streets
and the playgrounds of poor at-risk communities are often not safe places in which to
exercise or play.

The availability of open space also serves as a catalyst for habitual physical

activity of higher intensities. Approximately 2% of the variation in mean daily total

67

estimated energy expenditure (kcal/day) of 12 to 14 year old Taiwanese school boys was
attributed to SES (Huang, 1994). On the other hand, the locality of the school - urban or
rural - accounted for 6% of the difference in energy expenditure, with rural boys being
more active than urban boys (Huang, 1994). Approximately 2% of the variation in
estimated energy expenditure in MVPA (kcal/day) was explained by school differences.
Additionally, urban-rural locality of schools explained 6%, 4% and 4% of variation in
the mile run (time), sit-ups (per minute) and sit and reach (cm) tests, respectively.

In low SES urban and rural families, there is a heavy burden of participation in
domestic chores and waged labor at an early age (Berg, 1973). Children from upper SES
families, on the other hand, participate more often in structured extracurricular physical
activities. In South Africa, European boys spend considerably more time in organized
sport activities outside of school than African, “Colored”, or Indian boys (Sloan, 1966a;
Goslin and Burden, 1986).

‘Habitual physical activity has a signiﬁcant cultural component and reﬂects the
customs of the family and a society. In a study of 1,610 subjects from 375 Quebec
families, Pérusse et a1. (1989) estimated that genotype accounted for 29% of the
transmissible variance in the estimated level of habitual physical activity. It was
suggested that children could acquire the behaviors towards physical activity from his/her
parents. Cultural transmission accounted for 12% of the transmissible variation in
physical activity. Non-transmissible environmental factors, such as common
environment and lifestyle, accounted for 71% and 88% of the phenotypic variation of the

habitual physical activity and exercise participation, respectively.

68

Measuring and quantifying physical activity

Quantifying physical activity and exercise habits is difﬁcult. The validation of
methods of assessing physical activity is complicated by the fact that it is “impossible to
obtain a ‘gold standard’” (Blair, 1984; Baranowski et al., 1992). Physical activity is an
observable phenomenon. Thus, the validation of methods of measuring and quantifying
physical activity have traditionally been against observational data. Observation,
however, is subjective and the reliability of observation needs ﬁrrther research
(Baranowski et al., 1992).

The recall method is particularly easy and useful, but is limited to older children
and adults. The recall method is also conﬁned to assessing activity in the few days
immediately preceding the recall (Blair, 1984). The use of short term recall reduces the
amount of time available for memory decay, but there is no way to ensure that the
week/period is representative of the physical activity of the individual (W ashbum and
Montoye, 1986). A combination of methods for the evaluation of physical activity is
proposed by some since it provides the highest level of accuracy while maintaining
practicality (Saris et al., 1980; Saris, 1986).

The activity of children has been estimated in numerous ways, including parental
reports (Seliger et al., 1974), self-reports by children under the guidance of a teacher
(Shephard et al., 1980), and reports by the teacher on the child that were completed at the
end of each day (Saris et al., 1980), among others. One of the major problems of having
a parent or teacher report on the activity of a child is that the parent or teacher is only

aware of the activity that the child engages in while in the presence of the adult, or by

69

what the child tells the adult. This potentially fails to account for a signiﬁcant portion of
activity that occurs in the non-observed time. The greatest advantage of the self-report is
that it can be used easily with large numbers of subjects. However, the recall technique
is not useful for children under 10 years of age (Bouchard et al., 1983; Baranowski et al.,
1992)

Some of the problems with recall questionnaires include estimation of the
duration and intensity of activity, selective recall, discrimination of activity intensity at
high values, developmental appropriateness, and conversion of data (Baranowski et al.,
1984, 1992). The ability of children to recall and report the intensity and duration of
physical activity is limited, and is related to culture, education and age. Thus, young
children are less reliable in recalling their activity accurately, and daily ﬂuctuations in
activity levels make recall more difﬁcult (Baranowski et al., 1992). Further,
chronological time is a culturally bound concept and is oﬁen a factor of education. In
many developing countries children and adults do not own watches. Reporting the
duration of an activity presupposes ownerships and/or frequent access to a clock, as well
as a ﬁnely developed sense of chronological time.

Segrnenting the day into functional components and having the child record
aerobic activity within each component rather than the whole day increases the accuracy
of self reports (Baranowski et al., 1984). Children in 3rd to 6th grades have difﬁculty
making accurate minute-by-minute-time estimates, and in estimating the duration of an

activity regardless of the response format used. Providing options for time estimation

7O

other than a statement of precise minutes of expired time enhances the accuracy of
reporting.

Simons-Morton et a1. (1990) observed that children consistently reported greater
intensity and duration for moderate to vigorous physical activities (MPVA). There was a
86.3% agreement between the reported and observed number of MPVA that were 2 10
minutes. Similar overreporting of high levels of physical activity was noted by
Baranowski et al. (1992), and may be a result of the limited ability of the children to
recall, or the tendency to report the total amount of time that they were engaged in the
activity, as opposed to the amount of time that they were truly active.

All types of recall are intrusive on the subject, but this is exacerbated when the
subject is required to make a record of all activities. Recording of activities places a
great burden for compliance with the measurement protocol on the individual and could
result in noncompliance, or the effort required could cause him/her to change his/her
normal activities (Blair, 1984). By focusing on the typical pattems or activity, there is a
smaller burden on the subject to recall all miscellaneous activities. The activity pattern
questionnaire is a less intrusive instrument since the subject is required to report only
general or typical patterns of activity rather than to recall speciﬁc activities.

Physical activity reports are used to estimate the time and energy expenditure at
different levels of activity, using standard conversion charts (Bouchard et al., 1983;
FAO, 1985). However, the energy expenditure of children and adults are not the same.
Children have a higher energy expenditure and oxygen cost during weight-bearing

activities per unit body weight than adults, and the differences are greater at younger

71

ages (Astrand, 1952; Bar-Or, 1983). Consequently, the use of standard conversion charts
leads to an underestimation of energy expenditure in people under the age of 20 (Sallis et
al,1991)

Torun et al. (1983) estimated that the energy cost of basal metabolism was 38
kcalkg'lmin"l for children < 10 years compared to 18 kcalkg'lmin'l for adults. This
increased to 44 kcal.kg'l .min'1 for children when lying down, compared to 20 kcalkg'l.
min’1 for adults. Furthermore, energy costs did not remain constant and the child-to-
adult ratio of energy expenditure declined from 2.2:] for lying down to 1.121 for walking
up and down hill and climbing stairs (Torun et al., 1983). These estimates were based on
the energy expenditure of subjects who had previously been undernourished but were
currently well-nourished. Sallis et al. (1991), on the other hand, based estimated energy
expenditure of children on an analysis of several studies. However, no information on
the studies, nutritional status of subjects, or number of studies was cited. Sallis et a1.
(1991) estimated that there was a consistent 20% underestimation of energy expenditure
for 10 year old children, with an age-adjusted energy cost ratio of 1.18 of standard adult

values.

Habitual physical activity patterns of children

From middle childhood onwards, there is a greater development of proﬁciency in
movement skills and the ability to participate in structured activities (Haubenstriker and
Seefeldt, 1986). Running, walking fast, sports and games, and bicycling accounted for
over 70% of the moderate to vigorous physical activities for a sample of third and fourth

grade girls and boys from Texas (Simons-Morton et al., 1990). Similarly, the ﬁve top-

72

ranked physical activities of Taiwanese boys (12-14 years of age) were basketball,
bicycling, jogging, walking and running (Huang, 1994).

The intensity of participation in physical activity decreases with increasing age.
In an examination of the activity patterns of 6 to 16 year undernourished and normally
nourished Colombian children, no signiﬁcant age effects in the 12 hour average heart rate
frequency were observed, with children spending progressively more time at lower heart
rates and less at higher rates in their daily activity as they grew older (Spurr and Reina,
1990). Between 7 to 10 hours a day were spent on workloads < 30% YO: mu,
approximately 1.5 - 4 hours a day were spent on activities that were between 30% to 50%
170, m, and on average, between 20 to 60 minutes a day were spent at levels >50% 1702
m. The average 12 hour percentage 1702 w ranged from 20% to 30% of 170; m and
increased with age. Additionally, undernourished 6 to 8 year old boys worked at
signiﬁcantly lower values of percentage 1{702 m during school time compared to both
free play and nutritionally normal boys. Nutritionally normal boys, on the other hand,
had less vigorous involvement in free play compared to physical activity during school
time (Spurr and Reina, 1990).

Nutritionally normal (3rd and 4th grade) boys in Texas report 1.6 times more
moderate to vigorous physical activities before and after, than during, school (Simons-
Morton et al., 1990). The lower level of physical activity during school time of
undernourished children could be energy conservation in undernourished populations,
which is often alluded to in the literature (Spurr and Reina, 1990). Additionally, many

children from low SES families are required to be active participants in the home and

73

income generation for the family, which often requires higher energy expenditure in non-
school time.

Disease states alter nutritional Status and affect habitual physical activity. In a
study of the effect of treatment for helminth infections on the growth, nutrition and
physical activity of Kenyan children (7.5 years of age), Stephenson et a1. (1994)
observed a 43% increase in free-play activity after albendazole-treatment for helminth
infection. This increase was signiﬁcantly greater than that for a placebo group.
Structured activity levels increased for both groups in the second exam, with a 53%
increase in the treatment group compared to the 33% increase for the placebo group.

Goslin and Burden (1986) estimated the physical activity of South African
European, African and “Colored” secondary school students (Table 15). Self-reported
physical activity was assigned values of 8 (inactive), 9 (fairly active), 10 (active), and 11
(very active). European males had the highest physical activity levels, while African
females had the lowest activity levels. Statistical signiﬁcance was not indicated. The
authors concluded that the ethnic-related differences in physical activity were reﬂected in
performance-related ﬁtness tests, and argued that the differences in the performance tests
were most likely to be an artifact of the culture and experience of the groups.

Differences in activity participation at school and after school, weekend-weekday
differences, and the context of physical activity such as sport or waged labor were factors
that could bias the results and conclusions of this study. Also, habitual physical activity
was rated on a limited scale (8 to 11) that reduced the range of possibilities to just four

categories.

74

 

Table 15. Estimated activity level (SD) of South African European, “Colored” and
African adolescent males.‘I

 

 

European African “Colored”
Age 16.8 (:tO.8) 18.6 (i 1.4) 18.0 (i 1.5)
Activity level 10.0 ($0.8) 9.9 (i 0.5) 9.0 (i 0.5)

 

' Adapted from Goslin and Burden (1986).

The energy expenditure of Taiwanese school children (12 to 14 years of age), was
estimated using the Bouchard et a1. (1983) protocol (Huang, 1994). The age-speciﬁc
health-related physical ﬁtness of boys in the lowest and highest quartiles of mean daily
total estimated energy expenditure (kcal/day) were compared. Boys in the lowest
quartile of mean daily total estimated energy expenditure were taller and heavier, took
signiﬁcantly longer to complete the mile run and had less ﬂexibility of the back on the sit
and reach tests compared to boys in the highest quartile of mean daily total estimated
energy expenditure. Boys with high mean daily total estimated energy expenditure

performed better on health-related ﬁtness tests.

Summary

The stratiﬁcation of apartheid capitalism in South Africa has resulted in SES,
growth, and nutritional status differences that are largely differentiated on the basis of
race. SES and ethnicity are inextricably bound. In developing countries, indigenous
populations are of lower SES compared to immigrant populations. In South Africa
Africans have the lowest I-IDI and Europeans the highest HDI. These differences in the

standard of living are manifest in nutritional status and growth differences. In general,

75

Europeans are taller and heavier than African, “Colored” and Indian subjects of the same
age.

Physical ﬁtness is a combination of environmental and genetic factors that
include culture, maturity, SES, ethnicity, nutritional status, growth, health, and habitual
physical activity patterns of children. Nutritionally normal children have better physical
performance-related ﬁtness scores compared to undernourished and overweight boys.
The data suggest that South African European boys have superior performance- and
health-related ﬁtness scores compared to African and Indian boys. Africans, on the other
hand, have the lowest motor performances and physical ﬁtness scores. Similar ethnic,
SES and/or nutritional status differences in growth and performance have been found in
Bolivia, Brazil, Colombia, India, Mexico and Zaire.

Physical ﬁtness is related to the habitual activity levels of children, but the
correlations are low to moderate. There is a need to clarify the relationship, if any,
between habitual physical activity and ﬁtness is different populations. There is evidence
to indicate that habitual physical activity affects working capacity, the quality of life and
productivity of an individual. The causal relationship between low 902m and habitual
physical activity has not been addressed.

Habitual physical activity is affected by environmental factors that are closely
tied with SES. Lower SES children have fewer opportunities to participate in habitual
physical activity. On the other hand, despite the availability of space, undernourished
children tend to engage in lower intensity habitual physical activities during the school

day, whereas nutritionally normal children engage in higher levels of physical activity at

76

school compared to free time. Thus, the level of habitual physical activity is determined
by factors other than nutritional status or SES alone.

While the effects of SES, race and nutritional status on growth status have been
addressed, there is a lacuna in the knowledge on the interaction effects of SES and race
on nutritional status, and speciﬁcally what race and SES mean in the nutritional status of
different populations in South Africa. There is a need to identify and address the effects
of each of these factors and the potential confounding inﬂuences of these variables on the

growth and physical performance of children.

77

Chapter 3

NIETHODS

The purpose of the study was to investigate the effects of socioeconomic status

(SES), race, and nutritional status on the growth status, physical ﬁtness and habitual
physical activity of 10 year old South African boys of European, African and Indian
ancestry. Data on sociodemographic characteristics, growth, estimated body
composition, performance- and health-related ﬁtness, and habitual physical activity were
collected between 10 March and 11 June 1997. Permission to conduct the study and
recruit subjects was solicited from approximately 30 schools; 25 schools indicated a
willingness to participate in the study. The study was conducted at 18 of the 25 schools,

due to the school calendar and time constraints.

South African gghmls
The end of the apartheid era has led to drastic rearrangement within South Africa,
but large remnants of the apartheid era are still visible. The shortage of schools in
traditional African townships and in peri-urban informal settlements resulted in large
scale commuting to schools. A large proportion of government school populations are

bussed in from the traditionally disadvantaged areas. They account for 20% to 80% of

78

the school populations. Schools are divided into three major categories: Government,
Model-C and Private.

Schools that are predominantly subsidized by the state, “Government schools,”
are those in the former Black, Indian and “Colored” residential areas, and include some
of the larger and poOrer schools in former White residential areas. School fees at the
Government schools range from R150 to R600 per year.

Model-C schools are semi-private. School fees are higher than government
schools and range from Rl,200 to R3,600 per year. These fees are used to employ more
teachers and reduce the pupil-teacher ratio below 45:1 found in Government schools, and
for the improvement of facilities (both movable and ﬁxed). Model-C schools are the
schools of choice for middle and upper middle class parents. Public education
scholarships funded by the national lottery are given to some students to attend some of
the Model-C schools. Overall, Model-C schools have more affluent students than the
Government schools.

Private schools receive no government funding and generally have greater
ﬂexibility and extensivity of the curriculum. School fees at the Private schools range
from R6,000 to R24,000 a year. A small proportion of the students are on academic

scholarships.

same
There was great reticence among upper SES parents for their children to

participate in the study. Many Model-C schools in the predominantly White upper

79

middle class suburbs did not grant permission for the study. A similar reluctance to
participate was expressed at most Private schools. There was a high refusal rate among
predominantly White Model-C and private schools. Further, at schools which granted
permission, parental consent was low. The main reason for the refusal was that the
parent-teacher associations did not want the academic programs to be disrupted. It was
thus, very difﬁcult to recruit subjects of European (White) ancestry. This is an inherent
bias in the present sample since the majority of the higher SES and European boys were
excluded from this study.

Subject selection was based on the area in which the school was located, since
schools drew the majority of their constituencies from the areas in which they were
located. However, many students were bussed to Government schools from other areas,
but the actual percentage of students bussed was not ascertained. Boys attending Public
schools in traditionally low SES regions were classiﬁed as lower SES, while boys
attending schools in higher SES residential areas were categorized as higher SES.
Additionally, teachers were asked to assist in classifying the boys into “afﬂuent” and
“indigent” categories because school admission was not restricted to children from a
speciﬁc residential area.

At the time of testing, all subjects were between 10.0 and 10.99 years of age.
Permission to conduct the study was obtained from the principals of schools. In addition,
all subjects read and understood the human subject forms, which was approved by the
University Committee on Research Involving Human Subjects (UCRIHS) at Michigan

State University, and had signed parental consent and subject assent forms on ﬁle

80

(Appendices A, B and C). The overall parental and student response was less than 50%,
but ranged from 7.8% to 100% at the different schools.

Three hundred and eighty six boys between 10.00 and 10.99 years of age,
attending schools in the greater Johannesburg area (Gauteng), South Africa, were
subjects. The traditional classiﬁcation of race in South Africa was used, i.e., an
anthroscopic appraisal as it is offensive and politically incorrect to ask pe0ple to classify
themselves by race. One of the major social changes that occurred in the post-apartheid
South Africa was the removal of classiﬁcation by race. Homeroom teachers helped to
classify children and all children of mixed race were excluded from the study.
Approximately 40% of boys in the schools samﬂed were excluded from the study by
virtue of their race, i.e., of mixed ancestry and immigrant origin.

Sociod m a hic uestionnai

Parents (or surrogate parents) of each subject completed the sociodemographic
questionnaire. This questionnaire established date of birth, age, race, area of residence,
household size, SES, and a gross health history (Appendix D). In two “parent” families,
the SES was based on midparent education, occupational prestige rating and income. If
the child came from a single parent family, SES was based on the level of education,
occupational prestige and income of the head of the household.

Occupation was categorized using the occupational prestige ratings for South
Africans (Schlemmer and Stopforth, 1979). Since occupational prestige is not uniform
across population groups in South Africa, Schlemmer and Stopforth (1979) provided

adjusted occupational prestige ratings for Africans. Occupational prestige ratings ranged

81

from 15 for unskilled manual labor to 84 for independently employed professionals.
Income of the head of the household was divided into six categories (Table 16).
Occupancy ratio (the number of people per room) was calculated as the number of rooms

in the house, divided by the number of people who lived in the house.

Table 16. Income of the head of household per month and per year in South
African Rands and U.S. Dollars.‘l

 

 

 

Income

Category Recorded as per month per year
1 R208 R417 ($87) less than R5,000 ($1,042)
2 R635 R417 to R833 R5,000 to R10,000
($87 to $174) ($1,042 to $2,083)
3 Rl,052 R834 to R1,667 R10, 001 to R20, 000
($174 to $347) ($2,084 to $4,167)
4 112,001 R 1,668 to 113,334 R20, 001 to R40, 000
(S348 to $695) ($4,167 to $8,333)
5 R5,001 R3,335 or R6,667 R40,001 to R80,000
($695 to $1.388) ($8,334 to $16,667)
6 R6,668+ R 6,668 ($1.389) more than R80,000 ($16,667)

 

' Conversions are based on the exchange rate of four rand and ﬁfty cents (R480) to one U.S. dollar ($1)
in March, 1997.

Education was classiﬁed according the number of years of formal education, and
was assigned the following scores:

0 no formal schooling
some elementary education
completed elementary school
10 some high school
12 completed high school
13 technical education or some college/university
15 couegc/tmiversity degree
17 graduate and post-graduate degree

82

Since education, occupational prestige and income were rated on different scales,
all variables were converted into a common 5 point scale. Scaling values were calculated
as follows: education weight = 17/5; occupation weight = 84/5; income weight = 8/5.
Raw scores were then converted into scaled values as follows: education (score * 0.294);
occupation (score * 0.006) and income (score * 0.625). The SES score was calculated as
follows:

SES = (education score + occupational prestige + income)/3.
A score of 3.00 was used as the SES cutoff (higher > 3.00, lower SES 5 3.00). In the
absence of SES data, 92 boys were classiﬁed as higher (n=56) or lower (n=36) SES
based on the location of the school and teacher ratings of SES.

Nutritional status was classiﬁed on the basis of BMI-for-age using the U.S.
reference data of Must et al. (1991). Boys with a BMI-for-age s 15 percentile were
classiﬁed as undernourished. Boys with a BMI-for-age 2 85th percentile were classiﬁed
as overweight. Boys with a BMI-for-age >15 and <85th percentile were classiﬁed as
nutritionally normal. The breakdown of the sample by race and SES is summarized in

Table 17.

Table 17. Total sample by race and SES.

 

 

Sample size

African

Lower SES 101

Higher SES 107
Indian

Lower SES 38

Higher SES 99
E

Higher SES 41
Total sample 386

 

83

Procedures

A battery of items which included anthropometry, tests of physical ﬁtness, and
physical activity questionnaires was used. Sociodemographic information was obtained

from parents/guardians.

Anthropometry

Anthropometry is routinely used to assess the growth and nutritional status of a
population (WHO, 1976), and to document within and between population variation.
Anthropometric dimensions included indicators of attained size, skeletal robustness,
relative muscularity and subcutaneous fat. All anthropometric dimensions were taken
using the methods described in Lohman et al. (1988). Bilateral measurements were taken
on the right side of the body. Skinfolds were taken with a Holtain (Harpenden) caliper to
the nearest 0.1 millimeter. Height, sitting height, and biiliac and biacromial breadths
were measured to the nearest millimeter with a Siber-Hegner (Martin) anthropometer.
Elbow and knee breadths were taken to the nearest 0.5 millimeter with a small sliding
caliper. Mass was measured to the nearest 100 grams on a calibrated Healthometer
Professional scale. All circumferences were taken to the nearest millimeter using a
retractable, nonstretchable ﬂexible tape.

Measurements.

Body mass (body weight) was measured on a leveled platform scale with the

subject dressed in light clothes and in bare feet.

84

Stature, or standing height, was measured with an anthropometer. The subject
was barefooted and body weight was evenly distributed on both feet with heels together.
The shoulders were relaxed with the arms hanging freely at the sides. The head was
positioned in the Frankfort horizontal plane. A gentle upward stretch was applied to the
neck area. The measurement was taken as the distance from the vertex of the skull to the
floor.

Sitting height was a measure of the distance from the sitting surface to the vertex
of the skull. Sitting height was obtained by having the subject seated erect on a table
with the legs hanging freely over the edge of the table. The head was in the Frankfort
horizontal position. A gentle upward stretch was applied to the neck area. The vertical
distance between the vertex of the skull and the table surface was recorded.

Relaxed arm circumference was taken with the subject standing erect with the
hands at the sides and the palms facing the thighs. The midpoint between the olecranon
and acromial processes was the level of measurement. It was marked on the arm. The
tape was placed around the midpoint of the arm so that it touched the skin, but did not
compress the underlying soft tissues, and was positioned perpendicular to the long axis of
the arm over the marked points.

Calf circumference was taken with the subject standing and body weight evenly
distributed on both feet. A tape measure was positioned horizontally around the calf and
moved up and down to locate the level of maximum circumference in a plane
perpendicular to the long axis of the calf. This level was marked. The tape was in

contact with the whole circumference but did not indent the skin.

85

Biacromial breadth was taken from the rear of the subject with the upper end of
the anthropometer. The subject stood with the heels together and the arms hanging at the
sides. The shoulders were relaxed, downwards and slightly forward so that the maximal
reading could be taken. The most lateral borders of the acromial processes were palpated
and the measurement was taken across the most lateral borders of the acromial processes.

Bicristal breadth was also taken from the rear. The subject stood with his feet 5
cm apart to prevent swaying, with the arms away from the area measured. The iliac
crests were palpated and the measurement was taken with the upper end of the
anthropometer applied downwards at a 45° angle, with ﬁrm pressure, across the
maximum breadth of the iliac crests.

Biepicondylar breadth was taken with a small sliding caliper as the distance
between the epicondyles of the right humerus to the nearest 0.5 mm.

Bicondylar breadth was taken with a small sliding caliper with the subject sitting
and the right knee ﬂexed at 90°. The medial and lateral epicondyles were palpated and
the distance between the medial and lateral condyles of the femur was measured. Firm
pressure was applied as the distance between the condyles was measured to the nearest
0.5 mm.

Triceps skinfold was taken as a vertical fold at the same level as relaxed arm
circumference, midway between the acromial and olecranon processes in the midline on
the posterior aspect of the arm over the triceps muscle.

Biceps skinfold was taken as a vertical fold on the anterior aspect of the arm over

the biceps muscle at the same level as the triceps skinfold.

86

Subscapular skinfold was taken with the subject standing erect with his hands
relaxed at the sides. The inferior angle of the scapula was palpated. A diagonal skinfold
was picked up at approximately 45° to the horizontal plane following the natural
cleavage lines of the skin below the inferior angle of the right scapula. The measurement
was taken 1 cm inferior-laterally to the thumb and foreﬁnger.

Suprailiac skinfold was measured in the midaxillary line immediately superior to
the iliac crest. The subject stood with feet together in an erect position. An oblique
skinfold was grasped just posterior to the midaxillary line following the natural cleavage
lines of the skin.

Abdominal skinfold was taken as a horizontal fold 3 cm lateral to the midpoint of
the umbilicus and 1 cm inferior to it.

Derived dimensions.

1. Skinfolds were used to estimate total subcutaneous fatness and relative

subcutaneous fat distribution (Malina and Bouchard, 1988).

o Extremity skinfolds: sum of the triceps and biceps skinfolds.

o Trunk skinfolds: sum of the abdominal and subscapular skinfolds.

0 Total skinfolds: sum of ﬁve skinfolds.

o T rank/Extremity ratio:

2 of 2 trunk skinfolds/ 23 of 2 extremity skinfolds (mm/mm).
2. Estimated leg (subischial) length:
stature minus sitting height.

3. Arm musculature (Malina, 1995b):

87

arm muscle circumference (cm) = C, - 7t/2 (S[ + Sb)

arm muscle area (cmz) = 1/(4 1:) [Ca - 1r/2 (St - Sb)]2

where: C21 is arm circumference (cm)

St is triceps skinfold (cm)

S, is biceps skinfold (cm).
4. Sitting height/stature ratio, an estimate of relative trunk or leg length:

sitting height/standing height x 100.
5. Bicristal/biacromial breadth ratio, an indicator of the conﬁguration of the trunk in
terms of relative shoulder and hip proportions:
bicristal/biacromial x 100.
6. Body mass index (BMI):
weight (kg) / height (m)2.

Tghnical errors of measurement_.

All measurements were taken by the researcher. Intra-observer measurement
variability was calculated on replicate measurements of 10% of the sample, taken about
four days after the ﬁrst set. The technical error of measurement was calculated (Malina
et al., 1973):

oe = \/ 2d2/2n.
It is deﬁned as the square root of the sum of the squared differences of replicates divided

by twice the number of pairs. The intra-observor variability of the anthropometry was

within ranges observed in studies on other populations (Table 18).

88

Table 18. Intra-ohserver technical errors of measurement in the present study

(n=42) compared to those for several other studies.

 

 

Present study Katzmarzyk Krilta Wellens 1188' zavaleta and Rocha
(1997) (1995) (1989) Malina (1981) Ferreira
(198L

Mass (kg) 0.14 0.72 0.23 0.53 0.48 0.23
Statute (cm) 0.41 0.54 0.21 0.19 0.49 0.54 0.35
Sitting height (cm) 0.31 0.95 0.22 0.20 0.53 0.69 0.38
Circumferences (cm):

Relaxed arm 0.26 0.51 0.24 0.13 0.35 0.37 0.29

ﬂexed am 0.21 0.40 0.20 0.17 0.38

Carr 0.24 0.34 0.25 0.13 0.87 0.30 0.45
Breadths (cm): .

Biacromial 0.34 0.57 0.18 0.17 0.54 0.72 0.34

Bicristal 0.34 0.58 0.26 0.15 0.71 0.24 0.38

Biepicondylar 0.07 0.10 0.12 0.04 0.12 0.07 0.10

Bioondylar 0.07 0.84 0.16 0.04 0.11 0.06 0.10
Skinfolds (mm):

Triceps 0.46 0.94 0.23 0.15 0.80 0.51 0.55

Biceps 0.32 0.96 0.34 0.58 0.19

Subscapular 0.29 0.03 0.34 0.17 1.83 0.55 0.26

Suprailiac 0.32 1.13 0.60 0.24 1.87 0.55 0.26

Abdominal 0.55 1.64 0.41 0.89 0.55

 

”U.S. Hearth Examination Survey, Johnston et a1. (1972), Marina eta1(1973)

Physical ﬁtngs,

Tests of health-related and performance-related ﬁtness were used. Physical ﬁtness

testing is routinely performed as part of the Physical Education curriculum in South

Africa and students were familiar with the battery of tests used in this study. Health-

related ﬁtness was assessed using the 9 minute run, sit-ups, push-ups, and the sit and

reach. Performance-related ﬁtness was assessed using the 50 meter dash, grip strength,

standing long jump, and shuttle run.

89

Research assistants conducted six of the performance- and health-related ﬁtness
tests: 9 minute run, 30 second sit-ups and push-ups, dash, long jump and shuttle run. The
sit and reach and grip strength tests were conducted by the author. The best of three
trials were recorded for grip strength, sit and reach, and the standing long jump. The
better of two trials were used for the dash. An all out single effort was used to record the
9 minute distance run, push-ups, sit-ups, and the shuttle run.

Measurement of Physical Fitness

9-minute run-walk provided an indicator of cardiorespiratory ﬁtness. Subjects
ran-walked the greatest distance that they could in 9 minutes. Most tests were conducted
on makeshift asphalt or grass surfaced 200 to 400 meter tracks, but some tests were
conducted on makeshift 100 meter tracks. The subject assumed a standing start at the
starting line on the command ‘Get Ready’. On the command ‘Go’, the subject jogged
and/or walked around the track continuously for nine minutes. After each lap the subject
was informed of how many minutes of the run were left. The countdown provided both
incentive for continued participation as well as cues for pacing the run. When the whistle
was blown to signal the end of the nine minutes, the subject stopped where he was and
sat down on the ground. The run was scored in terms of distance completed to the
nearest 10 meters.

5 0-meter dash was a measure of running speed. The test was conducted on a
track where possible. However, in some instances, the sidewalk or grass surfaces were
used. A short warm-up period preceded the dash. The subject assumed a standing

starting position at the starting line on the command ‘Get Ready’. On the command

9O

‘Go’, the subject ran as fast as possible to a point beyond the 50 meter mark. This
overrunning technique was used to maintain maximum acceleration throughout the dash.
The time elapsed to the nearest 0.01 second between the starting signal and crossing of
the ﬁnish line was record as the score.

Grip strength was a measure of static strength. The subject held an adjustable
Stoelting dynamometer comfortably in the right hand in line with the forearm. It hung
down beside the thigh. The second joint of the hand was ﬁtted snugly under the handle
and took the weight of the instrument. The dynamometer was gripped between the
ﬁngers and the palm at the base of the thumb. The ﬁrmly held dynamometer was raised
clear of the body and no contact with the body was allowed. During the squeezing of the
dynamometer, the arm was not allowed to swing or perform a pumping motion. Each
trial was recorded to the nearest 0.5 kilogram.

Modiﬁed push-ups for males provided a measure of endurance of the arm and
shoulder girdle musculature. The subject lay face down on the ﬂoor. The body was
straight with the arms bent and hands ﬂat on the ﬂoor beneath the shoulders. The lower
legs were crossed with the knees bent. The subject pushed upward to a straight arm
position, lowered his body until his chest touched the ﬂoor, and repeated the exercise as
many times as possible in 30 seconds (Chrysler Fund/Amateur Athletic Union, 1993-
1994)

Shuttle run was a measure of speed and agility. Two parallel lines 120 cm long,
were drawn on the ﬂoor 5 meters apart. Two cones marked the end of each line. The

subject stood behind one of the lines, with one foot just behind the line. On the signal

91

‘Go’, the subject ran to the other cone, crossed that line with both feet, returned to the
starting line and crossed the starting line with both feet, as quickly as possible. The
subject completed ﬁve of these cycles as quickly as possible. To guard against
deceleration at the end of the ﬁfth cycle, the overrunning technique as in the dash was
used. The score was the time, in tenths of a second, taken to complete ﬁve cycles
(Beunen et al., 1988).

Sit-ups provided a measure of abdominal strength and muscular endurance. The
subject assumed a supine position on a mat with the knees ﬂexed and heels ﬂat on the
mat. The arms were crossed on the chest with the hands on opposite shoulders. The
subject's feet were held down by a partner. On the ‘Go’, signal, the subject curled up to
touch his knees with his elbows, and curled back to the ﬂoor until the midback contacted
the mat. The subject performed as many sit-ups as possible in 30 seconds. Only
completed sit-ups were counted.

The standing long jump was a test of explosive power. The subject stood behind
the take-off line with his feet at a natural distance apart. In the preparation phase of the
jump, the subject bent at the knees and simultaneously swung his arms backwards. The
subject jumped forward by extending at the knees and simultaneously swinging his arms
forward. The subject took off with both feet simultaneously, jumped forwards and
landed as far away from the take-off line as possible. Iftlre subject fell backwards, or
took off on one foot, another trial was given. The distance to the nearest 0.5 cm between
the point of contact closest to the take-off line and the take-off line was measured as the

distance of the jump.

92

Sit and reach evaluated the ﬂexibility of the lower back and posterior thighs. The
subject sat down at the test apparatus with knees ﬁrlly extended and the bare feet
shoulder width apart. Feet were ﬂat against the end board. Arms were extended forward
with the hands placed on top of each other. The subject reached directly forward, palms
down, along the measuring tape four times and held the position of the maximum reach
of the fourth trial. The position of the maximum reach was held for one second. The
most distant point reached on the fourth trial was measured to the nearest 0.5 centimeter

Within-day and between day reliabilities were calculated on 10% of the sample
(Tables 19 and 20). The ﬁrst and third measurements were used to calculate the within
day reliability. Between day reliability was calculated on measurements taken four days

apart.

Table 19. Within day correlations of ﬁtness tests of trials one and three for 42 boys
with comparative data for three other samples.

 

 

 

 

 

 

Present Study Malina and Buschang Malina and Rocha Ferriera
(1985) Moriyarm (1991) (1988)
South African Mexican (Oaxaca) Philadelphia childremBrazilian children
boys children
IstandBrd lstand2nd bestandanbest 1stand2nd
trials
Dash 0.74 0.71 to 0.89 0.39 to 0.98 0.84
Grip 0.96 0.63 to 0.97 0.79 to 0.98 0.99
Sit and reach 0.92
W 0.97 0.82 to 0.94 0.72 to 0.95 0.99

 

Diaries and Questionnaires
A 3-day physical activity self-report and a habitual physical activity questionnaire

were completed by subjects and a sociodemographic questionnaire was completed by

93

Table 20. Between day correlations of fitness tests taken four days apart with
comparative data for Brazilian children.

 

 

 

Correlations
Present Study I Rocha Ferreira (1988)
Dash 0.53 0.74
Grip 0.95 0.95
Nine minute run 0.92 0.68
Push-ups 0.98
Shuttle run 0.79
Sit and reach 0.95
Sit-ups 0.95
Standing long jump 0.96 0.69

 

parents or guardians. Subjects completed the questionnaires and the diaries under the
supervision of the researcher or research assistants.

Phygigl activity guestionngig-a

Physical activity data were collected using the Bouchard et al. (1983) protocol
(Appendix E), which was adapted by dividing the day into meaningful segments as
opposed to 15 minute intervals. The day was divided into meaningful segments (before
school, before ﬁrst recess, ﬁrst recess, before second recess, second recess, after second
recess, after school, afternoon, before supper, after supper) to increase the accuracy of
reporting (Baranowski et al., 1984). The research assistant who administrated the
questionnaire prompted subjects to provide estimates of the duration of activities to the
nearest 15 minutes.

The physical activity questionnaire was a retrospective 24-hour recall. Activities
of the preceding 24 hours were recalled and recorded. All efforts were made to have the

subject recall the activity for one weekend day and two weekdays. The questionnaire

94

was administered on three successive school days -- Monday, Tuesday and Wednesday;
or Thursday, Friday and Monday. Activities of the preceding day (Sunday, Monday and
Tuesday; or Wednesday, Thursday and Sunday) were recalled. This ensured that a
weekend day (Sunday) was included in the survey. Activity questionnaires were
administered at the same time each day to ensure continuous coverage of the three day
period.

The Bouchard et al. (1983) protocol was used in an interview format with
individual children. A list of activities was given to each subject along with the 24-hour
record. The subject was prompted and signiﬁcant events such as before and after
breakfast were used to trigger the recall of the events of the previous day. The researcher
asked the subjects the following types of questions to prompt and cue recall:

0 What time did you get up and go to bed?

0 Did you participate in any domestic chores related to the activity (cleaning,
preparing meals, tending to garden, cleaning the car, etc)?

o What did you do immediately before or after a meaningful event (getting up,
breakfast, school, ﬁrst recess, second recess, aﬁer school, before snack, after
snack, before supper, aﬁer supper, before bed, bedtime) in the day?

0 How did you move (run, walk, hop, skip, fast, slow, moderate) about from one
location to another?

0 How much of effort (sweating, hyperventilating, tired) was involved in the

activity?

95

o In what types of activities (games, watching television, reading, solitary hobbies,

etc.) did you engage?

0 Were you engaged in extracurricular activities (choir, ballet, drama, martial arts,

sport clubs, etc)?

o What television program was on? Or what television programs did you watch?
The amount of time and the estimated energy expenditure in sedentary (1-2 METS), light
(3-5 METS) and moderate-to-vigorous physical activity (MVPA, 6-9 METS) were
estimated for each day, based on the criteria (Table 21) of Bouchard et al. (1983). The
mean amount of time spent in light and MVPA, as well as total daily estimated energy
expenditure and energy expenditure in MVPA for each 15 minute segment, were
calculated.

The physical activity of 57 subjects was observed during recess and lunch breaks
on the second day at the school. Activity observations were recorded as accurately as
possible and the time spent on each activity was recorded to the nearest ﬁve minutes.
The energy expenditure (METS) was calculated. The reliability of the recall records
were compared with the observed physical activity records. There was a high positive
correlation (r = 0.89) between observed and recalled physical activity in this subsarnple.
The correlation was similar to the correlation (r = 0.84) between observed and recalled
physical activity reported by Baranowski et al. (1984). Although there was reasonable
consensus between recalled and observed physical activity, there was a signiﬁcant
difference in the accuracy of the recalled values (Table 22). The boys reported higher

levels of habitual physical activity compared to observed values.

96

Table 21. Activity Code Sheet (after Bouchard et al., 1983).

 

METS

Example of activities in this code

 

Sleeping, resting in bed

Mg: eating, listening, writing, reading, studying, watching TV

Light activig standing: washing, combing my hair, cooking, and similar
activities

Slow walking (less than 4 kilometers per hour), driving a motorcycle, dressing,
showering, and similar activities

Light manual work: household chores such as sweeping the ﬂoor, washing
windows, vacuuming, painting, waiting on tables, walking at 4 to 6 km/hr --
km in 10 to 15 minutes

Leisurg hetivitig gd smrts in h rm’gngj environment: baseball/softball,
cricket, ﬁeld hockey, golf, volleyball, canoeing or rowing, cycling (less than

10 km/hr), table tennis, and similar leisure/recreational activities, drama classes
Mal wgrk at a mode_rgte pace: carpentry, raking and mowing the lawn
loading and unloading goods, chopping wood, and similar activities

Lgishre and smrt gctivitig hf higher intﬂity (no; mmpgh'h'vg): canoeing (5 to
8 km/hr), bicycling (more than 15 km/hr), dancing, water skiing, badminton,
gymnastics, swimming, tennis, horse riding, ﬁtness exercises like calisthenics,
walking (more than 6 kin/hr), karate, judo, etc.

 

judo, jogging and running (more than 9 krn/hr), racquetball, handball, squash,
badminton, tennis, swimming, hiking, rugby, soccer, carrying heavy loads,

aerobic dance, and similar intense activities

 

97

Table 22. Means, SD, correlations and AN OVA for observed and recalled
estimated energy expenditure (kcal/day) in physical activity of 57 boys.

 

 

 

 

 

 

Recess Lunch M
M SD M SD M SD
Observed 12.77 4.35 16.04 6.29 28.81 9.19
Reported 18.47 8.46 18.32 8.40 36.79 14.69
Correlation Sum of df Mean F
squares square

Recess 0.86" Between Groups 787.20 5 157.44 29.65“
Within Groups 270.83 51 5.31
Total 1058.04 56

Lunch 0.86" Between Groups 1687.52 6 281.25 26.51"
Within Groups 530.42 50 10.61
Total 2217.93 56

Total 0.89" Between Groups 3914.69 11 355.88 19.77"
Within Groups 810.19 45 18.00
Total 4724.88 56

**p<0.01

thitual Phygighl Am‘vity
The Physical Activity Questionnaire for Older Children (PAQ-C) was used to

collect habitual physical activity data (Appendix F) (Kowalski et al., 1997). The PAQ-C
is a self-administered 7-day recall questionnaire. Nine items assessing the habitual
physical activity of children were scored on a 5-point scale and were used to derive the
total activity score. Higher scores indicated higher levels of activity. Six questions
assessed activity in physical education classes, recess, lunch, immediately after school, in
the evening, and on weekends. One question asked the subject to categorize his activity
for the last seven-days based on ﬁve descriptions ranging from low to very high activity
levels. Another question required the subject to list how many times he had engaged in

physical activities for each day of the preceding week. The last question was an activity

98

checklist consisting of common sports, leisure activities and games, and had room for the
subject to list any other activities that he commonly participated in that were not included
on the checklist. The primary purpose of this item was to act as a memory cue. The
PAQ-C score which ranges from 1 to 5, was calculated:

PAQ-C score = [(2 of scores for questions 1 through 8) + (total score for question

nine/ number of items checked on activity checklist)] / 9.

Statistical Analyses

Descriptive statistics were calculated for all variables; SES-speciﬁc, race-speciﬁc
and nutritional status-speciﬁc descriptive statistics were also calculated. Multivariate
analysis of variance (MANOVA) was used to answer the question: “What is the
relationship of SES, race and nutritional status to the growth status, physical ﬁtness and
habitual physical activity of 10 year old South African boys?”.

SES, race and nutritional status were entered as independent variables with
growth (stature and mass), proportions (sitting height/stature ratio; bicristal/biacromial
ratio), body composition (2 of 5 skinfolds; trunk/extremity ratio; estimated arm
musculature), individual ﬁtness tests, estimated energy expenditure and habitual physical
activity as the dependent variables.

To speciﬁcally answer research question one: Are there any SES, race and
nutritional status differences in the growth status and physique of 10 year old South
African boys?” MANOVA and MANCOVA were performed. Anthropometric

variables were categorized as size attained, body proportions, and body composition.

99

Analysis of covariance (ANCOVA) was used to determine SES-speciﬁc, race-speciﬁc
and nutritional status-speciﬁc differences in grth status. SES, race and nutritional
status were the covariates. MANOVA was used to calculate the interaction of SES, race
and nutritional status on growth status. Speciﬁc between group differences for all
signiﬁcant MANOV‘As and MANCOVAs were tested using the Scheffe’ Post hoc test and
pairwise comparisons of estimated marginal means when SES, race and/or nutritional
status were used as covariates.

To speciﬁcally answer research question two: “Are there race, SES and
nutritional status differences in the health-related and performance-related ﬁtness of 10
year old South African boys?”, MANOVA and ANOVA were conducted. Components
of health-related ﬁtness included sit-ups, push-ups, 9 minute run, sit and reach and the
sum of 5 skinfolds, while components of performance-related ﬁtness included the dash,
shuttle run, grip strength and the standing long jump. The effects of race, SES and
nutritional status were considered individually, while the other two independent variables
were entered as covariates. Between group differences for all signiﬁcant ANOVAs and
ANCOVAs were identiﬁed using the Scheffé Post Hoc test and pairwise comparisons of
estimated marginal means, respectively. Multiple and stepwise regression analyses were
performed to determine which sociodemographic factor or factors (SES, race and/or
nutritional status ) was and/or were signiﬁcant predictor/s of health- and performance-
related ﬁtness items.

To speciﬁcally answer research question three: “Are there SES, race and

nutritional status differences in the habitual physical activity of .10 year old South

100

African boys?” MANOVA was performed. Estimated energy expenditure was sorted
into sedentary (1 - 2 METS), light (3 to 5 METS), MVPA (6 - 9 METS) and total daily
energy expenditure. Multiple and stepwise regression analyses were conducted to
determine which sociodemographic factor/s (SES, race and nutritional status) was and/or
were the most powerful predictor/s of sedentary, light and moderate to vigorous physical
activity.

To speciﬁcally answer research question four: “Are there nutritional status
differences in race-speciﬁc habitual physical activity of 10 year old South African boys?”
MANOVA was performed. The effects of SES were controlled statistically. Estimated
energy expenditure was sorted into sedentary (1 - 2 METS), light (3 to 5 METS), MVPA
(6 - 9 METS) and total daily energy expenditure. Stepwise regression analyses were
conducted to determine if nutritional status was a powerful predictor of sedentary, light
and moderate to vigorous physical activity.

To speciﬁcally answer research question ﬁve: “Are there race-speciﬁc differences
in the growth status and physical ﬁtness of 10 year old South African boys of African,
Indian and European ancestry with different levels of energy expenditure?”, race-speciﬁc
differences in the growth and physical ﬁtness of boys in the lowest (< 25th percentile )
and highest (> 75th percentile) quartiles of MVPA and total energy expenditure
(kcal/kg) were compared using ANCOVAs with the SES as the covariate. Pairwise
comparisons of estimated marginal means and Scheffe post hoc comparisons of

signiﬁcant MANCOVAs and MANOVAs were used.

101

Power of statistical analyses.

The sample size is large enough to provide adequate statistical power. Adequate
power (i.e., > 0.70) for multivariate analyses (MANOVA and MANCOVA) at an alpha
of 0.10, for two through seven variables at a moderate overall effect size of 0.64 requires
about 30 subjects per group (Stevens, 1980). The sample size that will provide adequate
power in regression analyses is roughly calculated as 15 subjects per variable (Stevens,
1992). Thomas et al. (1997) posit that setting power at 0.8 adjusts the level of Type II
errors to an acceptable one. Estimated sample size, based on the tables of Cohen (1988),
with an effect size of 0.5 and high power (0.8), would require a minimum of 26 subjects
per group.

The regression analyses that were calculated were to determine if ethnicity, SES,
or nutritional status were predictors of physical ﬁtness and estimated energy expenditure.
This analysis, using three variables, required a minimum of 45 subjects from each SES
group to provide adequate power. The sample sizes for higher (n = 101) and lower (n =
107) SES African, and higher (11 = 99) SES Indian boys were large enough to satisfy the
criteria for power stipulated by both Stephens (1992) and Cohen (1988). The sample
sizes of lower SES Indian (11 = 38) and European boys (11 = 41), on the other hand, were
smaller than recommended, indicating a lower power to predict the trends for European

and lower SES Indian boys

102

Chapter 4

RESULTS

The study examines the relationship of race, SES and nutritional status to the
growth status, physical ﬁtness and habitual physical activity of 10 year old South African
boys of African, Indian and European ancestry. The results are presented in four parts:
growth; physical ﬁtness; habitual physical activity; and relationships among habitual

physical activity, growth status and physical ﬁtness.

Descriptive Statistics

Means (M), standard deviations (SD), and medians (1nd) where appropriate, for
all variables for the total sample are shown in Table 23. Corresponding statistics by race
are shown in Table 24. Summary ANOVAs of differences among the three racial groups
and Scheffé post hoc tests are shown in Table 25.

Descriptive statistics by race (SES combined in African and Indian boys) are
summarized in Table 24. Between group differences for all signiﬁcant ANOVAs, which
required F values larger than 2.95 and 4.50, for p<0.05 and < 0.01, respectively, with two
degrees of freedom, were tested with the Scheffé post hoc test (Table 25). Height and
mass do not differ among the groups. European boys, however, have longer sitting

heights than Indian and African boys, who do not differ signiﬁcantly.

103

Table 23. Sample sizes, means, standard deviations, ranges and medians (where
appropriate) for all variables in the total sample.

 

 

Variables N Mean SD Median Min Max
Age, yrs 389 10.5 0.3 10.0 11.0
Mass, kg 389 32.7 6.9 31.5 15.8 61.6
Stature, cm 389 138.4 6.6 120.6 160.4
BMI, kg/m2 386 16.9 2.7 16.4 10.0 29.5
Leg length, cm 386 68.6 4.8 53.0 87.9
Sitting height, cm 386 69.7 3.6 59.0 81.0
Sitting height/Stature, % 386 50.4 1.9 43,.7 57.4
Skinfolds, mm
Biceps 386 6.9 4.0 5.8 2.3 25.0
Triceps 386 12.5 5.8 10.5 5.0 34.2
Subscapular 385 9.2 6.4 6.8 3.8 45.8
Suprailiac 386 7.7 5.6 5.6 2.2 36.4
Abdominal 386 12.1 8.1 8.9 3.4 42.0
2 of Trunk 385 21.3 14.0 15.9 7.2 87.8
r: of Extremity 386 19.4 9.4 16.2 8.0 59.2
2 of 5 385 48.4 27.9 37.6 19.5 168.2
Tank/Extremity, mm/mm 385 1.06 0.26 1.05 0.39 2.02
Circumferences, cm
Relaxed arm 387 20.3 2.7 19.8 15.1 29.6
Calf 386 27.6 2.8 27.2 20.6 36.5
Arm muscle area, cm2 386 30.5 8.5 28.5 16.6 66.3
Breadths, cm
Biacromial 387 28.8 2.3 20.1 35.1
Bicristal 387 19.9 2.1 14.5 27.0
Biepicondylar 385 5.5 0.4 4.0 6.6
Bicondylar 386 8.1 0.5 6.7 9.8
Bim'ghm mmial. % 386 69.1 52f 47.1 89.7

 

104

Table 23. (continued)

 

 

N Mean SD Min Max
Physical ﬁtness
Dash, sec 382 9.5 1.2 8.0 15.1
Grip, kg 383 19.0 3.5 10.0 30.0
Grip, kg/kg mass 383 0.6 0.1 0.3 1.0
Shuttle run, sec 383 23.0 2.6 19.0 35.0
Standing long jump, cm 379 151.1 21.8 76.0 197.0
9-minute run, m 388 1622.0 376.0 560.0 2840.0
Push-ups, #/30 sec 384 25.0 8.0 1.0 46.0
Sit and reach, cm 383 26.2 6.0 6.0 39.5
Sit-ups, #lsec 386 17.0 6.0 2.0 39.0
Time spent in sedentary activities, 3 day mean, hours
Television 337 2.7 1.6 0.0 10.0
Commute 350 0.6 0.6 0.0 2.9
Computer Games 337 2.9 1.7 0.0 10.0
Energy erqrenditure, 3-day mean, hours
Sedentary 372 20.1 1.3 15.0 22.9
Light 333 2.6 1.0 0.8 6.1
MVPA 333 1.4 0.8 0.0 4.8
Energy expenditure, 3-day mean (kcallday)
Sedentary 328 721.8 170.2 333.6 1386.4
Light 328 260.0 126.9 51.6 829.4
MVPA 328 274.0 174.6 0.0 953.2
Total 328 1255.8 297.3 732.8 2360.5
Energy expenditure, 3-day mean (kcal/kg/day)
Sedentary 328 22.0 2.0 16.7 28.2
Light 328 7.9 3.4 1.7 19.2
MVPA 328 8.6 5.4 0.0 29.6
Total 328 38.5 5.8 24.7 55.6
PAQ-C score 383 2.8 0.5 1.4 4.5

 

105

Table 24. Descriptive statistics by race.

 

 

African Indian European
N M SD Md N M SD Md N M SD Md
Age, yrs 209 10.5 0.3 10.5 139 10.5 0.3 10.5 41 10.5 0.4 10.5
Mass, kg 209 32.5 6.1 31.5 139 32.7 8.3 30.6 41 33.7 5.7 33.3
Stature, cm 209 137.9 6.4 139 139.0 6.9 41 139.2 6.4
BMI. lag/1112 209 17.1 2.6 16.6 139 16.6 2.9 15.9 41 17.4 2.6 17.4
Leg length. cm 209 68.4 4.7 139 69.3 5.0 41 67.5 4.2
Sitting height, cm 209 69.4 3.7 139 69.7 3.6 41 71.8 2.9
Sit lit/Stat, % 209 50.4 2.0 139 50.2 1.9 41 51.6 1.2
Skinfolds. mm
Biceps 209 6.6 3.7 5.2 139 7.4 4.3 5.8 41 6.6 3.6 5.4
Triceps 209 11 .9 5.6 10.0 139 13.1 6.0 11.2 41 13.2 5.9 11.2
Subscapular 209 8.7 5.5 6.8 139 10.3 7.7 7.3 41 8.0 5.1 6.3
Suprailiac 209 6.9 4.6 5.4 139 8.9 6.8 5.9 41 7.5 5.2 5.6
Abdominal 209 10.8 7.0 8.4 139 14.1 9.5 10.6 41 11.7 7.2 9.7
2 Trunk 209 19.5 12.1 15.4 139 24.3 16.7 17.6 41 19.7 11.8 15.9
E Extremity 209 18.5 9.0 15.2 139 20.5 10.0 17.3 41 19.9 9.3 16.7
2 5 209 45.0 24.7 36.8 139 53.8 32.2 41.0 41 47.0 25.3 38.8
TER. min/mm 209 1.03 0.23 1.04 139 1.13 0.28 1.13 41 0.98 0.23 0.95
Circumference, cm
Relaxed arm 209 20.2 2.5 19.7 139 20.3 3.1 19.6 41 21.0 2.5 21.0
Calf 209 27.7 2.6 27.3 139 27.3 3.1 26.6 41 28.1 2.7 28.3
Arm muscle area, cm2 209 29.6 8.0 28.0 139 31.7 9.5 29.2 41 30.9 7.0 30.7
Breadths, cm
Biacromial 209 28.6 2.4 139 28.6 2.3 41 29.8 1.3
Bicristal 209 19.6 2.1 139 19.8 2.1 41 21.4 1.3
Biepicondylar 209 5.5 0.4 139 5.5 0.4 41 5.6 0.4
Bicorrdylar 209 8.1 0.5 139 8.1 0.5 41 8.3 0.5
Bier/Biac, % 209 68.4 4.9 139. 69.3 5.7 41 72.0 3.6

 

106

Table 24. (continued)

 

 

 

Aﬁican Indian European
N M SD N M SD N M SD
Physical ﬁtness
Duh, sec 200 9.4 1.2 137 9.8 1.2 41 9.4 1.0
Grip, kg 205 19.3 3.6 136 18.6 3.4 41 19.3 3.0
Grip, kg/kg mass 205 0.6 0.1 136 0.6 0.1 41 0.6 0.1
Shuttle run, sec 205 23.0 2.5 137 23.2 2.8 41 22.5 2.2
Standing longjump, cm 205 150.5 21.5 137 149.8 21.9 41 157.7 21.9
9-minute run, in 205 1616.7 360.5 138 1556.2 336.9 41 1875.4 471.6
Push-ups, ill/30 sec. 205 23.7 8.2 135 25.2 8.3 41 28.0 8.0
Sit and reach, cm 205 27.8 5.7 137 24.0 5.9 41 25.7 5.4
Sit-ups, #l30 sec. 205 16.4 5.3 138 16.5 5.9 41 17.2 5.5
Time in sedentary activities, 3 day mean, hours
Television 179 2.2 1.3 121 2.5 1.7 41 1.4 0.8
Commute 179 1.4 0.9 121 0.2 0.4 41 0.4 0.7
Computer Game 179 0.1 0.3 121 0.2 0.6 41 0.6 1.2
Energy expenditure, 3-day mean, hours
Sedentary 179 19.8 1.3 121 20.5 1.1 41 19.7 1.3
Light 179 2.8 0.9 121 2.3 0.9 41 2.9 1.0
MVPA 179 1.6 0.8 121 1.2 0.7 41 1.5 0.8
Energy expenditure, 3-day mean (kcal/day)
Sedentary 179 719.0 155.8 121 732.7 194.6 41 692.6 144.8
Light 179 275.0 122.9 121 235.7 133.2 41 269.4 111.6
MVPA 179 305.7 175.4 121 219.2 152.2 41 307.8 204.9
Total 179 1299.7 273.0 121 1187.6 327.3 41 1269.8 263.7
Energy expenditure, 3-day mean (kcal/kg/day)
Sedentary 179 22.0 2.0 121 22.3 1.8 41 21.3 2.5
Light 179 8.4 3.3 121 7.2 3.3 41 8.4 3.3
MVPA 179 9.6 5.5 121 6.8 4.5 41 9.6 6.1
Total 179 40.0 5.5 121 36 3 5.1 41 39.2 6.6
PAQ-C score 205 2.8 0.4 137 2.8 0.6 41 2.8 0.5

 

107

Table 25. Results of ANOVA (F ratios) and Scheffé post hoc tests for mean
differences (MD) between races.

 

Scheffe post hoc race differences

 

 

 

Dependent variable F Race Race MD
Mass, kg 0.38 African Indian 0.66
African European 0.72
Indian European 0.06
Stature, cm 1.17 African Indian -1.10
African European -1 .46
Indian European -0.36
BMI, kg/rn2 2.56 African Indian 0.70
African European 0.70
Indian European 0.00
Leg length, cm 2.35 African Indian -1.14
African European 0.57
Indian European 1 .71
Sitting height, cm 4.06' African Indian 0.03
African European -2.00‘
Indian European -2.07‘
Sitting height! stature ratio, % 5.74" African Indian 0.22
African European -1.20”
Indian European 4.42"
Skinfolds, mm
Biceps 2.92 African Indian -0.60
African European 1.38
Indian European 1.98
Triceps l .93 African Indian -0.92
African European 1 .24
Indian European 2.16
Subscapular 3 .20‘ African Indian -1.15
African European 2.16
Indian European 3.30‘
Suprailiac 5.38" African Indian -1.81‘
African European 1.18
Indian European 2.99'
Abdominal 5.01 " African Indian -2.83‘
African European 0.82
Indian European 3.65
E of Trunk 5.28” African Indian 4.82”
African European -0.18
Indian European 4.65
E of Extremity 1.82 African Indian -1.94
African European -1.28
Indian European 0.66
2 of 5 4.24’ African Indian -8.81’
African European -2.06
Indian European 6.75
TER, mm/mm 8.95” African Indian -0.10”
African European 0.05
Indian European 0.15”
Circumferences, cm
Relaxed arm 0.11 African Indian 0.14
African European 0.12
Indian European -0.02
"' p<0.05 " p<0.01

108

Table 25. (continued)

 

Scheffe post hoc race differences

 

 

Dependent variable F Race Race MD
Calf 0.64 African Indian -1.32
African European 1.37
Indian European 2.68
Estimated muscle area, cm2
Arm 2.72 African Indian -2.33
African European -0.94
Indian European 1.39
Breadth, cm ,.
Biacromial 3.21“ African Indian 0.40
African European -0.73
Indian European -1 . 13‘
Bicristal 13.84" African Indian 0.25
African European 4.84"
Indian European -1 .60”
Biepicondylar 0.24 African Indian -0.01
African European -0.06
Indian European -0.05
Bicondylar 0.60 African Indian 0.01
African European -0.12
Indian European -0.12
Bicristal/biacromial ratio, % 6.01 ” African Indian -0.93
African European -3.69“
Indian European -2.76"
Physical Fitness
Dash, sec' 3.64‘ African Indian 036*
African European 0.05
Indian European -0.4l
Grip, kg 2.04 African Indian 0.83
African European -0.06
Indian European -0.89
Grip, kg/kg mass 0.46 African Indian 0.01
African European -0.01
Indian European -0.02
Shuttle run, sec' 0.24 African Indian 0.21
African European 0.22
Indian European 0.01
Standing long jump, cm 0.53 African Indian 0.10
African European -4.51
Indian European -4.61
9-minute run, m 7.29" African Indian 38.63
African European -23 1 .91 " *
Indian European -270.54“
Push-ups, #/30 sec 2.95" African Indian -1.46
African European -3.91
Indian European -2.46
Sit and reach, cm 16.50" African Indian 4.07"
African European 1.21
Indian European -2.86

 

" p<0.05

5-Signs for the dash and shuttle run are inverted.

109

" p<0.01

Table 25. (continued)

 

Scheffe’ post hoc race differences

 

 

 

Dependent variable F Race Race MD
Sit-ups, #/30 sec 0.11 African Indian 0.10
African European -0.46
Indian European -0.55
Time in sedentary activities, hours
Television 6.18" African Indian -0.38
African European 0.73
Indian European 1.12"
Commuting 135.90” African Indian 0.87"
African European 0.78"
Indian European -0.09
Computer Games 3.54‘ African Indian -0.51‘
African European 0.05
Indian European 0.56
Estimated energy expenditure 3-day mean, hours
Sedentary 18.01 ” African Indian -0.85"
African European 0.11
Indian European 0.96"
Light 9.27” African Indian 0.46”
African European -0.16
Indian European -0.62”
MVPA 8.98“ African Indian 0.39”
African European 0.03
Indian European -0.35
Estimated energy expenditure 3-day mean (kcal/day)
Sedentary 0.51 African Indian 3.06
African European 34.17
Indian European 31.12
Light 4.52” African Indian 46.41"
African European 1 5.68
Indian European -30.72
MVPA 9.41” African Indian 89.36"
African European 9.55
Indian European -79.81 ’
Total 8.01 " African Indian 138.83”
African European 59.41
Indian European -79.41
Energy expenditure, 3-day mean (kcal/kg/day)
Sedentary 3. 10‘ African Indian -0.37
African European 0.67
Indian European 1.04‘
Light 4.80“ African Indian 1.26”
African European 0.17
Indian European -1.09
MVPA 8.87" African Indian 2.67”
African European 0.23
Indian European -2.44‘
Total 13.97" African Indian 3.56"
African European 1.06
Indian European -2.50‘
PAQ-C score 0.15 African Indian -0.04
African European 0.00
Indian European 0.03
* p<0.05 ** p<0.01

110

The sitting height/stature ratios indicate proportionally shorter legs in European
boys, or conversely, proportionally longer legs in Indian boys, but African and Indian
boys do not differ in relative leg length. Racial differences in subcutaneous fat are
limited to the trunk skinfolds. The sum of trunk and total skinfolds are significantly
thicker in Indian than African boys. Indian boys also have higher TER, indicating that
proportionally more subcutaneous fat is accumulated on the trunk than on the extremities
compared to African and European boys. In contrast, skeletal (biacromial and bicristal)
breadths tend to be larger in European boys than in African and Indian boys, who do not
differ in these variables. Europeans have less variation in the bicristal and biacromial
breadths than in Africans and Indians. The biacromial/bicristal ratio is significantly
greater in European than African or Indian boys, indicating proportionately larger
shoulders relative to the hips.

The ﬁtness variables show no consistent patterns of racial variation. Indian boys
are slower than African boys on the dash. European boys perform better on the distance
run and push-ups than African and Indian boys. African boys are more ﬂexible on the sit
and reach than Indian boys.

Though signiﬁcant, estimated time spent in activities of different intensities
differs slightly among race groups. Indian boys spend slightly more time in sedentary
and light activities than African and European boys. Indian boys engage in less MVPA
and spend more time playing computer games and watching television than African boys.
African boys have higher light, MVPA and total estimated energy expenditure (kcal/day

and kcal/kg/day) than Indian boys, and spend more time commuting to school than

111

Indian and European boys. The estimated energy expenditure of European boys does not

differ signiﬁcantly from African or Indian boys.

Within race descriptive statistics

The means, standard deviations, and medians (where appropriate) for all variables
for African and Indian boys are presented by SES in Tables 26 and 27, respectively. The
tables also include results of the ANOVA comparing higher and lower SES boys within
each race group. A comparison of European and higher SES African and Indian boys is
shown in Table 28. Between group differences of higher SES boys for all signiﬁcant
AN OVAs were tested with the Scheﬂ‘é Post Hoe test (Table 29).

African boys from higher SES families are taller and heavier, have absolutely
longer legs, thicker skinfolds, larger limb and estimated arm muscle circumferences, and
larger bicondylar and biepicondylar breadths compared to African boys from lower SES
families (Table 26). There are no signiﬁcant differences in ﬁtness, except for absolute
grip strength, which is greater in higher SES boys. The SES-related difference in grip
strength is not signiﬁcant when expressed in per kilogram body mass. Time spent in
sedentary, light, and MVPA activities; in television viewing and computer games; and in
commuting to school does not differ between SES groups of African boys. In terms of
estimated absolute energy expenditure, higher SES African boys tend to expend more
energy. This is a function of their larger body mass because there are no SES differences
in estimated energy expenditure per unit mass.

In contrast to African boys, SES differences in Indian boys are relatively few

(Table 27). Higher SES Indian boys have larger arm and calf circumferences (related to

112

Table 26. Descriptive statistics for African boys by SES.

 

 

 

 

Lower SES Higher SES
N M so Md N M so Md r

Age, yrs 101 10.5 0.3 108 10.6 0.3 1.59
Mass, kg 101 31.0 5.0 29.9 108 34.1 6.5 33.3 15.10”
swim, cm 101 136.4 5.8 108 139.2 6.7 140.1 10.08”
BMI, leg/m2 101 16.6 2.2 16.2 108 17.6 2.8 7.31"
Sitting height. cm 101 69.0 3.4 108 69.8 3.9 2.76
Leg length, cm 101 67.5 4.5 108 69.4 4.8 8.56“
Sit Ill/Stat, % 101 50.6 2.0 108 50.2 2.1 1.90
Skinfolds, mm

Biceps 101 5.8 3.0 5.0 108 7.4 4.2 5.6 9.40"

Triceps 101 10.9 5.3 9.5 108 12.8 5.7 10.8 5.97“

Subscapular 101 7.7 4.2 6.4 108 9.6 6.4 7.0 6.62”

Suprailiac 101 5.8 3.6 4.9 108 7.9 5.2 5.8 10.35"

Abdominal 101 9.5 6.1 7.4 108 12.0 7.5 8.8 6.52“

2 omek 101 17.2 10.0 14.3 108 21.6 13.4 16.2 6.92“

1: ofExtremity 101 16.8 8.0 14.2 108 20.2 9.6 15.8 7.77"

2 of5 101 39.9 21.1 34.6 108 49.7 26.9 37.8 8.29“

TER. min/mm 1.02 0.21 1.01 108 1.05 0.25 1.06 0.93
Circumference, cm

Relaxed arm 101 19.7 2.1 19.3 108 0.6 2.7 20.2 7.87“

Calf 101 27.2 2.2 27.0 108 28.2 2.8 27.7 8.57”
Arm muscle area. can2 101 28.4 6.4 27.4 108 30.7 9.1 29.0 4.63‘
Breadth, cm

Biacromial 101 28.4 1.9 108 28.9 2.7 1.97

Bicristal 101 19.6 2.0 108 19.6 2.2 0.11

Biepicondylar 101 5.5 0.4 108 5.6 0.4 4.94.

Bicondylar 101 8.0 0.5 108 8.2 0.6 6.57"

Bier/Bias, % 101 69.0 5.5 108 67.9 4.5 2.98‘
9 p<0.05. with 17 ratios >3.66
” p<0.01. with F ratios >597

113

Table 26. (continued)

 

 

 

 

Lower SES Higher SES
N Mean SD N Mean SD F
Physical ﬁtness
Dash, sec 97 9.4 1.3 104 9.5 1.2 0.58
Grip, kg 97 18.6 3.1 108 19.9 3.9 6.53”
Grip, kg/kg mass 97 0.6 0.1 108 0.6 0.1 1.52
Shuttle run, sec 97 23.0 2.4 106 23.0 2.5 0.04
Standing long jump, cm 97 149.9 19.9 103 151.3 23.0 0.20
9 - minute run, m 97 1612.0 314.3 108 1627.6 397.9 0.10
Push-ups, #/30 see 97 24.1 7.6 107 23.4 8.9 0.37
Sit and reach, cm 97 28.4 5.6 107 27.4 5.8 1.59
Sit-ups, #/30 see 97 16.5 5.2 106 16.3 5.6 0.10
Time in sedentary activities, 3 day mean, hours
Television 90 2.2 1.3 97 2.2 1.3 0.12
Commute 90 1.4 0.8 97 1.5 0.8 0.55
Computer Games 90 0.1 0.4 97 0.1 0.3 0.45
Energy expenditure, 3-day mean, hours
Sedentary 90 19.6 1.4 97 19.8 1.2 2.27
Light 90 2.7 1.0 97 2.7 0.9 0.62
MVPA 90 1.7 0.7 97 1.5 0.9 2.85
Energy expenditure, 3-day mean (kcal/day)
Sedentary 90 670.7 128.2 97 762.7 166.1 16.93”
Light 90 263.1 116.6 97 285.7 128.0 1.52
MVPA 90 313.7 156.1 97 298.4 191.7 0.34
Total 90 1247.5 236.9 97 1346.8 295.3 6.07“
Energy expenditure, 3-day mean (kcal/kg/day)
Sedentary 90 21.7 2.1 97 22.2 1.9 0.06
Light 90 8.5 3.4 97 8.3 3.3 0.77
MVPA 90 10.3 5.2 97 8.9 5.8 3.66‘
Total 90 40.5 5.5 97 39.5 5.6 1.49
PAQ-C score 98 2.7 0.4 97 2.8 0.5 1.2
‘ p<0.05, with F ratios >3.66

” p<0.01, with F ratios >597

114

Table 27. Descriptive statistics for Indian boys by SES.

 

Lower SES (n=33-38)

Higher SES (n=86-99)

 

 

N M SD Md N M SD Md F
Age, yrs 38 10.6 0.3 101 10.5 0.3 2.74
Mass, kg 38 30.5 7.0 29.3 101 33.0 7.8 31.3 3.02
Stature, cm 38 137.6 7.0 101 139.5 6.8 2.05
BMI, kg/m2 38 15.9 2.5 15.2 101 16.8 3.0 16.1 2.72
Sitting height, cm 38 69.0 3.8 101 69.9 3.4 1.50
Leg length, cm 38 68.6 4.7 101 69.6 5.1 1.21
Sit Ht/Stat, % 38 50.2 1.8 101 50.1 1.9 0.03
Skinfolds, mm
Biceps 38 6.9 3.7 5.2 101 7.5 4.3 6.2 0.68
Triceps 38 11.5 4.9 9.8 101 13.5 6.1 12.2 3.16
Subscapular 38 8.7 5.4 6.5 101 10.6 8.0 7.6 1.88
Suprailiac 38 8.0 6.2 5.6 101 9.1 6.6 6.6 0.81
Abdominal 38 12.3 7.9 8.9 101 14.5 9.7 11.2 1.47
2 Trunk 38 21.0 12.8 15.6 101 25.6 17.9 18.8 2.12
2 Extremity 38 18.4 8.4 14.6 101 21.3 10.4 18.3 2.41
2 5 38 47.3 26.7 36.7 101 56.2 33.8 45.2 2.12
TER,mm/mm 1.10 0.22 1.14 101 1.14 0.30 1.13 0.67
Circumference, cm
Relaxed arm 38 19.3 2.7 18.7 101 20.5 3.0 20.0 4.10‘
Calf 38 26.3 2.8 26.2 101 27.5 3.0 27.0 4.77‘
Arm muscle area, em2 38 31.3 8.5 28.6 101 31.6 9.4 29.3 0.03
Breadth, cm
Biacromial 38 28.1 2.4 101 28.8 2.3 3.12
Bicristal 38 19.8 2.2 101 19.8 2.1 0.00
Biepicondylar 38 5.5 0.4 101 5.5 0.4 0.39
Bicondylar 38 8.1 0.5 101 8.1 0.5 0.58
Bier/biac, % 38 71.0 6.3 101 69.0 6.0 3.15

 

' p<0.05, with F ratios >4.10
” p<0.01

115

Table 27. (continued)

 

 

 

 

Lower SES Higher SES
N M SD N M SD F
Physical ﬁtness
Dash, sec 38 9.6 0.8 98 9.8 1.3 0.54
Grip, kg 38 17.5 3.0 98 18.9 3.4 4.46‘
Grip, kg/kg mass 38 0.59 0.09 98 0.59 0.11 0.00
Shuttle run, sec 38 22.4 2.7 98 23.5 2.8 4.42’
Standing long jump, cm 38 147.7 20.5 98 150.4 22.7 0.41
9-minute run, m 38 1665.0 344.0 98 1510.8 328.8 5.87‘
Push-ups, #/30 sec 38 23.9 8.6 98 25.8 8.2 1.38
Sit and reach, cm 38 23.6 5.7 98 24.2 6.0 0.30
Sit-ups, #l30 sec 38 17.7 6.9 98 15.9 5.4 2.70
Time in sedentary activities, 3 day mean, hours
Television 33 2.2 1.4 88 2.6 1.6 1.47
Commuting 33 0.2 0.3 88 0.2 0.4 0.04
Computer Games 33 0.2 0.5 88 0.2 0.6 0.09
Energy expenditure 3-day mean, hours
Sedentary 33 20.3 1.1 88 20.6 1.1 1.01
Light 33 2.4 0.9 88 2.3 0.9 0.95
MVPA 33 1.2 0.7 88 1.2 0.7 0.05
Energy expenditure, 3-day mean (kcal/day)
Sedentary 33 674.3 151.3 88 754.7 205.1 420'
Light 33 236.7 1 16.4 88 235.3 139.6 0.00
MVPA 33 206.6 143.5 88 223.9 155.9 0.31
Total 33 1117.6 278.4 88 1213.9 341.6 2.10
Energy expenditure, 3-day mean (kcal/kg/day)
Sedentary 33 22.2 2.0 88 22.3 1.8 0.06
Light 33 7.7 3.2 88 6.9 3.4 1.32
MVPA 33 6.9 4.0 88 6.8 4.7 0.00
Total 33 36.8 4.3 88 36.1 5.4 0.49
PAQ-C score 38 3.0 0.5 98 2.8 0.7 1.92
‘ p<0.05, with F ratios >4.10
” p<0.01

116

slightly larger limb skinfold thicknesses, respectively) and are stronger in grip strength
compared to Indian boys from lower SES families. Lower SES Indian boys perform
better (p < 0.05) in the 9 minute and shuttle runs, but higher SES Indian boys expend
more energy in sedentary activities. The higher estimated energy expenditure (kcal/day)
of higher SES boys apparently reﬂects their larger body mass since estimated energy
expenditure relative to body mass does not differ signiﬁcantly. All other variables do not
differ signiﬁcantly between higher and lower SES Indian boys.

Anthropometry, physical ﬁtness and estimated energy expenditure of higher SES
African, Indian and European boys are shown in Tables 28 and 29. Race differences in
the sitting height/stature and bicristal/biacromial ratios, and bicristal breadth are
signiﬁcant. European boys have proportionally longer trunks relative to their legs than
higher SES African and Indian boys. Higher SES African and Indian boys have
proportionally broader shoulder relative to the hips than European boys. The estimated
subcutaneous adipose tissue of higher SES African, Indian and European boys do not
differ signiﬁcantly.

Performances in the distance run, push-ups and sit and reach are signiﬁcantly
different among higher SES African, Indian and European boys. European boys run a
greater distance in nine minutes and complete more push-ups in 30 seconds than higher
SES African boys. European boys also have better distance run performances than
higher SES Indian boys. There are no signiﬁcant differences in the distance run and
push-ups performances of higher SES African and Indian boys, or the push-ups

performance of European and higher SES Indian boys. Higher SES African boys have

117

Table 28. Descriptive statistics for higher SES African, Indian and European boys.

 

 

 

 

 

African Indian European l
N M SD Md N M SD Md N M SD Md I F

Age, yrs 108 10.6 0.3 101 10.5 0.3 41 10.5 0.4 0.95
Mass, kg 108 34.1 6.5 33.3 101 33.6 8.6 31.3 41 33.7 5.7 33.3 0.16
Stature. cm 108 139.2 6.6 101 139.5 6.8 41 139.2 6.4 0.07
BMI. kg/rn2 108 17.6 2.8 16.9 101 16.8 3.0 16.1 41 17.4 2.6 17.4 1.64
Leg length. cm 108 69.4 4.8 101 69.6 5.1 41 67.5 4.2 1.45
Sitting height. cm 108 69.8 3.9 101 69.9 3.5 41 71.8 2.9 4.71“
SH/ST, % 108 50.2 2.1 101 50.1 1.9 41 51.6 1.2 9.61”
Skinfolds, mm

Biceps 108 7.4 4.2 5.6 101 7.7 4.5 6.2 41 6.6 3.6 5.4 0.82

Triceps 108 12.8 5.7 10.8 101 13.7 6.3 12.2 41 13.2 5.9 11.2 0.54

Subseapular 108 9.6 6.4 7.0 101 10.9 8.4 7.7 41 8.0 5.1 6.3 2.54

Suprailiac 108 7.9 5.2 5.8 101 9.3 7.0 6.6 41 7.5 5.2 5.6 2.05

Abdominal 108 12.0 7.5 8.8 101 14.7 10.0 11.2 41 11.7 7.2 9.7 3.01

2'. of Trunk 108 21.6 13.4 16.2 101 25.6 17.9 18.8 41 19.7 11.8 15.9 2.90

20f Extremity 108 20.2 9.6 15.8 101 21.3 10.4 18.3 41 19.9 9.3 16.7 0.47

 

Z of 5 108 49.7 26.9 37.8 101 56.2 33.8 45.2 41 47.0 25.3 38.8 1.92
TER. rum/mm 108 1.05 0.25 1.06 101 1.15 0.30 1.13 41 0.98 0.23 0.95 3.00
Circumference, cm
Relaxed arm 108 20.6 2.7 20.2 101 20.6 3.2 20.0 41 21.0 2.5 21.0 0.24
Calf 108 28.2 2.8 27.7 101 27.6 3.1 27.1 41 28.1 2.71 28.3 1.91
Arm muscle area, cm2 108 30.7 9.1 28.8 101 31.9 9.9 29.4 41 30.9 7.0 30.7 0.42
Beadth, cm
Biacromial 108 28.9 2.7 101 28.9 2.3 41 29.8 1.3 2.64
Bicristal 108 19.6 2.2 101 19.8 2.1 41 21.4 1.3 12.55"
Biepicondylar 108 5.6 0.4 101 5.6 0.4 41 5.6 0.4 0.53
Bicondylar 108 8.2 0.6 101 8.1 0.6 41 8.3 0.5 2.07
Bier/biac, % 108 67.9 4.7 101 68.8 5.5 41 72.0 3.6 10.51“
‘ p<0.05
" p<0.01

118

Table 28. (continued)

 

 

 

 

 

 

 

African Indian European
N M SD N M SD N M SD F
Physical ﬁtness
Dash, see 104 9.5 1.1 101 9.8 1.3 41 9.4 1.0 2.68
Grip, kg 108 19.9 3.8 98 19.0 3.5 41 19.3 3.0 1.90
Grip, kg/kg mass 108 0.6 0.1 98 0.6 0.1 41 0.6 0.1 0.02
. Shuttle run, see 106 23.0 2.5 101 23.5 2.7 41 22.5 2.2 2.19
Standing long jump, cm 103 151.1 22.9 101 150.6 22.5 41 157.7 21.9 1.61
9-minute run, at 108 1621.0 399.2 101 1512.7 326.8 41 1875.4 471.6 12.99”
Puahoups, #l30 sec 107 23.4 8.8 99 25.8 ' 8.2 41 28.0 8.0 4.78”
Sit and reach, cm 107 27.3 5.8 99 24.1 6.0 41 25.7 5.4 7.55”
Sit-ups, #30 sec 106 16.3 5.5 102 16.0 5.5 41 17.2 5.5 0.68
Time in sedentary activities, 3-day mean, hours
Television 94 2.5 1.4 88 3.1 1.8 28 1.8 1.1 8.04”
Commuting 94 2.7 1.4 88 0.2 0.3 28 0.3 0.46 101.10“
Computer games 94 1.2 0.6 88 3.5 1.9 28 2.6 1.90 5.79"
Energy expenditure 3-day mean, hours
Sedentary 94 20.0 1.2 88 20.6 1.1 28 19.7 1.3 10.09“
Light 94 2.7 0.9 88 2.2 0.9 28 2.9 1.0 8.58“
MVPA 94 1.5 0.9 88 1.2 0.7 28 1.5 0.8 4.01 ‘
Energy expenditure 3-day mean (keal/day)
Sedentary 94 762.7 166.1 88 754.7 205.1 28 692.6 144.8 1.67
Light 94 285.7 128.0 88 235.3 139.6 28 269.4 1 1 1.6 3.40‘
MVPA 94 298.4 191.7 88 223.9 155.9 28 307.8 204.9 4.69”
Total 94 1346.8 295.3 88 1213.9 341.6 28 1269.8 263.7 4.14‘
Energy expenditm 3-day men (bulked-y)
Sedentary 94 22.2 1.9 88 22.3 1.8 28 21.3 2.5 3.36‘
light 94 8.3 3.3 88 6.9 3.4 28 8.4 3.3 4.49“
MVPA 94 8.9 5.8 88 6.8 4.7 28 9.6 6.1 4.70”
Total 94 39.5 5.6 88 36.1 5.4 28 39.2 6.6 8.90“
PAQ—C score 108 2.8 0.5 101 2.8 0.7 41 2.8 0.5 0.15
‘ p<0.05
” p<0.01

119

Table 29. Scheffé post hoc tests of mean differences (MD) between races for
variables showing a signiﬁcant F ratio among higher SES boys.

 

 

 

 

Scheffé post hoc race differences
Dependent variable Race Race MD
Growth Status
Sitting height! stature ratio, % African Indian 0.05
African European -1.39“
Indian European -1.42“
Bicristal African Indian -0.02
African European -1.82“
Indian European -l.79“
Bicristal/biacromial ratio, % African Indian -1 .68
African European -5.06“
Indian European -3.38‘
Physical Fitness
Grip, kg African Indian 1.19‘
African European 0.54
Indian European -0.64
9-minute run, m African Indian 91.32
African European 460.76”
Indian European -352.08”
Push-ups, #30 sec African Indian -2.96‘
African European 547”
Indian Elnopean -2.52
Sit and reach, cm African Indian 2.90“
African European 1.57
Indian European -1.33
Time in sedentary activities, hours
Television African Indian -0.60‘
African European 0.68
Indian European 1.28“
Commuting African Indian 0.88“
African European 0.78“
Indian European 01]
Computer Games African Indian -0.79”
African European -0.03
Indian European 0.82
Estimated energy expenditure 3-day mean, hours
Sedentary African Indian -0.68"
African European 0.32
Indian European 1.00”
Light African Indian 0.40
African European -0.32
Indian European -0.72”
Estimated energy expenditure 3-day mean (kcal/day)
Total 'can Indian 136.40‘
African European 83.57
Indian European -52.83
Energy expenditure, 3-day mean (kcal/kg/day)
Total African Indian 3.23‘
African European 0.40
Indian European -2.83
"' p<0.05 8* p<0.01

120

greater ﬂexibility of the lower back on the sit and reach test than higher SES Indian boys;
higher SES Indian and European boys, however, do not differ signiﬁcantly.

There are signiﬁcant race differences among higher SES boys in the time spent in
commuting, television viewing and playing computer games. Higher SES African boys
spend more time commuting than higher SES Indian and European boys. Higher SES
Indian and European boys have similar commutes. Higher SES Indian boys watch more
television than African and European boys, who do not differ signiﬁcantly from each
other. Higher SES Indian boys also play signiﬁcantly more computer games than higher
SES African boys. Higher SES Indian boys spend more time in sedentary activities than
higher SES African and European boys and have a higher relative energy expenditure
(kcal/kg/day) in sedentary activity than higher SES African boys. European boys spend
more time engaged in light activities than higher SES Indian boys. There are no
signiﬁcant differences in the estimated energy expenditure (kcal/day and kcal/kg/day) in
light activities among European and higher SES African and Indian boys. Higher SES
African boys have higher mean daily total estimated energy expenditure (kcaI/day and
kcal/kg/day) than higher SES Indian boys.

Means, standard deviations and medians (where appropriate), ANOVAs and
Scheffe’ post hoc tests of all variables for African, Indian and European boys by
nutritional status are presented in Tables 30-31, 32-33, and 34, respectively. These are
shown separately to reduce the confounding eﬂ‘ects of race on nutritional status.

BMI-for-age was used as a proxy for nutritional status. Regardless of race,
overweight boys have larger skinfold thicknesses, circumferences and estimated arm

muscle areas, and proportionally more subcutaneous fat on the trunk relative to the

121

Table 30. Descriptive statistics of African boys by nutritional status based on the

 

 

 

 

BM].
5 15 percentile 15th>NN<85th 285th percentile
(BMI-for-Age) (BMI-for—Age) (BMI-for-Age)
Undemourished Nutritionally normal Overweight
% of sample 18.3% 68.3% 13.5%
N M SD Md N M SD Md N M SD Md F
Age, years 38 10.5 0.3 142 10.5 0.3 28 10.5 0.3 0.55
Mass, ltg 38 27.4 3.1 27.0 142 31.8 3.6 31.8 28 43.7 5.2 41.9 88.88“
Stature, cm 38 137.7 6.2 142 137.4 6.5 28 140.2 5.9 1.82
BMI 38 14.4 0.9 14.5 142 16.8 1.1 16.7 28 22.3 2.3 22.2 148.13“
Leg length, cm 38 69.7 4.3 142 68.0 4.7 28 68.9 5.5 0.81
Sitting height, cm 38 68.0 3.4 142 69.4 3.5 28 71.3 4.2 2.62
Sit ht! Stat,% 38 49.4 1.7 142 50.5 1.9 28 50.9 2.8 1.05
Skinfolds, nun
Biceps 38 4.4 1.3 4.4 142 6.0 2.8 5.2 28 12.8 3.8 12.9 26.66“
Triceps 38 8.1 2.2 7.9 142 10.9 3.8 9.8 28 22.0 4.9 21.1 56.48“
Subscapular 38 5.5 1.2 5.3 142 7.5 3.4 6.8 28 18.9 6.3 17.2 45.06”
Suprailiac 38 4.3 1.0 4.2 142 5.8 2.3 5.4 28 15.7 6.0 14.6 58.70”
Abdominal 38 6.1 1.5 5.6 142 9.2 3.9 8.5 28 25.0 5.9 24.9 95.02“
E of Trunk 38 11.6 2.3 11.4 142 16.8 6.5 15.6 28 43.9 11.7 41.1 82.15”
2‘. of Extremity 38 12.5 3.1 12.2 142 16.9 6.2 15.2 28 34.9 8.4 33.7 47.75”
2 of 5 38 28.4 5.4 27.5 142 39.6 13.5 36.6 28 94.5 24.0 89.6 77.28”
TER. mm/mm 38 0.95 0.19 0.98 142 1.01 0.21 1.02 28 1.28 0.25 1.23 8.56”
Circumference, cm
Relaxed arm 38 17.9 1.1 18.0 142 19.8 1.5 19.8 28 24.7 2.2 24.6 71.05“
Calf 38 25.6 1.4 26.0 142 27.5 1.7 27.4 28 32.0 2.7 32.1 85.09“
AMA', cm: 38 26.3 7.1 25.7 142 29.3 6.6 28.3 28 35.8 1 1.6 35.5 20.36“
Breadths, cm
Biacromial 38 27.7 2.6 142 28.6 2.1 28 30.3 2.6 6.67”
Bicristal 38 17.6 2.2 142 18.6 2.1 28 19.5 3.1 4.17‘
Biepicondylar 38 5.4 0.3 142 5.5 0.4 28 5.8 0.3 11.23"
Bicondylar 38 7.8 0.4 142 8.1 0.4 28 8.8 0.5 26.20"
Bier/Biac, % 38 63.6 5.5 142 65.2 5.7 28 64.1 6.4 0.44
0 p<0.05 ” p<0.01

' AMA - urn muscle area

122

Table 30. (continued)

 

 

 

 

S 15 percentile 15th > NN < 85th 2 85th percentile
(BMI-for-Age) (BMI-for-Age) (BMI-for-Age)
Undemourished Nutritionally normal Overweight
N M SD N M SD M SD F
Physical frtness

Dash, sec 38 9.2 1.4 138 9.4 1.2 28 9.6 1.2 1.55

Grip, kg 38 18.0 2.6 138 19.2 3.4 28 21.5 4.6 9.30”

Grip, kg/kg mass 38 0.7 0.1 138 0.6 0.1 28 0.5 0.1 9.11”

Shuttle run, see 38 22.8 2.3 138 23.0 2.6 28 23.3 2.2 0.89

Standing long jump, cm 38 152.9 21.0 138 149.8 22.1 28 151.5 19.7 0.43

9 - minute run, in 38 1566.2 335.5 138 1661.4 365.9 28 1489.9 326.9 2.34

Push-ups, #60 sec 38 23.1 7.3 138 24.6 8.4 28 20.4 8.5 1.64

Sit and reach, cm 38 27.3 5.8 138 27.9 5.4 28 28.1 7.0 0.02

Sit-ups, #80 sec 38 15.2 4.2 138 17.3 5.7 28 13.8 3.7 4.42‘ "
Time in sedentary activities, 3 day mean, hours

Television 29 2.0 1.1 125 2.3 1.3 24 2.0 1.4 0.09

Commute 29 1.7 0.8 125 1.3 0.8 24 1.4 1.0 0.34

Computer Game 29 0.1 0.2 125 0.1 0.3 24 0.2 0.4 2.32
Energy expenditure 3-day mean, hours

Sedentary 29 19.8 1.4 125 19.8 1.3 24 19.9 1.1 0.99

Light 29 2.5 0.9 125 2.8 0.9 24 3.0 0.9 0.44

MVPA 29 1.8 0.8 125 1.6 0.8 24 1.3 0.7 1.98
Energy expenditure, 3-day mean (Real/day)

Sedentary 29 590.7 80.1 125 704.0 107.4 24 968.1 157.3 48.13”

Light 29 207.8 90.8 125 269.7 1 10.0 24 388.0 148.3 5.36“

MVPA 29 314.9 135.8 125 296.8 177.7 24 343.2 207.8 0.24

Total 29 1113.4 144.3 125 1270.6 229.0 24 1699.3 201.1 36.46“
Energy cmdinre. 3-dny mean (Realms/day)

Sedentary 29 21.7 1.5 125 22.1 2.1 24 21.9 2.2 2.00

Light 29 7.8 3.5 125 8.4 3.3 24 8.9 3.5 0.60

MVPA 29 11.8 5.5 125 9.3 5.6 24 7.9 4.8 1.75

Total 29 41.2 6.0 125 39.9 5.6 24 38.6 4.2 2.10
PAQ-C score 38 2.7 0.3 138 2.8 0.5 28 2.6 0.3 1.06
‘ p<0.05 ‘ “ p<0.01

123

extremities (higher TER), while undernourished boys have the lowest fatness and
estimated arm musculature, proportionally smaller SAT on the trunk relative to the
extremities, and the smallest circumferences and breadths. For all signiﬁcant ANOVAs,
the Scheffé post hoc test was used to test speciﬁc pairs of means (Tables 31 and 33, for
African and Indian boys, respectively).

Skinfold thicknesses, TERs, circumferences, estimated arm muscle area, and
skeletal breadths of overweight boys are signiﬁcantly larger than undernourished and
nutritionally normal African boys. Nutritionally normal boys also have greater sum of
skinfolds, circumferences and estimated arrn muscle area compared to undernourished
boys

Overweight African boys have signiﬁcantly greater absolute and relative grip
strength than undernourished and nutritionally normal boys. But, when grip strength is
expressed relative to body weight, both undernourished and nutritionally normal African
boys have greater grip strength per kilogram mass than overweight African boys. Grip
strength relative to body mass (kg/kg) of undernourished and nutritionally normal
African boys do not differ signiﬁcantly. Nutritionally normal African boys have greater
absolute grip strength compared to undernourished African boys.

Overweight African boys have signiﬁcantly greater sedentary, light and mean
daily total estimated energy expenditure (kcaI/day) compared to undernourished and
nutritionally normal African boys, and nutritionally normal African boys have greater
estimated energy expenditure in sedentary and mean daily total estimated energy
expenditure (kcal/day) than undernourished African boys. The estimated energy

expenditure (kcal/day) differences between overweight. nutritionally normal and

124

Table 31. Scheffé post hoc tests of mean differences (MD) between nutritional
status of African boys for variables showing a significant F ratio.

 

 

 

Dependent variable Nutritional Status' MD
Mass, kg UN NN 4.65”
UN OW -18.37”
NN OW 43.72”
BMI, kg/m2 UN NN -235"
UN OW -8.47"
NN OW -6.12"
Skinfolds, mm
Biceps UN NN -2.01
UN OW -8.58”
NN OW -6.57“
Triceps UN NN -3.80“
UN OW 45.35"
NN OW -11.56”
Subscapular UN NN -2.43
UN OW 43.94”
NN OW -11.50"
Suprailiac UN NN -l.86
UN OW 42.78"
NN 0W 40.93”
Abdominal UN NN -3.42"
UN OW -19.61"
NN OW -16.19"
2 of Trunk UN NN -5.85‘
UN OW -33.55”
NN OW -27.70”
'2 of Extremity UN NN -5.81"
UN OW ~23.94”
NN OW 48.13”
22 of 5 UN NN -l3.51”
UN OW 40.27”
NN OW -56.75”
Circumferences, cm
Relaxed arm UN NN -2.26”
UN OW 4.25”
NN OW 4.99”
Calf UN NN -2.22”
UN OW -7.74“
NN OW -5.52”
Estimated muscle area, cm2
Arm UN NN -5.02‘
UN OW 46.86”
NN OW -1 1.84"
'UN: undernourished, NN: nutritionally normal, and OW: overweight.
‘ p<0.05
” p<0.01

125

Table 31. (continued)

 

 

 

Dependent variable Nutritional Status MD
Breadth, cm
Biacromial UN NN -0.81
UN OW -2.76“
NN OW -1.94”
Bicristal UN NN -0.93
UN OW -2.27‘
NN OW -1.34
Biepicondylar UN NN -0. 19
UN OW -0.65”
NN OW -0.46”
Bicondylar UN NN -0.20
UN OW -1.07"
NN OW -0.87”
Physical Fitness
Grip, kg UN NN -2.84‘
UN OW -5.82"
NN OW -2.99‘
Grip, kg/kg mass UN NN 0.00
UN OW 0.12”
NN OW 0.12”
Sit-ups, #l30 sec UN NN -2.76
UN OW 1.81
NN OW 4.58‘
Estimated energy expenditure 3-day mean (kcal/day)
Sedentary UN 435.20"
UN OW 420.50”
NN OW -285.30”
Light UN NN -15.05
UN OW -1 16.32‘
NN OW -101.27"
Total UN NN -1 53.33”
UN OW 674.42”
NN OW 421.09“
’ p<0.05
” p<0.01

126

undernourished African boys are related to body size since estimated energy expenditure
per kilogram body mass do not differ signiﬁcantly among African boys from different
nutritional status groups.

Overweight Indian boys are signiﬁcantly taller, heavier and have longer trunks
than nutritionally normal and undernourished Indian boys. Nutritionally normal Indian
boys are heavier and have longer trunks than undernourished Indian boys. Overweight
Indian boys have signiﬁcantly larger skinfold thicknesses, TER, circumferences and
breadths (biepicondylar and bicondylar) compared to nutritionally normal or
undernourished Indian boys (Tables 32 and 33). Additionally, overweight Indian boys
have signiﬁcantly greater TERs than undernourished and nutritionally normal boys,
indicating that overweight boys have a larger accumulation of SAT on the trunk relative
to their arms compared to undernourished and nutritionally normal Indian boys.
Overweight Indian are have broader shoulders and hips compared to nutritionally normal
boys, but do not differ in the bicristal/biacromial ratio. Nutritionally normal Indian boys
have thicker skinfolds and larger circumferences than undernourished Indian boys.

Overweight Indian boys are signiﬁcantly slower in the dash than nutritionally
normal boys. Overweight Indian boys, on the other hand, have greater absolute grip
strength (kgs) than undernourished or nutritionally normal boys, and nutritionally normal
Indian boys have greater grip strength (kg) than undernourished boys. This, however, is
factor of their larger body size since both nutritionally normal and undernourished Indian
boys have greater grip strength per unit body mass compared to overweight Indian boys,

and undernourished and nutritionally normal boys do not differ in grip strength per unit

127

Table 32. Descriptive statistics of Indian boys by nutritional status based on the

 

 

 

 

BMI.
S 15 percentile >15th > NN < 85th 2 85th percentile
(BMI-for-Age) (BMI-for-Age) (BMI-for-Age)
Undemourished Nutritionally normal Overweight
% of the sample 36.5% 47.4% 16.1%
N M SD Md N M SD Md N M SD Md F

Age 50 10.5 0.3 65 10.5 0.3 22 10.6 0.3
Mass, kg 50 26.5 3.0 26.2 65 32.2 3.9 31.7 22 45.9 6.0 44.9 89.98”
Stature, cm 50 136.7 6.8 65 139.0 6.3 22 144.1 6.3 10.00"
BMI. kg/nr2 50 14.1 0.7 14.3 65 16.6 1.1 16.2 22 22.1 2.0 21.7 166.08“
Leg length. cm 50 68.6 5.0 65 69.2 4.5 22 71.4 5.8 1.02
Sitting height, cm 50 68.1 3.5 65 69.8 3.2 22 72.7 2.5 15.83“
Sit ht/ Stat.% 50 49.9 2.0 65 50.2 1.6 22 50.5 2.3 1.19
Skinfolds, mm 50 65 i 22

Biceps 50 4.9 1.5 4.8 65 6.9 3.0 6.2 22 14.1 3.9 14.1 66.26”

Triceps 50 9.3 3.4 8.5 65 12.7 4.1 12.2 22 22.2 4.2 21.9 47.29”

Subaeapular 50 6.2 1.5 6.0 65 8.5 3.3 7.7 22 23.3 9.4 21.7 53.23”

Suprailiac 50 5.0 1.6 5.0 65 7.8 4.3 6.6 22 19.9 6.8 21.7 42.33”

Abdominal 50 7.7 3.0 7.3 65 13.1 6.1 11.6 22 30.2 7.2 29.1 63.29”

)3 of Trunk 50 13.9 3.8 14.2 65 21.6 8.9 19.1 22 53.5 15.4 52.8 71.39”

2 of Extremity 50 14.2 4.5 13.1 65 19.6 6.6 18.5 22 36.3 7.5 37.5 62.87“

2 of 5 50 33.2 8.5 32.6 65 49.0 18.4 43.2 22 109.7 24.8 112.4 75.66”

TER, mmlmm 50 1.01 0.21 1.03 65 1.11 0.23 1.14 22 1.47 0.30 1.49 9.06”
Cirotunference.

Relaxed arm 50 17.7 1.2 65 20.4 1.7 22 25.1 1.9 81.48“

Calf 50 24.8 1.7 65 27.4 1.6 22 31.9 2.7 40.21“
AMA. cm2 50 31.3 10.6 65 30.7 8.6 22 34.4 6.9 1.31
Breadths, em 50 65 22

Biacromial 50 27.8 2.1 65 28.7 2.2 22 30.1 2.5 3.81 "

Bicristal 50 18.2 1.9 65 18.9 2.3 22 20.3 2.1 5.05”

Biepicondylar 50 5.3 0.3 65 5.5 0.3 22 6.0 0.3 18.03”

Bicondylar 50 7.8 0.4 65 8.2 0.4 22 8.7 0.5 17.28”

Bier/Biae, % 50 65.5 5.9 65 65.7 6.3 22 67.6 5.9 0.93
‘ p<0.05 “ p<0.01

I28

Table 32. (continued)

 

 

 

 

S 15 percentile >151h>NN<85th 285thpercentile
(BMI-for-Age) (BMI-for-Age) (BMI-for-Age)
Undemourished Nutritionally normal Overweight
N Mean SD N Mean SD N Mean SD F
Physical ﬁtness
Dash, sec 45 9.7 1.1 65 9.5 1.1 22 10.6 1.4 3.94‘
Grip, kg 45 16.8 2.7 65 19.0 3.5 22 21.2 2.2 12.04”
Grip, kykg mass 45 0.6 0.1 65 0.6 0.1 22 0.5 0.1 22.34”
Shuttle run, sec 45 22.7 2.0 65 23.2 3.2 22 24.1 2.8 2.99
Long jump, cm 45 146.7 21.5 65 154.0 17.2 22 143.5 32.4 2.65
9 - minute run, in 45 1609.7 304.3 65 1560.9 358.6 22 1408.9 325.5 1.63
Push-ups, #30 sec 45 25.0 7.3 65 26.2 8.9 22 22.9 8.6 1.24
Sit and reach, cm 45 23.1 5.1 65 25.4 6.2 22 21.8 6.1 0.35
Sit-ups, #/30 sec 45 17.0 5.2 65 17.3 5.7 22 12.2 6.3 4.27‘
Tuneinsedentaryactivities,3daymean,hours
Television 45 2.6 1.6 56 2.1 1.3 22 3.1 2.0 0.88
Commute 45 0.2 0.1 56 0.2 0.4 22 0.2 0.3 0.86
Computer Game 45 0.2 0.5 56 0.3 0.7 22 0.2 0.3 0.05
Energy Expenditure 3-day mean, hours
Sedentary 45 20.6 1.1 56 20.4 1.0 22 20.5 1.5 1.78
Light 45 2.3 1.0 56 2.3 0.8 22 2.3 1.2 0.26
MVPA 45 1.1 0.6 56 1.3 0.6 22 1.1 0.9 2.15
Energy expenditure, 3-day mean (kcal/day)
Sedentary 45 592.7 84.9 56 719.0 103.2 22 1018.9 176.6 65.13“
Light 45 185.2 89. 1 56 230.1 100.0 22 339.2 187.8 6.61”
MVPA 45 178.2 128.5 56 236.6 142.1 22 253.2 209.5 4.42‘
Total 45 956.1 162.3 56 1185.7 217.8 22 1611.3 298.5 39.68‘ ‘
Energy expenditure, 3-day mean (keal/kg/day)
Sedentary 45 22.3 1.9 56 22.4 1.8 22 22.1 2.0 0.12
Light 45 7.0 3.2 56 7.2 3.0 22 7.5 4.4 0.76
MVPA 45 6.8 4.6 56 7.4 4.3 22 5.6 4.8 2.98
Total 45 36.1 4.9 56 36.8 5.1 22 35.1 5.6 3.51‘
PAQ-C score 50 2.7 0.6 65 2.9 0.7 22 3.0 0.6 0.52
a p<0.05 ” p<0.01

129

Table 33. Scheffé post hoc tests of mean differences (MD) between nutritional
status of Indian boys for variables showing a signiﬁcant F ratio.

 

 

 

Dependent variable Nutritional status' MD
Mass, kg UN NN -6.20”
UN OW -20.91“
NN OW -l4.71”
Stature, em UN NN -2.26
UN OW -7.43"
NN OW -5.17"I
BMI, kg/m2 UN NN -253"
UN OW -8.46”
NN OW -5.94“
Sitting height, cm UN NN -1.68‘
UN OW 4.64"
NN OW -2.96”
Skinfolds, mm
Biceps UN NN -2.36"
UN OW -9.55“
NN OW -7.19“
Triceps UN NN -3.23"
UN OW -13.67"
NN OW -10.44”
Subscapular UN NN -2. 19
UN OW 45.34”
NN OW -13.16”
Suprailiac UN NN -3.46"
UN OW -15.68”
NN OW 42.22”
Abdominal UN NN -6.61"
UN OW -23.32”
NN OW 46.70”
E of Trunk UN NN -8.80“
UN OW -38.66”
NN OW -29.86"
2 of Extremity UN NN -5.60”
UN OW -23.23”
NN OW 47.63”
2'. of 5 UN NN -17.85“
UN OW -77.57“
NN OW -59.7l “
TER, mm/mm UN NN -0.14
UN OW -0.41"
NN OW -0.27“
Circumferences, cm
Relaxed arm UN NN -2.66“
UN OW -7.71“
NN OW -5.06“
Calf UN NN -2.54"
UN OW -6.63”
NN OW -4.09“
'UN: undernourished, NN: nutritionally normal, and OW: overweight.
‘ p<0.05 ” p<0.01

130

Table 33. (continued)

 

 

 

Dependent variable Nutritional status - MD
Breadth, cm
Biacromial UN NN -l.31
UN OW -2.34‘
NN OW -1.02
Bicristal UN NN -1.55‘
UN OW -2.46‘
NN OW -0.91
Biepicondylar UN NN -0.20
UN OW -0.76”
NN OW -0.56”
Bicondylar UN NN -0.42“
UN OW -0.91”
NN OW -0.49"
Physical Fitness
Dash, sec UN NN 0.30
UN OW 0.80
NN OW 1.10“
Grip, kg UN NN -3.25”
UN OW 4.26“
NN 0W -1.00
Grip, kg/kg mass UN NN 0.02
UN OW 0.18“
NN OW 0.16“
Sit-ups, #/30 sec UN NN 0.30
UN OW 5.82‘
NN OW 5.52‘
Estimated energy expenditure 3-day mean (kcal/day)
Sedentary UN NN 431.89"
UN OW 468.92”
NN OW -337.04”
Light UN NN -60.30
UN OW -1 19.21”
NN OW -58.91
MVPA UN NN -100.61‘
UN OW -52.13
NN OW 48.47
Total UN NN 492.79”
UN OW -640.27”
NN OW 847.48”
Energy expenditure, 3-day mean (kcal/kg/day)
Total UN NN -1.94
UN OW 2.33
NN OW 4.27‘
‘ p<0-05 " p<0.01

131

mass. Nutritionally normal and undernourished Indian boys also complete more sit-ups
in 30 seconds than overweight boys.

Overweight Indian boys have greater sedentary and mean daily total estimated
energy expenditure (kcal/day) than nutritionally normal and undernourished Indian boys,
and nutritionally normal boys have greater estimated mean daily total energy expenditure
(keel/day) than undernourished Indian boys. The higher estimated energy expenditure is
due to the larger body size of overweight and nutritionally normal boys, respectively,
since there are no signiﬁcant differences in estimated energy expenditure relative to body
mass (kcal/kg/day). However, nutritionally normal Indian boys have a higher mean daily
total estimated energy expenditure per unit mass (keal/kg/day) than overweight Indian
boys. The number of underweight (n=6-7) and overweight (n=l-3) European boys are

too small for comparisons.

H i u h si 'vi

Subjects indicated the number of times that they participated in popular sport and
physical activities over the past seven days. The frequency of participation based on
participation or non-participation is summarized in Table 34. The eight top ranked
physical activities, in order, are soccer, cricket, jogging, bicycling, walking, playing
catch, basketball, and hide and seek. The activities with the greatest frequency of
participation (6 or more times a week, scored as 5), in order of ranking, are soccer,
bicycling, cricket, jogging, walking, roller skating, basketball, and playing catch. Only
37 (11%) boys indicated that they did not participate in soccer in the past seven days,

while 153 (44%) participated in soccer six or more times. The least popular activities, in

132

Table 34. Participation in specific physical activities by frequency of
participation in the past seven days for the total sample (n=350).

 

 

 

 

 

Rank Frequency of participation
Mean Md None 1 to 2 3 to 4 5 to 6 6 or more
Red:
1 2 3 4 5

Soccer 3.7 4 37 52 49 59 153
Cricket 2.9 3 41 67 52 56 72
Jogging 2.7 3 100 63 81 47 59
Bicycling 2.7 2 126 55 40 48 81

Walk 2.5 2 111 95 51 38 55
Catch 2.2 2 148 83 54 29 36
Basketball 2.2 2 166 68 41 32 42
Hide and seek 2.1 1 175 66 47 35 26
Roller skating 2.0 1 194 60 33 18 45
Softball 1.8 1 212 61 32 18 27
Swim 1.8 1 221 59 16 30 24
Table tennis 1.5 1 243 60 24 11 12
Volleyball 1.5 1 253 46 29 7 15
Dance 1.5 1 267 32 25 6 20
Skateboard 1 .5 1 268 33 14 13 22
Aerobics 1 .5 1 268 40 12 8 22
Skip 1.4 1 263 52 15 7 12
Hop 1.4 1 280 35 21 4 10
Field Hockey 1.4 l 294 22 10 10 14
Badminton 1 .2 l 316 18 6 6 4

Netball 1.1 1 327 14 3 0 6

 

order, are netball, badminton, ﬁeld hockey, hopping, aerobics, skipping, skateboarding
and dance. Netball, aerobics, dance and skipping are often perceived as feminine
activities, while ﬁeld hockey, badminton and skateboarding are not freely available in the
residential areas studied.

Race speciﬁc differences in the ten top ranked habitual physical activities are
presented in Table 35. The frequency of participation was rated ma 5 point scale with
1: no participation, 2: 1 to 2 times , 3: 3 to 4 times, 4: 5 to 6 times, and 5: more than 6

times a week. The scores (1 to 5) for frequency of participation in each physical activity

133

were used for all statistical calculations. Soccer has the highest frequency of
participation among African, Indian and European boys. The top-ﬁve ranked habitual
physical activities, in rank order, among African boys are soccer, bicycling, cricket,
jogging and walking. Soccer, jogging, walking, cricket, and roller skating are the top-
ﬁve ranked habitual physical activities among Indian boys. The most popular habitual
physical activities among European boys are soccer, cricket, bicycling, basketball and
walking. African boys play signiﬁcantly more soccer than Indian boys, but roller skate
less than European boys. All other race-speciﬁc differences in habitual physical activity
are not signiﬁcant. Indian boys have the highest rates of non-participation in all physical
activities (Table 35).

Among Indian boys, 50% or more do not cycle, play catch, baseball, hide and
seek, roller skate and play softball. However, while only a small sample (37.4%) of
Indian boys roller skate, they skate frequently, making roller skating the ﬁfth highest
ranked habitual physical activity of Indian boys. On the other hand, at least one-half of
the sample of African boys participate in the eight top ranked sport activities, more than
one half of European boys participate in the top-ten ranked habitual physical activities.
Indian boys have the lowest rate of participation at high frequencies (6 or more times a
week) in cricket, walking, bicycling, catch, basketball, roller-skating and softball.

The top ranked activities among lower SES boys, races combined, are soccer,
bicycling, jogging, cricket and walking (Table 36). The top ranked activities among
higher SES boys are soccer, bicycling, cricket, walking and jogging. There are no
signiﬁcant SES- or nutritional status-related differences (races combined) in habitual

physical activity scores (Tables 36 and 37).

134

Sex. :

 

 

 

«2.x. .

.2 «.2 3.. «.2 «.«« e.« «.« . «s «..«. 3... 2: «... «... «..« «.«« 2
2... « c e v 2 « « v . a 2 .c . 2 « 2 c... «a . .123
e.«« 2.. «.« «.«« «.«« «.2 3. «2 «.2 e2. 3.. 3 e2 «.2 «..« 2
.2... e « . « «. « 2 n c o« 2. . «« m 2 c« R . «5.3.25.3.
«.e e.«« c.«. «.2 «.2. «e «e «.2 «.2 «.«« ca 3. «.2 .2 «..... 2
.e... « .. a. c 2 « m w 2 e« «e . 2 2 «« ..« 2. « deem .5. 8...
o...« ...2 2: «.2 c...« «s «.n e... «.2 «..e e.«. .2 «.2 «.«« «.8 2
8... c n « v c « c .. e. 2 2 . c« 2 «« «c «e « .333.
4.2 «e «.«« 4.2 c.«« c... 3 3.. ...« 4.8 .2 «... v.2 «..« :2 2
««... e « « e c « . . e 2 e« «e . 2 c« v« S. 2. « :63
e.«« 2.2 .2 «.«« «... 3.. e... o.«. «.«« «.«« «.2 «... 6.2 «.«« «..« 2
««... e e « 2 « « 2 z 2 .« 3. « o« c« «« 2. 2 « «5...?
«.«« «.6 «.e 4.2 «.«« «... «.4 3.. ...« 4.2 «.e« «.2 «.2 3. c.«« 2
3.... o. « « e 2 « . . e 2 e« «e « 2. «« «« «« a « «5.6262
«... e.«« «.«« e.«« «.2 .2 «.2 «..« «.2 ....« e2 6.... «..« 4.2 e.«« 2
3.... « e 2 « a. « .« c« «« c« ..« « «« 2 «« «« «n « «p.32
«.«« «e «.e 4.2 °.c« «.2 .2 «.2 ...« «..« «..« «.2 c2 «.2 «.«« 2
«..« 2 « « e c « c. .« c« e« m« « «.. o« «« «« 2. « 83.65
4.2. ..2 «... ..2 2.2 «.«e «.c o.«. «.2 «.2 «..v «..« e2 «... .2 2
:22 2 n « « e v 8 2 2 c« 2 v z. «« m« .« 2 v 868m

.6 e.« v-« «-. o a: .6 e.« r« «-. o .02 +e ea c-« «.. c e:

.._ catamaran he zoeoeroi newsman?!— ue hoses—.8"— :eweﬁomtae we 20:933....—
a«u5 52.25 G215 5...... 82".: 59:2.

 

 

 

.839»
«— .2... 63:38.. .339... 658% E «b... m .23 a... 5 5.33322— ..e «39:230.... .2... «5:12: 9583.35— .mn 93:.

135

8.9....
8.9...

.2328 :28 .8 «238:5 «8.: 532.322. 2.. 323 :03» a. 9.2m .2. 5:25.22. .3 .\e 2:.

 

 

 

 

 

 

e.« «... «... «.2 ..«e e.« 3 o... «.2 ..«« 2
«..... 2 c 2 «« .2 . c « 2 .« «e . ..eeccm
«.2 «.« ed 92 ..«« «.2 «... ..«« .2 «.2 2
9.... «« 2 c. ..« 2 . . 2 m c 2 ..« . «5.3.22.3.
e... o... v... «.2 «.«« «c o... 6.2 «.«« «.3 2
««... 2 .« ..« v« «.. . m 2 2 «« «m « U.ecm .65. .8...
2.2 e.« «.2 e2 e2. «... c... o... «.2 e2. 2
«..... «« 2 e« «.« .v « 2 2 2 «« «n « ..ee.s.=m
«.2 «..« «.2 «.«« «...V e.« «.2 «.2 e...« 4.2. 2
.2... e« 2 «« «... 2. « m 2 2 m« an « :25
«.2 Z: «.2 o.«« «..« e2 «.2 m2 «.«« ..«« 2
2... «« «« 2« mm 8 « 2 2 o« «« e« « «53......»
«.«« «.2 «.2 «.2 o.e« c.«« e.«. c... 2.2 «.v« 2
a... o« o« «« «« e« « «« 2 2 c« :. « «5.9.62
e2 «.2 «.«« ..«. v.2 «..« e2 e2 «..« ..«« 2
««... «« ..« o« e« «e « v« 2 «« v« «« « «E26.
«.«« «.2 «.2 «.2 o.o« e2 «..« e2 «.2 c.«« 2
.«.c 2. 2 «« .e «e « «« ..« 2 «« «« « 36.6
«.2 «.2 «.2 «.2 o... «... «.2 ..2 e2 e2 2
..«.. ea «« «« o« .« v 2. 2 2 2 2 « 38%
.6 e-« v-« «-. o 3,. +6 «..« .V-« «-. o e:
m :o_.m&o_..ma ..o 22.2.09...“ «8:223:2— uo 22.03.05
2.21.5 222...... 2.1529633
2.2:...» .— eea .8532. .33.»...—
9589. 5 «use b .2:— o... 5 333.32.:— ue Ave—.358 «one... «03:33...— E... «5:12.. ocﬁonummm in 03:.

136

adv...
ncva—o

559.8 :80 .8 33:5... 8.! 203.225.. 05 30.3 :02» a. 9.2» .2. 5:22.222. ..c .x. 2:.

 

 

 

 

 

..v ..N I. ..2 N..... ..2 e. .... ...N ..o. ... 5. ...v. ..2 o... 2
a... N . . e c. . 2 2 . 2 2. v... . . e N. v. .. . =22...
N2 m... ..N N2 ..... ...2 .... N... ..2 2.. ..2 ..N 2. 2.. 2.. 2
3.... .. . . m ..N . ..N N. N. .. 8. . a N .. 2 .. . .....s.............
..N N... 3.. e... ...S 2.. ...N. ...N. u... N... «.. o... ...N. ...N ...2. 2
..N... . N o . .N . 2 NN NN .. .N N . 2 .. 2 N.. N .02.. ..5. 8...
v... w... .2 a... N... 2. q... ...N. N...N w... ..2 m. o... a... .2. 2
N..... . n .. N .N . .N ..N NN N. 2 N 2 . 2 .. 2. N =33”...
.2 .2 .3 a... N... ...N. N... ..z ...NN ..z. .... as ...2 ...N a... 2
.2. v v . 2 a. N NN 2 .N N.. 2. N .. e 2 .N o. N ..26
...N w... o... N..: 2.. ..2 2.... a... NNN ...N ..2 .2 ...N. ...N 2... 2
2... N . .. a N. N v. ..N .N ... .. N c... ...2 c. . ...N ...N N .53..
...NN ..NN ..N. ..N N... ...N N.. v... ..2 2.. .2 1.. .2 ...N 2... 2
.3. o. 2 .. . 2 . N. N. .N ..n S N 2 2 a ..N .N N 220......
N2 N2 2... 2. 3.. ..2 5.. ...N o... ...N .2 o... ...N a... ...N 2
NN... .. m 2 v 2 n v. NN a. o. a. . ...2 ...2 ...NN 3.. ..N . 2.2....
N2 .... v... N2 ...2 ...N ..2 ...N. ...N. ...N ..2 ...N. ...z «...N ...N 2
2... N .. . .. 2 N N.. a. NN 2 N. . ...z .2 ...N. ...N ...N N 30.6
2... v... ...N N.. .2 2.... o... ..N. 2.. ...N a... .2 ...N. .2 .2 2
a... 2 . .. n .. e x. 2 .N ..N 2 v ...2. ...2 a... ...2 ...2 c .88.
$ 2. I. N.. .. a. $ 2. v. N-. .. ..2 _ .6 e. v... N.. .. ..2
.— :o_...&ou.aq ..o .8553...— co..u&o.a.u._ ..o anagram comiﬁowim mo mucus—.2..—
.Su... 22:55 .3. T... .25.. 2.80.252 ...N»... 333058....
32.2.2. .33....— quu... a.
9.2. b .2... a... E 5:25.22— ..o 32.358 82.... 8.95:5...— E... 2.2—.0... 953...-..58. $535.2 ﬁn 033—.

137

Research Question One

To answer the question: “Are there signiﬁcant SES, race and nutritional status
differences in the growth status of 10 year old South African boys?”, MANOVA and
MANCOVA were performed. Race, SES and nutritional status were entered as the
independent variables and growth (stature and mass), proportions (sitting height/stature
and bicristal/biacromial ratio), and body composition (2 5 skinfolds; trunk/extremity
skinfold ratio; estimated arm musculature) as the dependent variables. The effects of
SES, race and nutritional status were explored independently, while the effects of the
other two variables (SES and nutritional status) were controlled statistically. SES and
race were entered as dummy variables (0-1, and 0-1-2 respectively), and nutritional status
was entered as a categorical variable (1-2-3). Interactions of SES, race and nutritional
status were also examined. Speciﬁc between group differences for all signiﬁcant
MANOVAs were tested using the pairwise comparisons of estimated marginal means
when SES, race and/or nutritional status were used as covariates. Marginal means are
deﬁned as the unconditional distribution of a single variable or a combination of
variables in a multivariate distribution without regard to the values of the independent
variables (Marriott, 1990; Kotz and Johnson, 1985).

Regression analyses of SES, race and nutritional status on growth status variables
were conducted to determine how much of the variation in growth status could be
attributed to SES, race or nutritional status (Table 38). Some of the variation in growth
(stature and mass) is explained by SES and nutritional status differences. Race, however,

does not contribute signiﬁcantly to the variation in stature or mass. SES accounts for a

138

small proportion (4%) of the variation in stature, and nutritional status differences
account for an additional 3% of the variation in stature. Nutritional status, on the other
hand, accounts for a large proportion (58%) of the variation in mass, with SES
accounting for an additional 2% of the variation. However, only a small proportion (3%)
of the difference in the sitting height/stature ratio is accounted for by nutritional status,
while race explains another 2% of the variation. Approximately 4% of the variation in
the bicristal/biacromial ratio is due to race, and SES differences account for an additional
1% of the variation. Nutritional status differences account for 49% of the variation in
total fatness (sum of 5 skinfolds), 15% of the variation in the TER, and 5% of the
variation in estimated AMA. Race accounts for an additional 2% of the variation in total
fatness, and 1% of the difference in estimated AMA.

MANCOVA analyses were used to explore the effects of SES, race and
nutritional status independently on growth (Table 39). Race differences in growth
(stature and mass) are not signiﬁcant when the effects of nutritional status and SES are
controlled statistically. When the effects of nutritional status and race are controlled
statistically, higher SES boys are taller and heavier than lower SES boys (Table 40).
Overweight boys are heavier than undernourished and nutritionally normal boys, and
nutritionally normal boys are heavier than undernourished boys, when SES and race are
used as covariates. There are no signiﬁcant SES, race, and/or nutritional status
interactions.

There are signiﬁcant race differences in proportions. European boys have

proportionally shorter legs relative to stature than African or Indian boys when SES and

139

Table 38. Regression analysis of selected growth status variables by SES, race and
nutritional status for total sample (n=380).

 

 

 

Dependent Variable Independent Variable R2 B-value' [3b F
Growth
Stature (cm) SES 0.04 2.6 0.19 14.04"
SES + 0.07 2.36 0.17 13.06"
Nutritional status 1.85 0.17
Mass (kg) Nutritional status 0.58 8.25 7.63 527.60“
Nutritional status + 0.60 8.12 7.51 280.70"
SES 1.73 0.13
Proportions
Sitting height/stature ratio, % Nutritional status 0.03 0.58 0.18 13.16”
Nutritional status + 0.05 0.62 0.19 9.84"
Race 0.36 0.13
Bicristal/biacromial ratio, % Race 0.04 1.50 0.20 15.10”
Race + 0.05 1.83 0.24 10.76”
SES -l.13 -0.13
Body Composition
2 of 5 SF (mm) Nutritional status 0.49 31.27 0.70 360.85
Nutritional status + 0.51 31.75 0.71 193.89“
Race 5.52 0.14
TER (mm/mm) Nutritional status 0.15 0.16 0.39 68.47
AMA (cm) Nutritional status 0.05 2.90 0.22 18.34”
Nutritional status + 0.06 3.02 0.22 12.17“
Race 1.50 0.12
'regression coefﬁcient b standardized regression coefﬁcient
' p<0.05 ” p<0.01

Table 39. Summary table of MAN OVA and MANCOVA F-values for SES, race and
nutritional status effects on growth status.

 

Covariate
SES Race Race
NS NS SES
Independent Variable Race SES NS Race SES NS SES ‘ SES ’ Race SES ‘
Race NS " NS Race ‘

 

 

NS

Growth

Stature (cm) 1.79 4.69‘ 2.86 0.98 8.75” 7.61 1.22 1.21 2.08 0.05

Mass (kg) 0.36 8.55“180.87" 1.09 10.03”352.63” 0.61 0.29 1.49 0.03
Proportions

SH/ST, % 5.33“ 1.67 5.47" 10.18” 3.14 9.42” 0.26 0.05 0.39 0.31

BBR, % 9.35” 6.61” 0.81 11.90" 7.21” 1.98 1.04 0.39 1.14 0.61
Body Composition

2 of 6 SF (mm) 7.92" 2.22 204.99” 17.65“ 2.64 372.57” 0.04 0.06 1.14 0.53

TER(rnrn/mm) 7.94” 1.99 22.14“ 14.75” 0.02 52.72” 0.01 2.36 0.97 0.16

AMA (cm) 1.60 1.57 9.65" 2.90 2.12 12.59” 2.84 1.78 2.09 4.75“
‘ p<0.05 " P4101

140

Table 40. Pairwise comparisons of estimated marginal means for selected growth
status variables with signiﬁcant F ratios of nutritional status, race and SES groups
with nutritional status, race and SES as covariates.

 

 

 

 

 

 

 

NS Mean' NS Mean“ MD"
Mass, kg UN 29.1 NN 32.2 -5.19*"‘
UN 29.1 OW 44.6 47.58"
NN 32.2 0W 44.6 42.39"
Sitting height/stature ratio, % UN 49.7 NN 50.6 -O.92**
UN 49.7 OW 50.9 -l.l9"
NN 50.6 OW 50.9 -0.28
2 of 5 Skinfolds, mm UN 30.9 NN 43.0 42.03"
UN 30.9 0W 101.7 -70.79**
NN 43.0 OW 101.7 -58.77"
Tnmk/Extremity Ratio, % UN 0.97 NN 1.03 -0.05**
UN 0.97 OW 1.40 -0.38**
NN 1.03 OW 1.40 -0.33“
AMA, cm UN 28.8 NN 29.9 -1.05
UN 28.8 OW 35.4 655"
NN 29.9 OW 35.4 -5.50**
SES Mean' SES Mean‘ MD”
Stature, cm Lower 137.1 Higher 139.2 -2.11*"‘
Mass, kg Lower 31.7 Higher 33.2 4.53"
Bicristal/biacromial ratio, % Lower 70.1 Higher 68.6 1.53“
Race Mean‘ Race Mean‘ MDb
Sitting height/stature ratio, % Afriean 50.3 Indian 50.2 0.04
Afriean 50.3 European 51.7 -l.4l"""
Indian 50.2 European 51.7 4.45"
Bicristal/biacromial ratio, % African 68.2 Indian 69.5 -1.37"
African 68.2 European 72.5 4.34"
Indian 69.5 European 72.5 4.98"
2 of 5 skinfolds, mm African 43.7 Indian 56.0 42.26"
African 43.7 European 45.4 -1.89
Indian 56.0 Emopean 45.4 10.36"
Tnmk/extremity ratio, % African 1.03 Indian 1.15 -0.12"
African 1.03 European 0.98 0.05
Indian 1.15 European 0.98 0.17"
AMA, cm African - 29.5 Indian 31.7 -2.l9*
African 29.5 European 30.7 -1.25
Indian 31.7 European 30.7 0.94
' Adjusted means
MD = mean difference
* p<0.05
** p<0.01

141

nutritional status are used as covariates. European boys have proportionally broader hips
relative to the shoulders than African and Indian boys when the effects of nutritional
status and SES are removed statistically. Further, Indian boys have proportionally
broader hips relative to the shoulders compared to African boys.

Nutritional status-related differences in the sitting height/stature ratio are
signiﬁcant when SES and race are used as covariates. Undemourished boys have
proportionally longer legs relative to their stature than nutritionally normal or overweight
boys, when the effects of SES and race are controlled.

SES differences in bicristal/biacromial ratio indicate that higher SES boys have
proportionally broader hips relative to their shoulders compared to lower SES boys when
the effects of race and nutritional status are controlled. There are no signiﬁcant
interaction effects of SES, race and/or nutritional status on the sitting height/stature and
bicristal/biacromial ratios, and there are no signiﬁcant SES, race and/or nutritional status
interactions for proportions. i

There are signiﬁcant race differences in total skinfolds and TER when SES and
nutritional status are used as covariates. Indian boys are fatter and have proportionally
more fat on the trunk compared to the extremities than African and European boys when
SES and nutritional status are entered as covariates. Additionally, Indian boys have
larger estimated arm muscle areas compared to African boys, which are related to the
larger accumulation of subcutaneous adipose tissue. When the effects of SES and race
are controlled statistically, nutritional status differences in total fatness, TER and

estimated AMA are signiﬁcant.

142

Overweight boys have larger accumulations of estimated SAT on the trunk
relative to the extremities than undernourished and nutritionally normal boys, and
nutritionally normal boys more than undernourished boys, when the effects of SES and
race are removed statistically. Overweight boys have larger estimated arm muscle areas
than undernourished and nutritionally normal boys when race and SES are used as
covariates. Nutritionally normal and overweight boys do not differ signiﬁcantly.

There is a signiﬁcant SES, race and nutritional status interaction for estimated
arm muscle area (Figure 2). Nutritionally normal Indian boys from lower SES families
have smaller estimated arm muscle areas compared to both undernourished and
overweight lower SES Indian boys. On the other hand, both higher and lower SES
African boys and higher SES Indian boys demonstrate a linear increase in estimated
AMA with improved nutritional status. Additionally, overweight African boys from
higher SES families have a markedly higher estimated AMA compared to nutritionally

normal African boys from higher SES families.

Research Question Two

To answer the question: “Are there signiﬁcant race, SES and nutritional status
differences in the health- and performance-related ﬁtness of 10 year old South African
boys?”, MANCOVAs were conducted, and speciﬁc between group differences for all
signiﬁcant MANCOVAs were tested using pairwise comparisons of estimated marginal
means. The estimated marginal means procedure uses the predicted means without

correcting for covariates, thus allowing for pairwise comparisons between groups.

143

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

Lower SES boys
40 4
an 36 «~
:
Q
E .African
3 32 “P ulndian
«a
3
,3
< 23 a-
24 -_- ., . .__.
m m ow
Nutritional Status
Higher SES boys
40 v
a 36 ~~
=
2
[African
'0 32 _.
.3 ulndian
3
%
< 23 ..
24 - . r —+
NN
Nutritional Status

 

 

Figure 2. SES adjusted means for nutritional status - race interactions by SES for
estimated arm muscle area.

144

Regression analyses of race, SES and nutritional status on the components of
health-related and performance-related ﬁtness were also conducted. Components of
health-related ﬁtness include sit-ups, push-ups, 9 minute run, sit and reach, and the sum
of 5 skinfolds, while components of performance-related ﬁtness include the dash, shuttle
run, grip strength, and standing long jump. The effects of race, SES and nutritional
status were considered individually, after the effects of the other two independent

variables were removed statistically.

Heglth-glgtﬂ ﬁtness

SES and race do not contribute signiﬁcantly to differences in sit-ups, and
nutritional status and SES do not contribute signiﬁcantly to variation in the push-ups,
distance run and sit and reach (Table 41). Nutritional status accounts for only 1% of the
variation in sit-ups, while race accounts for 3%, 2% and 5% of the variation in push-ups,
distance run, and sit and reach, respectively. Nutritional status accounts for 49% of the
variation in the sum of skinfolds, and race accounts for an additional 2% of the variation
in skinfold thicknesses.

The effects of race on health-related ﬁtness were explored independently with
SES and nutritional status used as covariates (Tables 42 and 43). There are signiﬁcant
race differences in push-ups, the distance run, sit and reach, and the sum of skinfolds
when the effects of SES and nutritional status are controlled statistically. European boys
complete more push-ups in 30 seconds and have greater endurance on the 9 minute run

compared to African and Indian boys, who do not differ. However, African boys have

145

Table 41. Regression analysis of selected physical ﬁtness variables by SES, race and
nutritional status for total sample (n=380).

 

 

 

Dependent Variable Independent variable R2 B-value' [3" F
Health-related ﬁtness
Sit-ups, #/30 sec Nutritional status 0.01 -0.97 -0.10 4.15'
Push-ups. #/30 sec Race 0.03 2.09 0.17 10.99”
9 minute run, m Race 0.02 76.36 0.14 7.16"
Sit and reach, cm Race 0.05 -2.00 -0.23 19.77“
2 of 5 SF (mm) Nutritional status 0.49 31.27 0.70 360.85
Nutritional status + 0.51 31.75 0.71 193.89”
Race 5.52 0.14
Performance-related ﬁtness
Dash, sec Nutritional status 0.01 0.21 0.10 3.95'
Grip, kg Nutritional status 0.11 1.88 0.33 45.14"
Nutritional status + 0.13 1.81 0.32 25.99"
SES 0.88 0.12
Grip, kg/kg Nutritional status 0.22 -0.08 -0.47 100.12
Nutritional status + 0.23 -0.08 -0.48 52.65
Race -0.02 -0.10
'regression coefﬁcient b standardized regression coefﬁcient
‘ p<0.05 ” p<0.01

Table 42. Summary table of MANOVA and MANCOVA F—values for SES, race and
nutritional status effects on physical ﬁtness.

 

Covariate

SES Race Race

NS NS SES
Independent variable Race SES NS Race SES NS Race ‘ Race ‘ SES ’ Race ‘
SES NS NS SES ‘

 

 

NS
Health-related ﬁtness
Sit-ups (t!) 1.80 1.75 3.85' 1.09 1.19 12.02”I 3.39 1.81 0.20 1.34
Push-ups (it) 2.71 0.00 2.64 5.23” 0.09 5.68"' 2.28 0.15 0.00 0.08

9 minute run (m) 0.11 1.46 11.65” 12.44" 0.39 9.17" 1.67 4.40” 0.50 0.12
Sit and reach (cm) 11.56” 1.02 1.66 14.51” 0.92 4.49" 0.82 0.31 0.99 1.94
2 of 5 skinfolds (mm) 7.34” 1.98 197.84" 15.26” 1.54 344.60" 0.02 0.87 0.02 0.48

Performance-related ﬁtness

 

Dash(sec) 3.99 0.15 5.13" 4.74" 0.05 6.53“ 1.59 1.09 0.08 1.72
Grip (kg) 0.90 7.05" 8.00" 2.34 7.77" 19.42" 0.63 0.88 0.56 0.49
Grip (kg/kg) 1.09 0.45 36.96" 3.77. 0.13 60.11” 0.07 0.53 0.25 0.46
Shuttle run (sec) 0.07 0.82 0.69 1.08 0.80 1.15 4.66' 1.02 0.05 0.73
Long jump (cm) 1.55 0.01 3.70. 2.00 0.81 2.28 0.47 3.29" 0.62 0.98
‘p<0.05 "p<0.01

146

Table 43. Pairwise comparisons of estimated marginal means for selected health-
related ﬁtness variables with signiﬁcant F ratios of nutritional status, race and SES
groups with nutritional status, race and SES as covariates.

 

 

 

 

Race Mean' Race Mean' MD"
Health-related m
Push-ups, #/30 sec African 23.6 Indian 25.2 -1.54
African 23.6 European 28.4 4.75“
Indian 25.2 European 28.4 -3.21"'
9 minute run, m Afriean 1610 Indian 1560 50.75
African 1610 European 1887 476.87"
Indian 1560 European 1887 427.62"
Sit and reach, cm Afriean 27.8 Indian 24.2 3.63"
African 27.8 European 25.9 1.91
Indian 24.2 European 25.9 -l.72
Z of 5 skinfolds, mm African 43.8 Indian 55.7 41.91"
Afriean 43.8 European 45.8 -2.04
Indian 55.7 European 45.8 9.87"
Performance-related ﬁtn__;_ess
Dash, sec Afriean 9.4 Indian 9.8 -0.39**
Afriean 9.4 Emopean 9.3 0.06
Indian 9.8 European 9.3 0.45‘I
Grip, kg/kg African 0.61 Indian 0.58 0.03"
African 0.61 European 0.59 0.02
Indian 0.58 Etn'opean 0.59 -0.01
SES Mean' SES Mean‘ rvm"
Performance-related ﬁ_tn§ _
Grip, kg Lower 18.4 Higher 19.5 -1.05"

 

" Adjusted means

MD = mean difference
* p<0.05
** p<0.01

147

Table 43. (continued)

 

 

 

 

NS' Meanb NS Mean‘ MD"

Health-related ﬁtness

Sit-ups, #/30 sec UN 16 NN 17.4 -1.40"I
UN 16 OW 13.4 2.63"
NN 17.4 OW 13.4 4.03"

Push-ups, #/30 sec UN 23.9 NN 25.8 -l.88
UN 23.9 OW 21.8 2.11
NN 25.8 OW 21.8 4.00"

9 minute run, m UN 1595 NN 1676 -80.90
UN 1595 OW 1439 156.30‘
NN 1676 OW 1439 237.21”

Sit and reach, cm UN 24.9 NN 27.0 4.03"
UN 24.9 OW 26.4 -0.43
NN 27.0 OW 26.4 1.60

2 of 5 skinfolds, mm UN 30.9 NN 43.0 42.05“
UN 30.9 OW ' 101.2 40.53“
NN 43.0 OW 101.5 -58.48“

Performance-related ﬁtness

Dash, sec UN 9.5 NN 9.4 0.11
UN 9.5 OW 10.1 -0.55“
NN 9.4 OW 10.1 066"

Grip, kg UN 17.5 NN 19.25 -1.73"
UN 17.5 OW 21.0 -3.52"
NN 19.3 OW 21.0 -1.80"

Grip, kg/kg UN 0.65 NN 0.60 0.05”
UN 0.65 OW 0.47 0.18"
NN 0.60 OW 0.47 0.13”

Standing long jump, cm UN 149.0 NN 153.5 4.58"
UN 149.0 OW 147.9 1.08
NN 153.5 OW 147.9 5.66

'UN: undernourished, NN: nutritionally normal, and OW: overweight.

b Adjusted means

°MD = mean difference

"' p<0.05

” p<0.01

148

greater flexibility of the lower back on the sit and reach compared to Indian boys. There
are no signiﬁcant SES differences in health-related ﬁtness.

Race and SES were used as covariates to explore the independent effects on
nutritional status on health-related ﬁtness. There are signiﬁcant nutritional status
differences in the sit-ups, push-ups, distance run, sit and reach, and total fatness when
SES and race are entered as covariates. Undemourished and nutritionally normal boys
complete more sit-ups in 30 seconds and run farther in nine minutes compared to
overweight boys. Nutritionally normal boys have better performances on the sit-ups and
sit and reach compared to undernourished boys. Additionally, nutritionally normal boys
have greater endurance of the arms and shoulder girdle on the push-ups compared to
overweight boys.

There is a signiﬁcant race by nutritional status interaction for the 9 minute run
(Figure 3). Nutritionally normal Indian boys run a shorter distance in nine minutes
compared to undernourished Indian boys, whereas undernourished African and European
boys have poorer endurance on the distance run compared to nutritionally normal African
and European boys. All other SES, race and nutritional status interactions on health-

related ﬁtness are not signiﬁcant.

Performance-related ﬁtness

SES, race and nutritional status do not contribute signiﬁcantly to the variation in
the shuttle run and standing long jump (Table 41). SES and race also do not contribute
signiﬁcantly to the variation in the dash, while nutritional status accounts for only 1% of

the variation. Nutritional status accounts for 11% of the variation in absolute grip

149

23.2

23.0

22.8

Adjusted means

22.6

22.4
22.2

1700

1500

Adjusted means

1400

1300

Adjusted means
8

Shuttle run: SES X Race

II Race

I - Aﬁim

- _ _ I 1 Indian
Lower Higher

SES

 

 

 

 

 

 

 

9-minute run: Nutritional status X Race

 

 

 

 

 

 

 

 

 

 

 

 

a
Race
- African
l‘ﬂ Indlln
NN
Nutritional status
Standing long jump: Nutritional status X Race
" Race
. - m
‘ r—J India:

 

 

 

 

 

 

 

Figure 3. Adjusted means and interaction effects for the shuttle run (Race X SES), 9
minute run (Nutritional status X Race), and standing long jump (Nutritional status

X Race).

150

strength (kg), and 22% of the variation in grip strength relative to body mass (kg/kg).
SES accounts for an additional 2% of the variation in absolute grip strength (kg), while
race accounts for an additional 1% of the variation in relative grip strength (kg/kg).

The effects of SES, race and nutritional status on performance-related ﬁtness
were explored independently using MANCOVA (Tables 42 and 43). When the effects of
SES and nutritional status are controlled statistically, race-speciﬁc differences in the dash
and grip (kg/kg) are signiﬁcant. African and European boys have greater speed in the
dash compared to Indian boys. Further, African boys have greater grip strength per
kilogram body weight compared to Indian boys when the effects of SES and nutritional
status are statistically removed.

There is a signiﬁcant SES difference in absolute grip strength when race and
nutritional status are used as covariates. Higher SES boys have greater absolute grip
strength compared to lower SES boys. This, however, is an artifact of the larger body
size of higher SES boys, since differences in grip strength relative to kilogram body mass
are not signiﬁcant.

When the effects of SES and race are statistically removed, overweight boys have
poorer performances on the 50 meter dash compared to undernourished and nutritionally
normal boys. Overweight boys have greater absolute grip strength compared to
undernourished and nutritionally normal boys, and nutritionally normal boys have better
absolute grip strength compared to undernourished boys, when race and SES are
controlled statistically. However, undernourished boys have greater grip strength per

kilogram body mass compared to nutritionally normal and overweight boys, and

151

nutritionally normal boys have greater grip strength per kilogram mass compared to
overweight boys.

There are signiﬁcant race by nutritional status interactions for the standing long
jump and SES by race interactions for the shuttle run (Figure 3). Lower SES Indian boys
have better performances on the shuttle run compared to higher SES Indian boys. On the
other hand, lower SES African boys have poorer performances on the shuttle run
compared to higher SES African boys. Nutritionally normal African boys have poorer
performances on the standing long jump compared to undernourished and overweight
African boys. Both nutritionally normal Indian and European boys have better explosive
strength on the standing long jump compared to undernourished and overweight Indian
and European boys respectively. All other interaction effects for performance-related

ﬁtness are not signiﬁcant.

March ngtjon Thm

To answer the question: “Are there SES, race and nutritional status differences in
the habitual physical activity of 10 year old South African boys?” MANOVA was
performed. Habitual physical activity was assessed as the time spent on light activity (3
to 5 METS) and MVPA (6 to 9 METS), and mean estimated total daily energy
expenditure. Stepwise regression analyses were conducted to determine which
sociodemographic factor (SES, race and nutritional status) was the most powerﬁrl
predictor of sedentary, light and moderate to vigorous physical activity (Table 44).

Nutritional status accounts for 46% and 41% of the variation in absolute
(kcal/day) and relative (kcal/kg/day) estimated energy expenditure in sedentary activity,

152

respectively. SES accounts for an additional 2% of the variation in sedentary activity.
Nutritional status accounts for approximately 16% of the variation in estimated energy
expenditure (kcal/day) in light activity, but does not account for any of the variance in
estimated energy expenditure in light activity per kilogram body mass. Nutritional status
explains 2% of variation in estimated energy expenditure in MVPA (kcal/day), and race
contributes another 1%. On the other hand, race accounts for 2% of the variation in

MVPA (kcal/kg/day).

Table 44. Regression analysis of components of the energy expenditure by SES, race
and nutritional status for total sample (n=228).

 

 

 

Dependent variable Independent Variable R2 Beta-value' a” F
Estimated energy expenditure (kcal/day)
Sedentary Nutritional status 0.46 183.50 0.68 277.07"
Nutritional status 0.48 178.82 0.66 148.73"
SES 46.66 0.14 148.73"
Light Nutritional status 0.16 81.68 0.40 60.95"
MVPA Nutritional status 0.02 39.46 0.14 6.17“
Nutritional status 0.03 35.54 0.12 5.25"
Race -30.73 -0.11
Total Nutritional status 0.41 304.64 0.64 221.42"
Estimated energy expendittn'e (keal/kg/day)
Sedentary Race 0.04 -1.65 -0. 19 11.76"
MVPA Race 0.02 -1.07 -0.13 5.49“
'regression coefficient " standardized regression eoemcient
* p<0.05 ** p<0.01

Variation in estimated energy expenditure (kcal/day) in sedentary and light
activity among racial groups is confounded by SES and nutritional status. When the
effects of SES and nutritional status are centrolled statistically, racial variation in
sedentary and light estimated energy expenditure (kcal/day) is not signiﬁcant (Table 45).

There are, however, differences in MVPA and mean daily total estimated energy

153

expenditure (kcal/day) when SES and nutritional status are used as covariates. African

and European boys are more active at higher intensities of estimated energy expenditure

compared to Indian boys (Table 46). The differences remain signiﬁcant when MVPA is

expressed per kilogram body mass. African boys have a higher daily total estimated

energy expenditure (kcal/day) compared to Indian boys, but expressed relative to body

mass, both African and European boys are more active than Indian boys.

Table 45. MANOVA of estimated energy expenditure variables by SES, race and

nutritional status for total sample (n=228).

 

 

 

 

 

 

Covariates
NS and NS and Race
SES Race Ed SES
RACE SES NS Race SES NS RACE ‘ RACE ‘ SES ‘ RACE “
SES NS NS SES ‘
NS
Estimated energy expenditure (kcal/day)
Sedentary 4.12' 7.49" 40.51" 2.34 9.80” 159.21" 0.26 0.10 0.75 0.37
Light 4.88” 0.84 17.99” 2.68 0.07 31.13" 0.60 0.21 0.76 0.19
MVPA 6.00" 0.60 2.06 8.52"I 0.20 2.78 0.05 0.60 0.45 0.28
Total 4.92" 0.11 41.97“ 5.45" 1.40 111.13“ 0.24 0.87 0.64 0.08
Estimated energy expenditure (kcal/kg/day)
Sedentary 3.96' 0.66 1.60 3.81‘ 2.29 0.39 0.01 0.64 1.65 0.11
Light 5.80“ 2.38 3.77‘ 4.18‘ 1.52 1.47 0.20 0.11 1.20 0.24
MVPA 6.92" 1.20 0.09 10.01“ 1.11 2.37 0.34 0.40 0.63 0.31
Total 9.73" 2.85 0.31 14.27” 1.40 1.83 0.06 0.53 3.36‘ 0.62
" p<0.05 ” p<0.01

SES variation in absolute estimated energy expenditure in sedentary (kcal/day)

activity is signiﬁcant when the effects of race and nutritional status are controlled.

Higher SES boys are more sedentary (chday) than lower SES boys. This is an artifact

of body size, since SES differences in estimated energy expenditure in sedentary

(kcal/kg/day) activity are not signiﬁcant when expressed in terms of body mass.

154

Table 46. Pairwise comparisons of estimated marginal means for selected physical
activity variables with signiﬁcant F ratios of nutritional status, race and SES groups
with nutritional status, race and SES as covariates.

 

 

 

Race Mean' Race Mean' MDb

Estimated energy expenditure (kcal/day)

MVPA African 302.2 Indian 222.4 79.88"
Afriean 302.2 European 314.5 -12.31
Indian 222.4 European 314.5 -92.19"

Total African 1285 Indran' 1198.6 86.12“
African 1285 European 1281.3 3.42
Indian 1199 European 1281.3 -82.70

Estimated energy expenditure (keal/kg/day) .

Sedentary African 22.0 Indian 22.3 -0.25
African 22.0 Emopean 21.1 092*
Indian 22.3 European 21.1 1.17"

Light African 8.3 Indran' 7.2 1.05"
African 8.3 European 8.6 -0.33
Indran' 7.2 Emopean 8.6 -l.38*

MVPA Afriean 9.5 Indian 6.8 2.68"
Afriean 9.5 European 9.8 -0.26
Indian 6.8 European 9.8 4.94“

Total Afriean 39.9 Indian 36.4 3.48"
Afriean 39.9 Emopean 39.5 0.33
Indian 36.4 European 39.5 -3.15**
ﬂ m §E_s m _Mn‘-'

Estimated energy expenditure (keel/day) -

Sedentary Lower 690.6 Higher 736.2 45.69“
N_S Mean! N_S Mean-J w

Estimated energy expenditure (kcal/day)

Sedentary UN 594.7 NN 710.3 -1 15.67"
UN 594.7 OW 978.2 683.55“
NN 710.3 OW 978.2 -267.88**

Light UN 200.2 NN 257.0 -56.78"'*
UN 200.2 OW 371.4 -171.21"
NN 257.0 OW 371.4 -1 14.43"

Total UN 1030 NN 1250 419.93"
UN , 1030 OW 1654 624.12"
NN 1250 OW 1654 404.19"

' Adjusted means

MD = mean diﬂ'erence
* p<0.05 “ p<0.01

155

Variation in sedentary, light and total estimated energy expenditure (kcal/day) by
nutritional status are signiﬁcant when the effects of SES and race are removed
statistically. Nutritional status variation is related to body size as differences are not
signiﬁcant when expressed per unit mass.

The interactions of SES and nutritional status in daily total estimated energy
expenditure (kcal/kg/day) are signiﬁcant (Figure 4). Mean daily total estimated energy
expenditure per unit mass (kcal/kg/day) of undernourished lower SES boys is higher than
that of nutritionally normal and overweight lower SES boys. Nutritionally normal lower
SES boys have a higher mean daily total estimated energy expenditure per unit mass
compared to overweight lower SES boys. The mean daily total estimated energy
expenditure per unit mass (kcal/kg/day) of nutritionally normal higher SES boys is
higher than that of undernourished and overweight higher SES boys. Undemourished
higher SES boys have lower mean daily total estimated energy expenditure per unit mass
than overweight higher SES boys. All other SES, race and/or nutritional status

interactions are not signiﬁcant.

Rggrch Question Four
To answer the question: “Are there nutritional status differences in habitual
physical activity of 10 year old South African boys grouped by race?” MANOVA was
performed. The effects of SES were controlled statistically. Estimated energy
expenditure was sorted into sedentary (l - 2 METS), light (3 to 5 METS), MVPA (6 - 9

METS) and total daily energy expenditure. Stepwise regression analyses were conducted

156

 

Total EEE (kcang/day): SES by NS

 

 

 

 

m

C
(D
d)

E I higher
g D Lower
.2.

2

 

 

 

 

 

 

l
j

 

 

   

NN
Nutritional status

 

 

 

Figure 4. Adjusted means for the interaction of SES and nutritional status on daily
total estimated energy expenditure (kcal/kg/day).

to determine if nutritional status was a signiﬁcant predictor of sedentary, light and
moderate to vigorous physical activity.

There is signiﬁcant nutritional status variation in estimated energy expenditure in
sedentary, light and total activity of African boys (Tables 47-49). This, however, is a
factor of body size since there are no differences in estimated energy expenditure relative
to body mass of African boys. In contrast, nutritional status variations in all categories of
absolute estimated energy expenditure (sedentary, light, MVPA and total) is signiﬁcant
for Indian boys, and the differences for MVPA and total estimated energy expenditure of
Indian boys remain signiﬁcant when expressed per unit mass. Undemourished Indian
boys engage in less MVPA (kcal/kg/day) compared to nutritionally normal and
overweight Indian boys. Additionally, overweight Indian boys have higher mean daily
total estimated energy expenditure relative to body mass compared to nutritionally

normal boys. Differences in MVPA and mean daily total estimated energy expenditure

157

between undernourished and nutritionally normal Indian boys are not signiﬁcant. There
are nutritional status differences in absolute estimated energy expenditure in light, total
and habitual physical activity (PAQ-C Score) of the small sample of European boys.
When estimated energy expenditure is expressed relative to body mass, overweight
European boys engage in more light activity and have a higher PAQ-C Score compared to
undernourished and nutritionally normal European boys.

Stepwise regression analyses were conducted to determine how much of the
variation in estimated energy expenditure of the total sample was explained by nutritional
status (Table 50). Nutritional status explains 2% to 44% of the variation in estimated
energy expenditure in the total sample. When estimated energy expenditure is expressed
per unit mass, a smaller proportion of the variation (4% versus 39%) in mean daily total
estimated energy expenditure is explained by nutritional status. On the other hand,
nutritional status explains only 2% of the variation in both absolute and relative estimated

energy expenditure in MVPA.

Table 47. Summary table of race speciﬁc AN COVA F values, controlling for SES,
of nutritional status differences in estimated energy expenditure (kcal/day and

 

kcal/kg/day).
African & European

Estimated energy expenditure (kcal/day)

Sedentary 48.55" 56.78" 1.53

Light 5.96" " 5.77" 6.99"

MVPA 0.50 5.04" 0.87

Total 35.05" 33.93" 4.12"
Estimated energy expenditure (kg/day)

Sedentary 0.21 0.58 0.99

Light 0.74 1.55 3.54"

MVPA 0.18 3.72“ 0.60

Total 0.16 5.51" 0.97
PAQ-C Score 0.29 0.29 6.08"

 

158

Table 48. Pairwise comparisons of estimated marginal means for selected estimated
energy expenditure (kcal/day) and physical activity variables with signiﬁcant F
ratios of nutritional status for African, Indian and European boys.

 

 

 

NS Mean' NS Mean' MD"
Afriean

Sedentary UN 590.7 NN 704.0 -121.47**
UN 590.7 OW 968.1 401.84"
NN 704.0 OW 968.1 -280.37**

Light UN 207.8 NN 269.7 -23.26
UN 207.8 OW 388.0 -124. 14"
NN 269.7 OW 388.0 400.88"

Total UN 113 NN 1271 438.85"
UN 113 OW 1699 -57 1.61"
NN 1271 OW 1699 432.76"

Indian

Sedentary UN 592.7 NN 719. 429.38“
UN 592.7 OW 101.9 440.67“
NN 719.0 OW 101.9 -311.29“

Light UN 185.2 NN 230.1 -64.63"
UN 185.2 OW 339.2 404.20”
NN 230.1 OW 339.2 -39.57

MVPA UN 178.2 NN 236.6 404.30"
UN 178.2 OW 253.2 -47.18
NN 236.6 OW 253.2 57.14

Total UN 956 NN 1186 498.30"
UN 956 OW 1611 692.05"
NN 1186 OW 1611 493.75"

1.3mm

Light UN 261.4 NN 252.4 8.98
UN 261.4 OW 617.1 -355.71"“"
NN 252.4 OW 617.1 664.69"

Total UN 1043 NN 1307.5 -263.46*
UN 1043 OW 1707.4 -663.42‘
NN 1308 OW 1707.4 -399.96

PAQ-C Score UN 2.45 NN 2.80 035
UN 2.45 OW 4.06 -1.60"
NN 2.80 OW 4.06 4.26“

' Adjusted means

bMD = mean diﬁerence

‘ p<0.05

" p<0.01

159

Table 49. Pairwise comparisons of estimated marginal means for selected estimated
energy expenditure (kcal/kg/day) and physical activity variables with signiﬁcant F
ratios of nutritional status for Indian and European boys.

 

 

 

NS Mean‘ NS Mean‘ MD"
Mag
MVPA UN 6.8 NN 7.4 -l.78
UN 6.8 OW 5.6 -3.47"'
NN 7.4 OW 5.6 1.70
Total UN 36.1 NN 37.0 2.47
UN 36.1 OW 35.2 2.55
NN 37.0 OW 35.2 -5.02"
European
Light UN 9.1 NN 7.9 1.24
UN 9.1 OW 16.6 4.46"
7.9 OW 16.6 -8.70"I
; Adjusted means
MD = mean drﬂerence
* p<0.05
" p<0.01

Table 50. Regression analyses of estimated energy expenditure variables by
nutritional status for the total sample.

 

 

Dependent variable Independent variable R2 Beta-value' 13" F
Estimated energy expenditure (keal/day)
Sedentary nutritional status 0.44 189.77 0.66 170.64'MI
Light nutritional status 0.09 56.36 0.31 22.26"
MVPA nutritional status 0.02 38.78 0.13 3.90"I
Total nutritional status 0.39 284.91 0.62 135.04'"

Estimated energy expenditure (kcal/kg/day)

 

MVPA nutritional status 0.02 -l.07 -0.13 5.49"
Total nutritional status 0.04 -l.65 -0. I9 11.76“I
'regression coefficient b standardized regression coefﬁcient
a P<0.05 "‘ p<0.01

160

Research Questign Five

To answer the question “Are there any race-speciﬁc differences in the growth
status and physical ﬁtness of 10 year old South African boys with different levels of
energy expenditure?”, the growth status and physical ﬁtness of boys in the lowest and
highest quartiles of 3-day mean MVPA and mean daily total estimated energy
expenditure (kcal/kg/day) were compared with MAN OVA and MANCOVA. Pairwise
comparisons of estimated marginal means and Scheffe post hoc comparisons of
signiﬁcant MANOVAs and MANCOVAs were used. SES was used as a covariate.

The racial composition of the lowest and highest quartiles of MVPA and mean
daily total estimated energy expenditure differ signiﬁcantly (Chi-square statistic of x2 =
18.3 for 2 df, p < 0.01; x2 = 17.95 for 2 (if, p < 0.01, respectively) (Tables 51 and 52).
There are signiﬁcantly more African boys in the highest quartiles of estimated energy
expenditure than Indian boys. Conversely, there are signiﬁcantly more Indian boys in
the lowest quartiles of MVPA and mean daily total estimated energy expenditure. Thus,
some of the growth and physical ﬁtness differences between boys in the highest and
lowest quartiles of estimated energy expenditure may reﬂect racial variation.

Race-speciﬁc differences in growth and physical ﬁtness of boys in the lowest and
highest quartiles of mean daily total estimated energy expenditure and MVPA were
compared using MANOVA (Tables 53 to 58). African boys with the lowest overall
habitual physical activity are taller, heavier, and have a higher BMI than the most active
African boys. Similarly, Indian boys in the lowest quartile are also taller and heavier
than Indian boys in the lowest quartile of mean daily total estimated energy expenditure.
On the other hand, Indian boys in the lowest quartile have a smaller sitting height/stature

161

Table 51. Racial distribution, mean and standard deviations of estimated energy
expenditure of boys in the highest and lowest categories of MVPA and mean daily
total estimated energy expenditure (kcal/kg/day) for the total sample by race.

 

African Indian Europgn Total
M SD M SD M SD M SD

 

Total estimated energy exgnditure (kcal/kg/day)

Lowest quartile

N 33 43 6 82

Energy expenditure 31.8 1.9 30.6 2.5 30.2 3.0 31.1 2.4
Highest quartile

N 57 17 8 82

Energy expenditure 46.2 3.1 44.2 1.8 47.3 3.0 45.9 3.0
MVPA (kcal/kg/day)
Lowest Quartile

N 33 45 5 83

Energy expenditure 2.4 1.0 2.5 1.0 1.6 1.6 2.4 1.0
Highest

N 57 18 7 82

Energy expenditure 16.2 3.3 14.9 2.2 18.0 4.2 16.0 3.3

 

Table 52. Chi square statistics for race differences for boys in the highest and
lowest categories of MVPA and mean daily total estimated energy expenditure

 

 

(kcal/kg/day).
Total MVPA
Highest quartile 49.78" 50.50"
Lowest quartile 26.81" 30.46"

 

" p<0.01

ratio indicating proportionally longer legs relative to stature than Indian boys in the
highest quartile of mean daily total estimated energy expenditure. All other differences
in growth status of African and Indian boys at the extremes of mean daily total estimated
energy expenditure are not signiﬁcant. There are no signiﬁcant differences in the growth
status or physical ﬁtness of European boys in the highest and lowest quartiles of mean
daily total estimated energy expenditure.

162

African boys in the lowest quartile of mean daily total estimated energy
expenditure (kcal/kg/day) have greater absolute grip strength compared to African boys
in the highest quartile. This, however, is a function of their larger body since grip
strength per unit mass does not differ between boys at the extremes of mean daily total
estimated energy expenditure (kcal/kg/day). All other growth and physical ﬁtness
differences between African boys in the highest and lowest quartiles of mean daily total
estimated energy expenditure do not differ. There are no signiﬁcant differences in the
physical ﬁtness of Indian boys at the extremes of mean daily total estimated energy
expenditure.

There is a difference in the relationship between absolute (kcal/day) and relative
estimated energy expenditure (kcal/kg/day) and growth and physical ﬁtness (Tables 56 to
58). African boys in the lowest quartile of MVPA are heavier than African boys who are
most active at high levels of estimated energy expenditure. European boys in the lowest
quartile of MVPA are taller than European boys in the highest quartile of MVPA. All
other differences in the growth status of African and European boys are not signiﬁcant.
There are no signiﬁcant differences in the growth status of Indian boys at the extremes of
MVPA.

African boys who are most active at high intensities complete more sit-ups in 30
seconds than to boys in the lowest quartile of MVPA. All other physical ﬁtness
differences between African boys at the extremes of MVPA are not signiﬁcant. The
physical ﬁtness of Indian and European boys in the highest and lowest quartiles of

MVPA do not differ.

163

Table 53. Means, standard deviations and F ratios for the growth status and
physical ﬁtness of African boys grouped into the highest and lowest categories of

mean total daily estimated energy expenditure.

 

 

 

 

Lowest Quartile Highest quartile
Mean SD Md Mean SD Md F

Growth

Stature, cm 137.3 8.2 138.5 137.0 6.0 137.1 213*

Mass, kg 33.8 8.3 32.1 31.7 5.4 31.0 17.31"

BMI, ltg/rn2 17.8 3.5 16.7 16.8 2.3 16.7 15.91"
Proportions

SH/ST, % 50.6 1.8 50.5 50.5 1.8 50.7 1.33

BBIL % 69.2 4.9 70.2 69.3 4.8 69.2 2.52
Body composition

2 of 5 SF, mm 48.0 29.2 38.3 42.4 25.7 33.7 0.00

TER, mm/mm 1.06 0.26 1.08 1.04 0.27 1.04 0.24

AMA, cm 30.5 9.5 27.3 29.2 7.9 27.7 0.04
Health-related ﬁtness

Sit-ups, #30 sec 16.7 4.2 17.0 15.9 5.2 16.0 0.33

Pushups, #30 sec 23.2 6.2 23.0 24.6 8.8 23.0 0.01

9 minute mu, m 1604.9 392.4 1675.0 1544.5 335.4 1540.0 0.21

Sit and reach, cm 27.0 6.4 26.8 27.8 5.5 28.0 1.59
Performance-related fitness

Dash, sec 9.5 1.2 9.2 9.3 1.1 9.0 0.24

Grip, kg 19.6 4.0 19.0 18.9 3.8 19.0 7.78“

Grip, kg/kg 0.6 0.1 0.6 0.6 0.1 0.6 0.30

Shuttle run, sec 22.4 2.3 21.8 23.3 2.7 22.9 0.81

Standing long jump, cm 150.3 22.2 149.5 149.1 24.1 143.0 0.62
PAQ-C Score 2.79 0.52 2.72 2.73 0.37 2.67 0.21
" p<0.05 " p<0.01

164

Table 54. Means, standard deviations and F ratios for growth status and physical
ﬁtness of Indian boys grouped into the highest and lowest categories of mean total
daily estimated energy expenditure.

 

 

 

 

Lowest Quartile Highest quartile
Mean SD Md Mean SD Md F

Growth

Stature, cm 140.2 6.0 139.6 137.5 8.4 139.6 13.76"

Mass, kg 33.2 9.1 30.1 31.6 7.0 32.4 4.55""

BMI, kg/m2 16.4 3.0 15.3 16.6 3.0 16.2 3.81
Proportions

SH/ST, % 49.7 2.0 50.2 50.1 2.7 50.3 555*

BBR, % 68.7 6.3 68.8 70.4 4.5 70.5 0.07
Body composition

2 of 5 SF, mm 54.0 33.1 37.6 51.0 23.4 48.0 1.73

TER, mm/mm 1.13 0.31 1.13 1.15 0.30 1.08 0.64

AMA, cm 31.5 8.5 29.7 33.4 8.5 31.4 0.60
Health-related ﬁtness

Sit-ups, #/30 sec 14.5 4.9 16.0 16.9 5.3 17.0 0.36

Pushups, #/30 sec 22.8 7.0 21.5 23.6 10.3 22.0 0.91

9 minute run, m 1521.3 306.8 1524.0 1465.8 394.9 1420.0 0.12

Sit and reach, cm 23.9 6.0 . 24.5 23.6 6.3 24.0 2.75
Performance-related ﬁtness

Dash, sec 10.1 1.3 9.7 9.5 0.6 9.5 3.02

Grip, kg 18.5 3.4 18.0 18.5 3.6 20.0 0.15

Grip, kg/kg 0.6 0.1 0.6 0.6 0.1 0.6 1.19

Shuttle rlm, sec 23.0 2.8 22.7 24.8 3.8 24.1 0.17

Standing long jump, cm 149.8 26.4 148.0 147.3 19.0 150.0 0.02
PAQ-C Score 3.0 0.7 2.8 2.7 0.5 2.8 2.52
" p<0.05 " p<0.01

165

Table 55. Means, standard deviations and F ratios for growth status and physical
ﬁtness of European boys grouped into the highest and lowest categories of mean
total daily estimated energy expenditure.

 

 

 

 

Lowest Quartile Highest quartile
Mean SD Md Mean SD Md F

Growth

Stature, cm 144.6 6.3 144.6 136.3 3.4 136.6 2.61

Mass, kg 34.9 5.3 33.7 32.3 3.6 32.4 0.79

BMI, ltg/rn2 16.6 1.4 16.2 17.4 1.3 17.3 0.01
Proportions

SH/ST, % 51.3 0.9 51.4 51.6 0.9 51.9 0.05

BBR, % 72.7 3.5 73.7 72.8 4.1 71.6 0.06
Body composition

2‘. of 5 SF, mm 40.0 13.3 35.5 49.2 24.5 37.3 0.03

TER, mm/mm 1.00 0.15 1.00 1.07 0.28. 1.01 0.06 ~

AMA, cm 30.9 6.6 32.2 33.1 7.8 31.4 0.03
Health-related ﬁtness

Sit-ups, #l30 sec 16.8 5.5 14.5 20.0 3.2 20.0 0.93

Pushups, #/30 sec 23.7 5.8 22.5 31.0 7.8 30.0 0.37

9 minute run, m 1682.5 238.0 1580.0 1788.1 306.4 1818.0 1.26

Sit and reach, cm 25.5 4.4 24.5 25.0 4.1 23.5 0.26
Performance-related ﬁtness

Dash, sec 9.4 0.7 9.3 9.2 0.7 9.4 0.21

Grip, kg 21.6 2.5 20.5 18.8 1.6 20.0 1.69

Grip, kg/kg 0.6 0.1 0.7 0.6 0.1 0.6 0.02

Shuttle run, sec 23.9 2.0 24.1 22.4 2.1 21.9 2.36

Standing long jump, cm 159.0 19.5 158.0 153.4 21.2 157.0 2.63
PAQ-C Score 2.6 0.3 2.6 2.8 0.7 2.6 0.29
"‘ p<0.05 " p<0.01

166

Table 56. Means, standard deviations and F ratios for growth status and physical
ﬁtness of African boys grouped into the highest and lowest categories of estimated
energ expenditure in MVPA (kcal/kg/day).

 

 

 

 

Lowest Quartile Highest quartile
Mean SD Md Mean SD Md F

Growth

Stature, cm 139.3 6.9 138.9 137.3 6.3 139.6 2.62

Mass, kg 35.6 7.5 34.4 31.8 5.8 31.0 4.55“

BMI, lrg/rn2 18.2 3.1 17.3 16.8 2.5 16.7 2.56
Proportions

SH/ST, % 50.4 1.7 50.5 50.5 2.0 50.6 0.46

BBR, % 68.6 5.3 68.9 69.2 4.7 69.2 0.13
Body composition

2 of5 SF, mm 55.0 28.5 45.3 41.1 24.5 33.7 3.84

TER, mm/mm 1.10 0.25 1.13 1.04 0.27 1.02 1.90

AMA, cm 32.6 10.7 29.5 29.2 7.1 28.9 1.66
Health-related ﬁtness

Sit-ups, #l30 sec 15.8 4.0 15.5 16.0 5.3 16.0 5.14"

Pushups, #/30 sec 22.8 9.0 22.5 24.5 9.1 23.0 0.04

9 minute run, at 1600.8 366.4 1700.0 1546.6 321.4 1550.0 0.12

Sit and reach, cm 25.3 6.7 25.0 28.1 5.1 28.0 1.17
Performance-related ﬁtness

Dash, sec 9.7 1.4 9.5 9.3 1.0 9.0 0.96

Grip, kg 19.8 4.4 19.0 18.8 3.6 19.0 2.08

Grip, kg/kg 0.6 0.1 0.6 0.6 0.1 0.6 0.19

Shuttle run, sec 22.3 2.4 21.9 23.4 2.7 22.9 1.73

Long jump, cm 147.9 23.0 150.0 148.9 23.3 143.0 0.26
PAQ-C Score 2.8 0.5 2.7 2.8 0.4 2.7 0.20
* p<0.05 . "- p<0.01

167

Table 57. Means, standard deviations and F ratios for growth status and physical
ﬁtness of Indian boys grouped into the highest and lowest categories of estimated
energy expenditure in MVPA (kcal/kg/day).

 

 

 

 

Lowest Quartile Highest quartile
Mean SD Md Mean SD Md F

Growth

Stature, cm 141.0 6.5 141.3 138.8 8.9 141.9 2.53

Mass, kg 34.9 9.2 31.7 32.5 6.8 32.9 1.09

BMI, kg/m2 17.1 3.3 16.0 16.8 2.9 15.8 0.16
Proportions

SH/ST, % 49.8 1.9 50.2 50.8 2.5 50.6 2.43

BBK % 69.2 6.9 69.0 70.2 5.8 70.7 0.25
Body composition

2 of 5 SF, mm 60.3 36.6 42.5 49.1 22.8 46.3 0.53

TER, mm/mm 1.19 0.32 1.17 1.09 0.26 . 1.07 2.39

AMA, cm 31.3 8.2 29.8 34.0 12.8 28.5 0.01
Health-related ﬁtness

Sit-ups, #/30 sec 14.7 5.4 15.0 18.2 4.8 17.5 3.55

Pushups, #l30 sec 23.0 7.6 23.0 27.3 9.6 29.5 0.65

9 minute run, at 1565.4 287.8 1564.0 1514.3 427.5 1460.0 0.03

Sit and reach, cm 22.9 5.9 23.0 24.0 5.6 24.8 0.02
Performance-related ﬁtness

Dash, sec 10.3 1.4 10.0 9.7 1.3 9.5 0.04

Grip, kg 19.1 3.1 19.0 18.8 3.3 20.0 0.22

Grip, kg/kg 0.6 0.1 0.6 0.6 0.1 0.6 0.41

Shuttle run, sec 23.0 2.5 22.8 24.5 3.8 23.5 0.70

Standing long jump, cm 149.7 23.1 150.0 152.4 21.5 160.0 0.19
PAQ-C Score 3.0 0.7 2.8 2.5 0.6 2.4 3.36
* p<0.05 " p<0.01

168

Table 58. Means, standard deviations and F ratios for growth status and physical
ﬁtness of European boys grouped into the highest and lowest categories of estimated
energy expenditure in MVPA (kcal/kg/day).

 

 

 

 

Lowest Quartile Highest quartile
Mean SD Md Mean SD Md F

Growth

Stature, cm 144.9 7.9 144.8 136.1 3.6 136.1 695*

Mass, kg 34.9 5.8 31.0 31.5 3.3 32.1 1.62

BMI, ltg/rn2 16.6 2.0 17.5 17.0 1.6 16.5 0.18
Proportions

SH/ST, % 51.0 1.8 51.4 51.6 1.0 51.9 0.61

BBR, % 72.1 3.8 72.7 72.9 4.4 70.8 0.11
Body composition

2 of5 SF, mm 39.7 15.8 37.5 45.5 24.1 37.0 0.23

TER, mm/mm 1.10 0.19 1.02 1.02 0.27 0.95 0.33

AMA, cm 29.8 7.3 27.8 31.5 6.9 29.0 0.17
Health-related fitness

Sit-ups, #/30 sec 16.2 6.5 14.0 19.4 2.9 20.0 1.37

Pushups, #/30 sec 22.6 7.1 19.0 30.4 8.2 26.0 2.95

9 minute run, 111 1691.0 237.7 1610.0 1828.6 307.0 1890.0 0.70

Sit and reach, cm 27.0 3.1 27.0 25.6 4.0 24.0 0.44
Performance-related ﬁtness

Dash, sec 9.3 0.5 9.1 9.2 0.7 9.5 0.03

Grip, kg 19.8 2.3 20.0 18.6 1.7 20.0 1.02

Grip, kg/kg 0.6 0.1 0.6 0.6 0.1 0.6 0.16

Shuttle run, sec 24.3 2.6 22.8 22.5 2.2 22.1 1.63

Standing long jump, cm 150.8 12.1 151.0 151.3 22.0 148.0 0.00
PAQ-C Score 2.5 0.3 2.5 2.7 0.6 2.6 0.19
"' p<0.05 “ p<0.01

169

Chapter 5

DISCUSSION

This study examined the relationship of SES, race and nutritional status with
growth (size, proportion and body composition), physical ﬁtness and habitual physical
activity of 10 year old South African boys. The growth, physical ﬁtness and habitual
physical activity of 10 year old boys are ﬁrst summarized. The discussion then focuses
on growth, physical ﬁtness and habitual physical activity with reference to SES, ethnicity

and nutritional status.

Growth

Growth is addressed in terms of growth status (stature, mass ), fatness (skinfolds),
muscularity, and secular trends in South African populations. Variability in growth
status in infancy and childhood is greater within than between races (F rongillo and
Hanson, 1995), and key confounders in population variation are SES and nutritional
status. This trend is evident in the present study. There are no signiﬁcant race
differences in growth status (stature and mass); SES accounts for 4% and nutritional
status for an additional 3% of the variation in stature. On the other hand, nutritional
status accounts for 58% of the variation in mass, with SES explaining an additional 2%.

In general, higher SES boys within a racial group are taller and heavier compared to

170

lower SES boys. Overweight boys are also taller and heavier than undernourished and
nutritionally normal boys, and nutritionally normal boys are taller and heavier than
undernourished boys.

Optimal growth is affected by a number of factors that are manifest in population
diversity. Environmental factors are the primary causes of growth stunting rather than
ethnicity or genetic background (Karlberg et al., 1994). Some of the factors related to
linear growth retardation include poor hygiene, overcrowding and high disease loads
(Karlberg et al., 1994), all of which are prevalent in South Africa at present.

Compared to U.S. reference data (Table 59), the present sample of boys is, on
average, shorter and lighter, and falls between the 25th and 50th centiles. On the other
hand, the present sample of boys is taller than many sub-Saharan populations, who fall
between the 5th and 25th centiles of U.S. reference data for mass and stature (Eveleth,
1986, Eveleth and Tanner, 1976, 1990; Cameron, 1991). The BMI of the present sample
of boys is around the 50th percentile of the U.S. reference, while the triceps skinfold is
within :1: 1 SD of the U.S. reference. The subscapular skinfolds of African and European
boys fall within 3: 1 SD of the reference, while Indian boys have larger subscapular
skinfolds than the U.S. reference, which suggests proportionately greater accumulation of

subcutaneous adipose tissue on the trunk.

Race and SES

While race does not account for a signiﬁcant proportion of the variation in the
growth status of the present sample, some of the trends are similar to those observed in

previous studies of South African populations. In earlier studies, African boys were the

171

 

Table 59. Size and fatness of 10 year old South African boys compared to U.S.

 

 

 

 

 

 

reference values.
Reference data' African Indian European Total
5%.

25 I 50 l 75 Md Md Md Md
Mass, kg 31.4 34.8 39.2 31.5 30.6 33.3 31.5
Stature, cm " 134.0 137.0 141.5 137.9 139.0 139.2 138.4
BMI, kg/mz 16.1 17.3 19.1 16.6 15.9 17.4 16.4

Skinfolds, mm

Triceps 7.5 11.0 16.5 10.0 11.2 11.2 10.5
Subscapular 5.0 6.0 9.5 6.8 7.3 6.3 6.8

 

‘NCHS reference values (Najjar and Rowland, 1987).
b Mean values are given for stature

shortest and lightest, and European boys were the tallest and heaviest (Smit and
Potgieter, 1967 ; Richardson, 1978; Kotze et al., 1987). In the present study, European
boys are the tallest and heaviest, and have the highest BMI compared to African and
Indian boys. Further, the median BMI of European boys is similar to the U.S. reference
median, and Indians have a BM] below the median (Figure 5). On average, African boys
are shortest, while Indian boys are lightest with the lowest BMI.

Racial variation in stature have been observed in the U.S. (Krogrnan, 1970; Garn
and Clark, 1976), England (Rona and Chinn, 1986), and Guyana (Ashcroft and Desai,
1976), where children of African ancestry are slightly taller than children of European
ancestry. Trends among immigrant Indian populations in England (Rona and Chinn,
1986) and Guyana (Ashcroft and Desai, 1976), suggest that children of Indian ancestry

are shorter than children of African and European ancestry.

172

 

 

 

 

 

 

 

 

 

22
21 di-
20 a» Q
9 o
19 ~>
18 +
S 17 a. US
9' 0
16 .1
15 at -
i
14 ~» -
13 -+
12 i +
African Indian European
Race

 

 

 

Figure 5. BMI (mean 1 SD) of African, Indian and European boys compared to the
U.S. reference median.

People of African ancestry in the western world are oftenmost likely of West
African origin. Similarly, immigrants from the Indian subcontinent are not a
homogenous group. Thus, comparisons of trends between sub-Saharan and West African
populations, or pe0ple of Indian ancestry on other continents are tenuous. The mass and
stature of higher and lower SES South Africa Indian boys of the current sample are
similar to those of higher SES Indian boys from India (Raghavan et al., 1971), but are
taller than those of two other Indian samples (Pakrasi et al., 1988; and Singh et al,. 1987)

(Table 60).

173

Racial comparisons are confounded by SES. The short stature of sub-Saharan
African boys compared to the U.S. reference may be due to the inﬂuence of
environmental variables on growth potential. The tall stature of a single sample of rural
African boys (Badenhorst et al., 1992), however, points out that this relationship is not
unilateral. Systematic examination of the genetic and environmental matrix of South

Africa, as well as their effects on attained growth is necessary.

Table 60. Mass and Stature of Indian boys from India and South Africa

 

Statlne Mass

Median Mean SD Median Mean SD

 

Indian (Raghavan et al., 1971)

Lower SES 126.7 127.0 5.73 22.4 22.3 2.6

Higher SES 138.0 138.5 6.8 31.0 32.4 5.9
Calcutta (Pakrasi et al., 1988) 133.0 5.5 25.5 5.8
Patella City (Singh et a1,. 1987) 132.6 5.9 24.1 3.1
Current sample

Lower SES 137.6 137.6 7.0 29.3 30.5 7.0

Higher SES 139.9 139.5 6.8 31.3 33.0 7.8

 

In the present study, Indian boys are taller and heavier than Aﬁican boys. The
smaller body size of Africans compared to Indians in the present study and in the trends
in earlier studies is probably a reﬂection of secular, SES, nutritional, and perhaps
geographic variation among samples (Smit and Potgieter, 1967; Richardson, 1978, 1986;
Kotze et al., 1986). However, the differences most likely reﬂect SES variation between
Africans and Indians in South Africa.

In the present study, SES contributes differently to the variance in the mass and

stature of Africa and Indian boys. SES accounts for 5% and 7% of the variation in

174

stature and mass of African boys, respectively. On the other hand, SES does not account
for any of the variation in the stature and mass of Indian boys.

The signiﬁcant race difference in the overall SES ratings of the sample is due to
the composition of the African, Indian and European samples (Table 61). There are
proportionally more lower SES African boys compared to Indian boys, and all European
boys are of higher SES. Additionally, there are differences in the SES groups. On
average, lower SES Africans come from families with relatively lower incomes and less
educated parents compared to lower SES Indian boys. The heads of lower SES African
households, however, have greater occupational prestige compared to lower SES Indians.
A possible explanation for this incongruity is that some vestiges of the apartheid era
remain, and Africans are still paid less compared to others for the same job.
Additionally, afﬁrmative action may have resulted in a larger number of Africans getting
jobs with better titles, but the pay is commensurate with education, training and
experience. On the other hand, many Indians are traders, which is allocated an
occupational prestige rating of 15. This rating is lower than that of a factory worker
(21), cook (20), gardener (16), general merchandise sales assistant (21), or sales manager
(69). This is probably a factor of semiotics since the term trader is used colloquially by
Indians to describe all types of retail business owners and includes hawkers and informal
vendors to chain store owners. The traditional use of the word trader has not been
considered in the classiﬁcation criteria offered by Schlemmer and Stopforth (1979).

Higher SES Indian boys come from households with a higher occupancy ratio and

a lower mean education of parents compared to European boys. Higher SES African

175

Table 61. Indicators of SES (mean i SD) by race for upper and lower SES.

 

 

 

Africa_n Indian Europeag_

Lower SES

Household Occupancy, # of people / room 1.3 :l: 0.9 1.4 :1: 0.7

Education of parents, years of schooling 8.7 :1: 4.6 10.4 :1: 3.7

Mid-parent income‘ 3.0 r 1.4 3.6 11.3

Occupational prestige, score 33.5 :1: 16.3 27.4 i 19.4

SES rating 1.9 :1: 0.7 1.8 :t: 0.8
Upper SES

Household Occupancy, # of people Iroom 1.1 :1: 0.8 1.2 :1: 0.6 1.3 :l: 0.8

Education of parents, years of schooling 12.8 :1: 2.1 12.3 i 1.8 13.3 :t 3.2

Mid-parent income' 3.9 :1: 1.3 4.2 1 1.3 4.8 i 1.5

Occupational prestige score 67.4 :1: 10.0 67.5 i 7.0 65.0 r: 16.6

SES rating 3.8 :t 0.5 3.8 :1: 0.4 3.9 i 0.5
Overall SES 2.9 11.1 3.2 :t 1.0 3.9 :1: 0.5

 

' midparent income is expressed in terms of income categories (Table 16).

boys, on the other hand, come from families with the lowest mean income and European
boys come from families with the highest mean income.

The relationship between sociodemographic factors and stature appears to be
different for South African and Swedish children. There is a signiﬁcant association
between height and number of children in the family (r = -0.60, p<0.01), household
density (r = -0.52. p<0.01) and occupation of the father (r = -0.55, p<0.01) of 10 year old
Swedish children (Cemerud, 1993). Household density (number of people per room),
number of children in the family, household number (number of people in household),
and the occupation of the father are not signiﬁcantly related to stature (r = 0.03;

= -0.10, r = 0.10; r= 0.09, respectively) in the present study (Table 62).

In summary, European boys are the tallest and heaviest, and African boys are the
shortest and lightest, in the present study. While environmental conditions in South
Africa have improved for Africans, these improvements may or may not have been

adequate for Africans to attain their growth potential. The difference among racial

176

groups in stature is most likely explained in the context of variation in SES and living

conditions between groups.

Table 62. Regression analyses of stature by sociodemographic characteristics for
the total sample (n=330)

 

 

 

Independent Variable R2 Beta-valuea B” F
Household number 0.00 0.02 -0.06 0.74
Occupation of parent 0.00 0.15 0.05 0.19
Household occupancy ratio 0.00 0.00 0.02 0.33
Number of children in household 0.01 0.02 -0.11 2.36
'regression coefficient b standardized regression coefficient

Nutritional status: BMI-for-age

The BMI was used as the criterion for nutritional status. The use of the BMI as
an indicator of nutritional status, however, is limited. BMI is an indicator of total body
fat but is a poorer estimate of percentage body fat compared to relative weight, weight
/stature3, and skinfolds (Roche et al., 1981). The correlation between BMI and
percentage body fat is moderate (r = 0.68) in boys 6 to 12 years of age (Roche et al.,
1981). Thus, some of the boys in the present sample that are classiﬁed as
undernourished may, in reality, be adequately nourished small boys since the BMI
probably reﬂects their current weight-for-height and not their nutritional history.

Based on the BMI, 24.6% of the present sample is categorized as undernourished
(BMI s 15th centile) and 13.7% is classiﬁed as overweight (BMI 2 85th centile).
Proportionally more Indian boys are underweight, but only 3 of the small sample of

European boys are overweight (Table 63). The BMI of Indians, on average, is lower

177

than that of Africans and Europeans. SES variation among Indian boys affects their
nutritional status. There is a higher incidence of underweight Indian boys from lower
than upper SES families (50.0% vs 31.3%).

The major ethnic groups among Indians are Moslem, Gujerati, Hindi, Tamil and

Telegu. Each group has different cultural practices that could potentially impact the

Table 63. Means, standard deviations, medians and ranges of the BMI by
nutritional status of African, Indian and European boys

 

Undemourished Nutritionally normal Overweight

 

(s 15 centile) (15111 > NN < 85th) (2 8Sth percentile)
African Mean 14.4 i 0.9 16.8: 1.1 22.321: 2.3
Median 14.5 16.7 22.2
Range 10.0-15.1 15.3-19.6 19.7-29.5
Number (%) 38 (18.3%) 142 (68.3%) 28 (13.5%)
Indian Mean 14.1:1: 0.7 16.6 :1: 1.1 22.1:1: 2
Median 14.3 16.2 21.7
Range 12.6-15.1 15.2-19.6 19.6-26.6
Number (%) 72 (45.3%) 65 (40.9%) 22 (13.8%)
European Mean 14.15: 0.7 17.4 :1: 1.3 24.1 :t: 3.9
Median 14.5 17.7 25.6
Range 13.1-15.0 15.4-19.2 19.8-27.0
Number (%) 7 (17.1%) 31 (75.6%) 3 (7.3%)

 

nutritional status of the individual, and may offer possible explanations for the
polarization of BMI among the Indian boys. Peters and Ulijaszek (1992) found
differences in arm circumference, and the triceps and subscapular skinfolds among
children of different Indo-Mediterranean ethnic groups (Sikh, East African Hindu, Indian
Hindu, Indian Moslem, Pakistani Moslem and Bangladeshi) in England. The practice of
abstinence from food for prolonged periods of time is common among Moslem and

Hindu children and adults, and may affect the nutritional status of individuals.

178

Ethnic/cultural differences among Indians are large, but were not considered in this
study. An investigation of the ethnic differences and practices on nutritional status may

provide insights into high prevalence of low BMIs observed among Indian boys.

Fatness

Adiposity is associated with physical activity, lifestyle variables and economic
prosperity. In populations of the developing world, there is ordinarily greater adiposity
among upper SES children compared to lower SES children (McGarvey et al., 1989),
while this trend is often reversed in populations of developed countries, especially among
females. In the present sample, there are no signiﬁcant SES differences in adiposity,
among boys.

SES and cultural practices affect nutritional status. Nutritional status explains a
large proportion (49%) and race only for a small proportion (2%) of the variation in the
accumulation of subcutaneous adipose tissue in the present sample. Overweight boys are
the fattest and undernourished boys are the leanest. Additionally, Indian boys have
larger accumulations of subcutaneous adipose tissue compared to African and European
boys

In the present study, Indian boys have a lower BNﬂ, and are lighter but fatter than
African and European boys. Indian boys also have greater accumulations of adipose
tissue on the trunk relative to the extremities. Smit and Potgieter (1967) observed that
Indians are smaller (shorter and thinner) than European boys, but have similar triceps,

subscapular and abdominal skinfold thicknesses. Similar trends are observed in the

179

present study. The apparent paradox between a lower BMI and thicker skinfolds of
Indian boys was not discussed by Smit and Potgieter (1967).

The higher adiposity of Indians has also been observed in studies in Guyana. In
Guyana, Indians appear to have a greater overall accumulation of subcutaneous fat
regardless of their nutritional status compared to Africans (Ashcroft and Desai, 1978).
Indians tend to have larger skinfolds compared to African and Europeans in the present
study (Table 64).

Similar BMI-fatness relationships have been observed among Chinese
populations (Wang et al., 1993; Wang et al., 1994). Both male and female of
Singaporean Chinese youth (17 to 21 years of age) have a lower mean BMI but larger
mean triceps skinfolds (Wang et al., 1993) compared to U.S. reference data (Najjar and
Rowland, 1987). Asians in the New York City area show the same trend (Wang et al.,
1994). The differences in the BMI-fatness relationship appear to be related to the
population variation in relative subcutaneous fat distribution and perhaps body
proportions. The results emphasize the need for care in utilizing the BMI as an indicator
of nutritional status in different populations.

African boys in the present study, on the other hand, have a smaller accumulation
of subcutaneous adipose tissue than Indian or European boys. Corresponding trends
were observed in South Africa by Richardson (1978), who noted that African males are
signiﬁcantly thinner than Indian and European males, with smaller skinfolds. African

boys have skinfolds that are between 16% to 24% lower than NCHS reference values

180

Table 64. Means and standard deviations for the BMI, skinfolds, limb
circumferences and skeletal breadths of 10 year old boys from selected South
African studies, by race.

 

 

 

 

 

 

 

 

AMmm
Present Study Badenhorst Smit and Potgieter Pedi Carey
(1992)" (1967)' (Leary, 1968)‘ (1980)
BMI', kg/m2 17.1 r 2.6 15.0
EkmmMs
Biceps, mm 6.6 : 3.7 3.8 : 1.0
memnmn “9:56 60:21 57:L2 52:07 63:35
&mmthnmm 87:55 47:04 43:08 52:25 55:32
&mmmuamm 69:46 36:06
AhhmmﬂJmn 108:70 50:08 35:07 39:06
2 of2 trunk, mm 19.5 : 12.1 9.7 7.8 9.2
2 of 2 extremity, mm 18.5 :1: 9.0 9.8
2 of 5, mm 44.9 : 24.7 23.1
TER", mm 1.0 a 0.2 0.99 "
Circumferences .
Anmcm 202:25 190:14 168:14
Cﬂﬁcm 277:26 255:2J 233:2J
Breadths
Biacromial, cm 28.6 : 2.4 27.6 :1: 1.8 31.6 : 1.8
Bicristal, cm 19.6 : 2.1 19.1 :1: 1.2 20.8 :1: 1.1
European Indians
Present study Smit and Carey (1980) Present study Smit and
Potgieter Potgieter
(1967i (1967)'
BMI', ltg/rn2 17.4 1 2.6 16.6 16.6 :1: 2.9 14.8
Skinfolds
Triceps, mm 13.2 : 5.9 8.9: 3.3 7.4 : 2.5 13.1 : 6.0 6.9 :1: 1.4
&qumhnmm 80:31 59:29 56:L7 103:39 48:08
Ahhmmﬂnmn 1L7:72 69:43 141:95 58:22
2 of2 trunk, mm 19.7 :11.8 12.8 32.7 :1: 22.0 12.7
Circumference
Anmcm 2L0:25 2L0:30 203:3J 188:10
Calf, cm 28.1 : 2.7 28.0 : 2.2 27.3 : 3.1 24.6 :1.3
Breadths
Biacromial, cm 29.8 : 1.3 30.3 : 1.7 28.6 : 2.3 28.0 : 1.6
Bicristal, cm 21.4 :t 1.3 22.0 :1: 1.3 20.1 : 1.3

19.8 :2.1

 

' BMI is ealculated from means for stature and mass.
b The sum of 2 and 5 skinfolds, and the TER are based on means of the individual skinfolds

181

(Kotze et al., 1986). However, differences among individual skinfold sites are apparent.
The subscapular skinfold of African boys is only slightly larger than that of European
boys in the present study (Table 59). Kotze et al. (1986) also found that African boys
have the smallest triceps skinfold compared to Indian and European boys, but have
subscapular skinfolds that are close to reference values. The earlier South African
studies did not address relative fat distribution.

Similar racial variations in subcutaneous adipose tissue have been observed in
Dominica (Robson et al., 1971). In a Caribbean Island population from the Island of
Dominica, poorly nourished Africans had lower triceps skinfold thickness compared to
Europeans of similar nutritional status, and the triceps skinfold of Africans was 18.2%
lower than WHO reference values. However, there was no difference in the thickness of
the subscapular skinfold between Africans and Europeans from the Island of Dominica
(Robson et al., 1971).

Compared to previous reports on South Africa populations, Africans and Indians
in the present sample have larger absolute accumulations of subcutaneous fat on the trunk
and extremities (Table 64). These trends are similar to those observed for Indian
populations in England (Peters and Ulij aszek, 1992) and Guyana (Ashcroft and Desai,
1976). In the present sample, Europeans also have larger absolute accumulations of
subcutaneous fat at all sites, compared to previous studies.

Overall, the current sample of boys is fatter than previous samples. Additionally,

the arm and calf circumferences of African and Indian boys reﬂect the greater upper

182

extremity subcutaneous fat. The apparent fattening of African and Indian boys may be

due to the improved living conditions.

Secular Trends

Secular trends are generally characterized by an acceleration in growth and/or
maturation. Positive secular change is identiﬁed by an increase in height or weight
attained at given ages observed over a period of time (van Wieringen, 1986). Monitoring
secular change is an important indicator of public health. The deviation of the population
from the reference provides an indicator of the distance between attained and potential
growth, and is based on the assumption that the reference does not change. The
difference between actual and potential growth status is primarily associated with
environmental variables, particularly infectious and parasitic diseases and compromised
nutritional status. South Africa is characterized by a high infant mortality rate of 50 per
1 000 liveborn (Sidiropoulos et al., 1997) and an under ﬁve mortality rate of 69 per 1
000 (World Bank, 1995). There is a high rate of urbanization and a constantly high rate
of migration into informal settlements in the Guateng conurbation each month (Weekly
Mail, May 16, 1997). The lack of urban infrastructure, access to water, sewage facilities
and garbage removal (Weekly Mail, May 16, 1997) in the informal settlements is
associated with a high incidence of communicable and infectious diseases and
malnutrition (Y ach, 1988).

There is less variation in the stature, mass and BMI among African, Indian and
European boys in the present sample compared to previous studies (Figure 6). Data from

30 to 40 years ago indicate a trend in South Africa towards increased stature in European

183

boys over a six year period from 1958 through to 1962-1963 (Table 6). This increase in
stature coincided with the beginning the South African Nationalist party rule and the
enforcement of the apartheid policies of South Africa. There was a signiﬁcant bias of the
socioeconomic structure in favor of White South Africans (Wolpe, 1987). The greater
stature of European boys likely reﬂects the markedly improved living conditions of
Europeans during the apartheid era.

Walker and Walker (1977) reported a 5.8 cm increase in the stature and 3.0 kg
increase in the mass of 10 year old African children from Letaba between 1938 and
1975, almost 40 years. Despite the increase in stature and mass, the growth status of
African boys was still below the 10th percentile of the NCHS reference in 1975.

Sample means for stature, mass and the BMI of 10 year old South African boys
of African, Indian and European ancestry (Smit and Potgieter, 1967; Richardson, 1978,
Kotze, 1986; present study) are compared in Figure 6. The small sample (n=41) of
Europeans in the present study may not be representative of the pepulation of the greater
Johannesburg area.

Between 1962-1963 and 1997, there has been an approximately 5 cm increase in
stature and 6 kg increase in mass of Africans and Indians, but between 1978 and 1986,
there was a decline in stature for African and Indian boys to about the levels of 1962-
1963. Most of the increase in the mass of African and Indian boys and in the stature of
African boys occurred after 1980. The increase in stature of Indian boys, however,
appears to be more gradual, occurring over the 30 year period. Between 1967 and 1997,

there has been a 1.4 cm decrease in the stature of Europeans, with the largest decrease

184

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Stature
142
g 140 ‘ a
- 138 6 8
g 136
a 134 :1 e
g 13% o . .Aflican
1960 1970 1980 1990 2000 ”men
.European
Year
Mass
35 T ‘
3 g? ‘" ‘ ‘ U .Aflican
31' ‘” ‘ nlndian
a 29 “L g
5 27 «- O 3 ‘European
25 1 i i 1
1960 1970 1980 1990 2000
Year
BMI
18 -.
E: 17 ,. A 0
.5. a ‘ 1:1
3 16 ‘" 0 C1
g“ 15 -~ 0 n .Aﬁican
m 14 J' + i ﬁ' ulndian
Year

 

 

 

Figure 6: Stature, mass and the BMI of African, Indian and European boys
in 1962-1963 (Smit et al., 1967), 1978 (Richardson, 1978), 1981 (Kotze et al., 1987)
and 1997 (present study).

185

between 1967 and 1978. The decline in stature could be due to cohort and sampling
variation. However, the political climate in South Africa became hostile following the
Soweto uprising (1976) and was followed by a period of harsh repression until the late
1980s. There was stricter enforcement of inﬂux control laws and high incidence of
strikes, effectively reducing the earning capacity of the masses. The increase in both
civil disturbance and the repression may have provided negative stresses on growth.
Additionally, while the stature of European boys has been increasing since 1978, it is
still lower than it was in 1962-1963. In contrast, the stature and mass of European boys
does not differ between 1962-1963 and 1997. The gain in weight surpassed the gain in
height for Africans and Indians with an increase of approximately 3 kg/m2 in the BMI.
However, the slightly higher BMI of European boys in 1997 is related to a small decrease
in height relative to the previous survey. This probably reﬂects sampling variation, given
the small number of European boys in the present study.

The positive secular changes for Africans and Indians are presumably a reﬂection
of the social changes that have occurred during the last 20 years. However, the available
data does not provide an indication as to whether the majority of the positive secular
changes occurred in the post-apartheid period. There has been an increase in life
expectancy, reduced infant mortality rate, increased annual growth and an increase in the
overall human development index (Table 65). Improvement in living conditions and the
accompanying decline in morbidity and mortality are reﬂected in positive secular change

for Africans and Indians. The current improvements in public health with campaigns for

186

increasing the number immunized, free mother and child medical care, and free clinics

are likely to impact the growth of subsequent generations of South Africans.

Table 65. Changes in indicators of the quality of life of South Africans from 1970 to

 

 

1995.
1970' 1993' 1995"

Life expectancy at birth, years

Women 56 66 69.2

Men 50 60 62.2
Infant mortality rate per 1 000 liveborn 79 52 50.4
Average annual growth, % 1.3 2.7
Human Development Index 1980" 1991"

Africans 0.394 0.500

Indians 0.655 0.836

Europeans 0.739 0.901

Colored 0.532 0.663

Total 0.557 0.677

 

' 1970 and 1993 statistics from World Bank Report (1995)

" 1995 statistics from Sidiropoulos et a1. (1997).

Other positive environmental pressures that have occurred in the post apartheid

South Africa are increased employment possibilities, the relocation of large numbers of

urban and rural poor into the urban periphery, increasing access to public health services

and employment, and a more varied diet. Conversely, the crowding in peri-urban

squatter settlements, the lack of running water and sanitation facilities, and removal of

human and other waste exert negative environmental pressures on these populations.

However, the living conditions in the pen-urban squatter settlements are relatively better

than those in rural areas and possibly account for the positive secular changes observed in

the stature and mass of African boys. The environmental changes and pressures that are

taking place in the dynamic social structure of South Africa needs to be systematically

addressed within the context and history of the lives of the individuals.

187

A corollary of the improved living conditions is the increase in the BMI of 10
year old African (2.1 kg/mz), Indian (1.8 kg/mz) and European (0.8 kg/mz) South African
boys compared to boys 35 years ago. African boys have the greatest increase in the mean
BMI. The increase in BM] is resulting in a higher proportion of the South Africans
being classiﬁed as obese. Obesity is a risk factor for many cancers and coronary heart
disease (Must et al., 1992), as well as the metabolic cascade that results in type II
diabetes (Reaven, 1988). Additionally, Indian boys have signiﬁcantly higher
accumulations metabolically active subcutaneous adipose tissue on trunk which poses
major health risks. The change in the nutritional and growth status of the South African

population may have some implications on the health-related ﬁtness of the population.

Physical Fitngs

AAHPERD and NCYFS reference values were used for the comparison of some
ﬁtness items (50 meter dash; 9 minute run, sit and reach). Other items (standing long
jump, grip strength) were compared to available data on 10 year old boys from South
Africa, the U.S., and Mexico (Table 66). Reference data for the 100 meter shuttle run,
30-second sit-ups and 30-second push-ups for Belgian boys (Beunen et al., 1988) are not
available for 10 year old boys. Further, comparative data on the performances of 10 year
old South African boys are apparently not available. Comparisons for African, Indian
and European boys were made separately.

Variation in health- and performance-related ﬁtness of boys in the present study
was examined in the context of race, SES and nutritional status. SES and nutritional

status do not contribute to the variation in the push-ups, distance run, and sit and reach.

188

Table 66. Summary of means (SD, where available) for physical ﬁtness of 10 year
old boys from selected studies.

 

Dash, Grip, kg‘ Standing 9-min run. hash-ups". Sit and Sit-ups”

 

m/sec long jump, m reach. cm
“9.!
Present Study
African 5.3 19.3 150.5 1617 23.7 27.8 16.4
(3.6) (21.5) (360.5) (8.2) (5.7) (5.3)
Indian 5.1 18.6 149.8 1556 25.2 24.0 16.5
(3.4) (21.9) (336.9) (8.3) (5.9) (5.9)
European 5.3 19.3 157.7 1875 28.0 25.7 17.2
(3) (21.9) (471.6) (8.0) (5.4) (5.5)
South Africa
Smit (1968)c
Indians 123.1 , 29.7 18.3
(16.4) (10.7) (5.8)
Colored 138.7 41.6 23. 8
(18.5) (15.6) (8. 0)
as
Clarke (1976) 5.6 150.0 1570 24.0
AAHPERD (1984) 152.0 1545 25.0 34.0
Slaughter and Inhrnan(1982) 5. 4 147.0
(0. 7) (21.0)
Haubenstriclter and Seefeldt (1986) 6.2 153.0
NCYFS (1995) 34.3 34
Malina mid Roche (1983)
Combined 22.5
Black 5.6 156.0
White 5.4 144.0
Other countries
Beneﬁee and Malina (1996) - Senegalese 4.3 137.0
(0.3) (0.2)
Malinaeta1.(1991)-Mexican 13.5
(3-2)
Malina and Buschang (1985) 4.7 15.0 120
(2. 4) 110. 4)

 

:Diﬁ'crentdynamometerswereusedtomeasmegripsucngthineaehsmdy
:30secondpush-upsinthepreaentsnrdy, 60-secoodpushupsforothersurdies
:30secondsit-upsinthepresentsmdy, 60-secondpuahupsforothersnrdies
dSit-upstoexhaustion

189

Race accounts for 3%, 2% and 5%, of the variation in these items, respectively.
Nutritional status accounts for only 1% of the variation in the dash, 11% of the variation
in absolute grip strength (kg), and 22% of the variation in grip strength per unit mass
(kg/kg). SES accounts for an additional 2% of the variation in the absolute grip strength
(kg), while race accounts for an additional 1% of the variation in relative grip strength
(kg/kg)-

Other factors thus inﬂuence health-and performance-related ﬁtness. Performance
on ﬁtness tests is affected by past experiences in activity and ﬁtness testing. Sit-up and
push-up performances often are a result of practice and familiarity with the activity.
Similarly, the agility demonstrated on the shuttle run may be affected by practice. Many
children perform poorly on these items because they do not have previous experiences
with such standardized tests.

Most of the studies used in the comparisons grouped subjects from 10.0 to 10.9
years of age. The grouping by age, however, was not speciﬁed in several studies (Smit,
1967; Beneﬁce and Malina, 1996), and 10 years of age was interpreted as a mean age of
10.5 years. Estimated speed (m/sec) on the 50 meter dash was compared to performance
on the 50 yard and 35 yard dashes, after adjusting for meters and yards. Overall, the
present sample of boys has a running speed that is slower than those reported in studies
on U.S. samples. The performance of African and European boys on the distance run is
better than the median performance of U.S. boys, but that of Indian boys is slightly lower
compared to U.S. boys. Comparison of grip strength is confounded by testing apparatus.

The Martin manometer (Beneﬁce and Malina, 1996), Narragansett dynomometer (Malina

190

and Roche, 1983), and the Stoelting adjustable dynamometer (Malina et al., 1991,
present study) are examples. With these limitations, the grip strength of the current
sample is lower than that of the U.S. comparative values.

Push-ups and sit-ups scores could not be compared directly with those in the
literature because 30-second sit-ups and push-ups tests were used in the current study and
the 60-second sit-ups and push-ups tests were used in the comparative studies .
Performance on the 60-second sit-ups and push-ups are not equal to twice that of the 30-
second sit-ups and push-ups, respectively, since fatigue is a potential limiting factor.
Nonetheless, the 30-second push-ups and sit-ups scores in the current study are close to
those previously reported for Indians (Smit, 1968) on the 60-second push-ups and sit-
ups, suggesting better performances on these items by the current sample of Indian boys.

Comparisons indicate that American Black children have better performances on
the dash (Hutinger, 1959; Milne et al., 1976; Malina and Roche, 1983) and standing long
jump (Milne et al., 1976). In earlier studies of older South African boys, however,
Africans did not have a performance advantage (Table 9). Europeans had the best, and
Africans had the poorest, performances on all items compared - sit-ups, shuttle run,
standing long jump, 50 yard dash, and distance run (Sloan, 1966).

Despite the low association between race and physical ﬁtness, there are race
differences in ﬁtness in the present study. Europeans have the best performances on all
items except on the sit and reach, with mean performances on the standing long jump and
distance run that surpass the 50th centile of AAHPERD norms. There are no differences

in the performance of Africans and Europeans on the dash and absolute grip strength.

191

Africans have better performances on the sit and reach and greater grip strength per unit
mass (kg/kg) compared to European and Indian boys. Indians have the lowest scores on
most physical ﬁtness test items (dash, grip, distance run, shuttle run, sit and reach and
standing long jump), with Africans having the lowest scores only for the sit-ups and
pushups. Trends in physical ﬁtness associated with race are similar to those reported by
Goslin and Burden (1986) for African and European South African boys (16 to 18 years
of age).

The low physical ﬁtness of Indians relative to African, European and “Colored”
South Africans has been previously reported (Smit et al., 1968). The higher ranking of
the physical performance of the combined Indian-“Colored” group compared to Africans
reported by Sloan (1966) could be due to the better performances of “Coloreds”.

The ﬂexibility of the lower back and upper thighs of African and European boys
in the current sample is above the median, with Indian boys having performances slightly
below the 50th percentile of the AAPHERD (1984) reference values. The sample of
boys included in the NCYFS study, however, have better performances on the same sit
and reach test than the current sample (Table 66). Both African and European boys have
greater endurance on the distance run compared to U.S. samples (AAPHERD, 1984;
Clarke, 1976). Explosive strength on the standing long jump in the present study falls
within the range of means reported in the U.S., with European boys jumping further than
the U.S. mean.

The present sample of African boys have better performances on physical ﬁtness

tests compared to those of Sloan (1966). In addition to sampling differences, the

192

differences in ﬁtness may be due to changes in SES and nutritional status over the last 30
years. The change in the hegemony and political economy of the country in the 19905
resulted in a shift in the interrelationship of race, SES, and nutritional status. Other
factors, such as rearing conditions, health status, habitual physical activity, the
opportunity to practice skills, and greater involvement in organized sport could be
signiﬁcant determinants of physical performance.

If the U.S. reference medians (AAHPERD, 1984; NCHS, 1987) are used as the
cut-off point for speciﬁc ﬁtness tests, South African boys score above the median for
cardiorespiratory endurance and lower back ﬂexibility (Table 67). Skinfold thicknesses

are comparable, allowing for measurement variation and different calipers.

Table 67. Comparison of fatness and selected performance items of present study
relative to AAHPERD (1980) percentiles by race for higher SES boys and the total

 

 

 

 

sample .
Present Study AAPHERD
African Indian Buggy}: Total sample Median

BMI" 16.9 (16.6) 16.1 (15.9) 17.4 16.9 17.3
Distance run, 111 1621 (1617) 1513 (1556) 1875 1622 1545
Sit and reach, cm 27.3 (27.8) 24.1 (24.0) 25.7 26.2 25
Standing long jump, cm 151.1 (150.5) 150.6 (149.8) 157.7 151.1 152.0
Skinfolds, mm

Subsca ular" 7.0 (6.8) 7.6 (7.3) 6.3 6.8 6.0

Triceps 10.8 (10.0) 12.2 (11.2) 11.2 10.5 11.0

 

' Total sample values are indicated in brackets.
b NCHS reference data (Naj jar and Rowland, 1987).
° Medians are used for measures of fatness (BMI and skinfolds), and means are used for performance

Health-related physical ﬁtness as indicated in the distance run, sit and reach, and
subcutaneous adipose tissue at the subscapular and triceps sites of the current sample
differs by race. Indian boys have a lower median BMI compared to the AAPHERD

reference, but have slightly thicker skinfolds at the triceps (12.2 vs 11.0 mm) and

193

subscapular (7.7 vs 6.0 mm) sites. Thus, the present sample of Indian boys has slightly
more subcutaneous fat but a lower BMI than the reference population. This trend of
large accumulations of subcutaneous adipose tissue and low BMIs has not been observed
in earlier South African studies, but has been found in Asian populations (Wang et al.
1993; Wang et al., 1994). The higher subcutaneous fat with a lower BMI in the present
sample may be indicative of higher adiposity among the current sample. Additionally,
they have poorer performance on the distance run and the sit and reach tests compared to
U.S. boys.

African boys also have a lower BMI and a slightly thicker subcutaneous fat depot
at the subscapular (7.0 vs 6.0 mm) site compared to the AAPHERD reference. However,
African males have better endurance on the distance run and greater ﬂexibility of the
lower back compared to the reference. European boys, on the other hand, do not differ
signiﬁcantly from U.S. boys in the BMI or subcutaneous adipose tissue, but perform

better on the distance run and standing long jump than the AAPHERD references.

Fatness and ﬁtness

The relationship between fatness and ﬁtness is not clear for the present sample of
boys. There is evidence to indicate that the fatness of an individual affects his/her
performance on physical ﬁtness items. On one hand, fatness has a negative impact on
performances in which the body has to be projected through space such as the standing
long jump and dash (Malina, 1975; Malina, 1994). On the other hand, overweight
children are stronger on tests of absolute strength (Beunen et al., 1983). Altemately,

underweight individuals have poorer performances in absolute grip strength (Malina,

194

1984). In the present study, race appears to affect the relationship between fatness and
ﬁtness.

The relationship between fatness and ﬁtness of the present sample of boys was
further examined using stepwise regression analyses. Regression analyses for the total
sample, and separate analyses by race of fatness (sum of ﬁve skinfolds) and the BMI on
physical ﬁtness variables were performed (Tables 68 to 71). Total fatness explains a
small proportion of the variation in the dash (5%), distance run (5%), push-up (1%), sit
and reach (2%), sit-ups (4%) and standing long jump (2%) performances in the total
sample. On the other hand, total fatness accounts for 31% of the variation in grip
strength per unit mass. The BMI also accounts for some of the variation in grip strength
per unit mass (2%), the sit and reach (8%), and standing long jump (3%).

The relationship between fatness and physical ﬁtness differs across race groups,
and there are no clear patterns between regional accumulation of subcutaneous adipose
tissue and physical ﬁtness (Tables 68 to 71). Fatness explains a fraction of the variation
for only three of the eight physical ﬁtness items in European boys, ranging from 12% to
38%. In contrast, some of the variation in the performance of African and Indian boys
on 6 of the 8 ﬁtness items can be explained in part by fatness. Fatness accounts for 2%
to 27% and 3% to 41% of the variation in physical ﬁtness of African and Indian boys,
respectively. Variation in fatness explains a large proportion (27% to 41%) of the
variation in grip strength per unit mass. With the exception of grip strength, these trends
are consistent with those of Belgian boys, 12 to 20 years of age, who showed a negative

association between fatness and a variety of motor performances (Beunen et al., 1983).

195

Fatness was positively associated with static strength and absolute grip strength of boys
in the present study. However, relative grip strength (expressed per unit body weight)
was negatively associated with fatness, possibly pointing to a proportionally greater
contribution of inert fat tissue per unit mass for fatter boys compared to thinner boys.
There is a negative association between performance on physical ﬁtness tests and
fatness of African and Indian boys on all items except absolute grip strength (Tables 68
to 71). However, when expressed per unit body mass, the association between fatness
and grip strength is negative. Similar fatness and physical ﬁtness associations are evident

for European boys on grip strength per unit mass and the nine minute run.

SES variation in physical ﬁtness

One of the confounding factors in estimating the inﬂuence of SES on physical ﬁtness is
that SES is often related to nutritional status. There is, however, considerable evidence
that points to poorer physical ﬁtness in lower, compared to upper, SES boys in
developing countries (Desai et al., 1984; Ghesquiere and Eeckels, 1984; Perez et al.,
1991; Henneberg and Louw, 1998). Perez et al. (1991) reported that upper SES children
have better performances on the 12 minute run, 50 meter dash and vertical jump
compared to lower SES Argentinean children. The differences were attributed to upper
SES children having greater access to organized sport and equipment, and earlier
involvement in both sport activities and spectator sport (Perez et al., 1991). On the other
hand, some researchers (Ponthieux and Barker 1965) found no signiﬁcant SES variation

in physical performance of ﬁfth and sixth grade boys and girls in the U.S. There were no

196

Table 68. Coefﬁcients of the regression of the BMI and fatness on physical ﬁtness

variables for the total sample of South African boys (n=378).

 

 

 

Independent Variable Dependent variable R2 B-Value' Beta" r
Dash, sec 2 of 5 0.05 -0.01 -0.22 17.75“
Grip, kg BMI 0.14 0.46 0.37 56.40"
BMI and 0.16 0.78 0.63 35.24"
2 of 5 -0.04 -0.31

Grip, kg/kg 2 of 5 0.31 0.00 -0.56 161.47"
E of 5 and 0.33 -0.00 -0.36 86.81"
BMI -0.01 -0.23

9 minute run, m 2 of 5 0.05 -2.95 0.22 17.50“

Push-ups, #/30 sec 2 of 5 0.01 «0.03 -0.10 387*

Sit and reach, cm 2 of 5 0.02 -0.03 -0.14 6.98"
)2 of5 and 0.10 -0.12 -0.56 19.27"
BMI 1.11 0.51

Sit-ups, #/30 sec 2 of 5 0.04 -0.04 -0.20 15.52“

Standing long jump, cm 2'. of 5 0.02 -0.12 -0.16 8.82"
2 of 5 and 0.05 -0.32 -0.41 9.69"
BMI 2.39 -0.30

Note: Signs for the dash are inverted.

'regression coefﬁcient b standardized regression coefficient

Table 69. Coefﬁcients of the regression of the BMI and fatness on physical ﬁtness
variables for African boys (n=188).

 

b

 

 

Independent Variable Dependent variable rt2 B-Valuc' Beta F

Grip, kg BMI 0.11 0.45 0.33 23.14“
BMI and 0.14 0.83 0.60 15.27"
2 of 5 -0.05 -0.32

Grip, kg/kg BMI 0.25 -0.02 -0.50 63.65"
BMI and 0.27 -0.01 -0.29 34.98"
2 of 5 0.00 -0.26

9 minute run, m 2 of 5 0.04 -2.80 -0.20 7.71"

Push-ups, #l30 sec 2 of 5 0.02 -0.05 -0.15 4.25"

Sit-ups, #l30 sec 2 of 5 0.03 -0.03 -0.16 5.04“

Standing long jump, cm 2 of 5 0.03 -0.15 -0.18 6.31“

Note: Signs for the dash are inverted.

'regression coefficient b standardized regression coefficient

197

Table 70. Coefﬁcients of the regression of BMI and fatness on selected” physical
ﬁtness items for Indian boys (n=130).

 

 

Independent Variable Dependent variable 112 B-Value" Betac F
Dash, sec 2 of 5 0.12 -0.01 -0.00 17.04"
Grip, kg BMI 0.21 0.53 0.46 33.93"
Grip, kg/kg 2 of 5 0.41 -0.00 -0.64 91.06"
9 minute run, rn BMI 0.03 -2124 -0.18 447*
Sit-ups, #/30 sec 2‘. of 5 0.14 -0.04 -0.38 21.76"

 

Note: Signs for the dash are inverted.
' only variables which had a relationship are included
b regression coefficient ° standardized regression coefficient

Table 71. Coefﬁcients of regression of the BMI and fatness on selectedII physical
ﬁtness items for European boys (n=40).

 

 

 

Independent Variable Dependent variable R2 B-Valueb Betac F
Grip, kg/kg BMI 0.38 -0.02 -0.62 24.15"
9 minute run, In 2 of 5 0.17 -7.61 -0.41 7.78"
Sit-ups, #l30 sec 2 of 5 0.12 0.07 0.35 5.28"l
' only variables which had a relationship are included
b regression coefficient ° standardized regression coefficient

clear SES trends in ﬁtness in the present study, even when the effects of race, and both
race and nutritional status were controlled. SES was quantiﬁed differently in the present
study than in the Argentinean study. In the latter study, SES classiﬁcation was made on
the basis of the area of residence, classiﬁed as lower and upper SES. Cultural and
environmental variables of the shared environment are possible confounders in the ﬁtness
differences observed between upper and lower SES Argentinean boys, and may not be a
true reﬂection of SES differences in ﬁtness. Further, in the present study the effects of
SES on physical ﬁtness differed by race. Higher SES African boys had better

performances on more physical ﬁtness test items compared to lower SES boys. On the

198

other hand, lower SES Indian boys had better performances on more physical ﬁtness
items compared to higher SES boys. Factors that may confound the relationship between
physical ﬁtness and SES are rearing and parental supervision (Malina, 1988), among

others.

Nutritional status and physical ﬁtness

Nutritional status is a signiﬁcant factor affecting physical work capacity and
performance (Satyanarayana et al., 1979; Desai et al., 1984; Malina and Buschang, 1985;
Malina and Little, 1985; Malina, 1985). In the present study, grip strength varied most
with nutritional status. Absolute grip strength was greatest in overweight boys, but when
body mass was accounted for, undernourished boys had signiﬁcantly greater grip
strength per unit body mass, regardless of race. These results are consistent with other
ﬁndings (Ghesequieré and Eeckels, 1984; Malina and Buschang, 1985). The larger body
size of overweight boys was also negatively related to physical ﬁtness. Overweight boys
completed fewer sit-ups and ran a shorter distance in nine minutes compared to
nutritionally normal and undernourished boys. Undemourished boys had superior ﬁtness
on these items compared to nutritionally normal and overweight boys

There is variation in relationships between nutritional status and ﬁtness within
racial groups. Undemourished African boys ran the shortest distance in nine minutes
compared to nutritionally normal and overweight African boys. Undemourished Indian
boys, on the other hand, ran a longer distance in nine minutes compared to nutritionally
normal or overweight Indian boys. Undemourished Indian boys, however, had

signiﬁcantly larger estimated arm muscle area per kilogram body weight than

199

nutritionally normal or overweight Indian boys. Assuming that estimated arm muscle
reﬂects the proportion of muscle mass to body weight, this may contribute to the better
distance run performances. On the other hand, the difference in arm muscle could be an

artifact of measurement error and derivation of arm muscle.

Habitual physical activity and physical ﬁtness

Data on the activity patterns and estimated energy expenditure of South African
children are limited. Estimated energy expenditure in the present study is compared to
other populations of similar age in Table 72. The comparative data for estimated energy
expenditure are based on heart rate monitoring, doubly labeled water, and recall surveys.
Each of the estimates has a reasonable degree of error. Of these methods, doubly labeled
water is the most reliable and has an accuracy of between 90% to 95% of that measured
by indirect calorimetry (Schoeller, 1988). Heart rate monitoring is based on estimations
of energy expenditure from heart rates. However, this method does not discriminate
between activity and non-activity (i.e., stress, fatigue, and environmental factors among
others that may elevate heart rate) heart rates. Thus, estimates based on heart rate
monitoring may overestimate habitual physical activity.

The energy expenditure of Senegalese children (Beneﬁce, 1993) was estimated
from observations of physical activity over four two-day periods in which subjects were
observed for six hours each day. Since the activity was recorded by an observer, there
was greater accuracy of reported physical activity compared to the present study. Similar
values were used to estimate the energy cost of activities in all categories except for hard

intense physical activity, category nine. Beneﬁce (1993) assigned a lower median energy

200

Table 72 Summary of studies of physical activity and estimated energy expenditure

 

 

of children.
Study Sample Physical Activity Measurements
__ta_rv.Seden Light Mﬂ
Senegalese children (12 yrs) Rural children 12.3 hours 10.1 hours 1.3 hours
Observations (n=20)
(Beneﬁce, 1993)
Present study Urban children 20.1 hours 2.6 hours 1.4 hours
Bouchard protocol
Physical Activig Mm
Irish (7, 9 12 and 15 yrs). Younger children 474 mins/day 91 mins/day
(Livingstone et al., 1992) (7-9 years of age)
Doubly labeled water and heart-rate Older children (12 282 may 52 [mg/day
momtonng (n= 19 boys) and 15)
Present Study 240 mins/day" 84 mins/day
W W9 '1 en
in activig e_xpenditure
Colombian (Spurr and Reina, 1988) Undemourished 669 keal/day 1852 keal/day
(10 - 12 yrs)
Heart rate monitoring Nutritionally normal 304 keal/day 2017 kcal/day
Present study 729 krill/day 1256 kcaI/day
5.1% $9 m M k_£L.xcal/k d3
Torun et a1. (1996)' time in hours
(10 - 14 yrs) Developing urban 8.5 7.5 4.0 4.0 53.2
meta-analysis of 3 studies Industrial luban 10.5 5.5 4.5 3.5 56.3
Present study Developing tn‘ban 9.7 10.4 2.6 1.4 38.5

 

Note: ' Samples not indieated
" Includes light and MVPA

201

cost for hard intense physical activity of 7 METS versus 7.8 METS assigned by
Bouchard et a1. (1993) that was used in the present study.

The present sample of children spend considerably more time on sedentary and
less time on light activities compared to rural Senegalese children (Beneﬁce, 1993) and
children in developing and industrial communities (Torun et al., 1996). Rural Senegalese
children do not attend school whereas the boys in the present sample spend an average of
6 hours a day at school. The current sample of South African boys engages in more
MVPA than undernourished Senegalese boys, but less than reported for Irish boys
(Livingstone et al., 1992).

Total estimated energy expenditure is lower in the present sample compared to
other data (Table 72) on the basis of estimates of kcal/day (Spurr, 1988) or kcal/kg/day
(Torun et al., 1996). Heart rate recordings were used to estimate the energy expenditure
of the Colombian children. Thus, energy expenditure was less subjective despite the lack
of discrimination between physical activity and physiological states.

The F AO/WHO/UNU (1985) recommendation for estimated energy intake is 237
kJ/kg/day (56.7 kcal/kg/day), which is approximately 1843 kcal/kg/day for the present
sample of boys (mean weight of 32.5 kg). This sample of boys has a mean estimated
energy expenditure of 1255.8 kcal/day or 38.5 kcal/kg/day. The energy intake of the
current sample was not estimated, but the estimated energy expenditure of the current
sample of boys is low. This may be due to the use of the 3-day protocol and the ability

of the boys to complete the diary.

202

The reliability of recall surveys is related to the competence of the individual to
complete the survey accurately. Cultural differences change the reliability of this
technique across populations. Furthermore, activities at the highest intensity levels were
the most reliably recalled on the Bouchard protocol, in both the current study and the
study of Canadian youth and adults (Bouchard et al., 1983). These methodological
differences and limitations are inherent in the reported estimates.

In the present study, energy expenditure was estimated from a recall survey using
the protocol of Bouchard et al. (1983). Since the accuracy of the recall is dependent
upon the ability of the child to recall and record the activity of the preceding 24 hours
and to have a well developed sense of time, the accuracy of the recall may be low. In the
present study, there is a high association between activity records for lunch and recess
and direct observation (r=0.86), which is similar to that reported by Bouchard et al.
(1983)

The culture of the present South African sample may perhaps play a major role in
the accuracy of recording the data and may be a possible explanation for the low
estimated energy expenditure. Understanding the concept of time and being able to
provide reasonable estimates of the duration of activities is bound by the culture of
chronological time. Most subjects in the present sample did not wear watches and had no
means by which to make reasonable estimations of the time that they spent in activities.
Furthermore, subjects (n=57) reported higher levels of physical activity during lunch and
recess compared to observations, but this again may be due to an inability to estimate the

duration of the activity, and is consistent with other data. There is a tendency for

203

children to exaggerate the amount of time spent in MVPA (Baranowski et al., 1982;
Bouchard et al., 1983). Additionally, the inability of subjects to recall all activities that
they had engaged in over the preceding 24 hours is a limitation. The recall questionnaire
was developed and validated on a sample of Canadians who may have had a better
developed concept of chronological time compared to the present sample of boys. The
low estimated energy expenditure of the current sample may be due to methodological
limitations and the inappropriateness of the measurement tool on this sample of boys.

Physical activity is also affected by interpersonal, physical and social
environments, and public policy and interactions among these variables (McLeroy et al.,
1988; Sallis et al., 1997; King et al., 1995; Sallis and Owen, 1997). Environments rich in
resources such as sidewalks, parks, exercise facilities and health clubs make it easier for
people to be physically active (Sallis et al., 1997). Most South African residential areas
do not have environments that are conducive to physical activity, and are possible
reasons for the lower estimated energy expenditure.

The greater Johannesburg area is characterized by small homes and yards,
crowding, high crime rates, and a lack of parks and recreation facilities. These factors
limit physical activity. Additionally, both organized and petty crime are high in South
Africa and there is a general fear for the safety of self. Children in environments that
have high rates of crime, poverty and crowding engage in higher levels of sedentary
activity (Sallis, 1993; Goodway, 1994). Goodway (1994) found that Flint inner city
children in an environment with high crime, unemployment, small housing conditions,

and poverty engaged in predominantly sedentary indoor activities and had little or no

204

exposure to physical activity outside of school. Poor communities that have few
resources are less able to provide recreation and activity resources than more affluent
ones. Streets and the playgrounds of poor at-risk communities are often not safe places
in which to exercise or play.

Most of the government schools and the two private schools in this sample had
limited play ground areas. Two of the schools visited had play ﬁelds with an area less
than 100 square meters and no indoor facilities. The model-C schools had large playing
areas and signiﬁcantly more sport equipment available for recreational purposes. The
majority of the Government schools had large grounds but lacked facilities and
equipment for physical activity.

An additional mitigating factor for low levels of physical activity in the current
study is the unusually cold and rainy autumn season in 1997. Physical activity of U.S.
children is lower in fall and winter than in summer and spring (Ross et al., 1985;
Baranowski et al., 1993). Ross et al. (1985) reported that winter activity levels were
47% that of summer levels. Since most schools and communities do not have indoor
sport and recreation facilities, this restricts the possibilities of physical activity in
inclement weather conditions. Additionally, if schools have halls or gyms, they are often
used for table tennis and this limits the number of people that could be involved in the
activity. Overall, there is limited space for children to play in, contributing to a
reduction in physical activity, and there is a signiﬁcant relationship between the greater

availability of play spaces and higher activity levels in children (Sallis et al., 1993).

205

Lifestyle factors such as television viewing and time spent in commuting affect
physical activity. African children had an average commute of 1.4 hours a day. This
commute added to the school day of 6 to 6.5 hours, decreased the amount of time
available for physical activity. Furthermore, most African children traveled back and
forth on taxis, which are informally run shuttle services and are very unpredictable.
Children often have to wait for over an hour for the taxi to drop-off the ﬁrst load of
children and return for the next load. The fear of missing the taxi causes children to mill
around the pick-up point waiting for the taxi’s return.

Many Indian children spend between 2.5 to 3.5 hours at vernacular (Indian
language and religious) schools in the afternoon, in addition to their regular school day.
Many European boys and upper SES African and Indian boys go to after-school care
programs until 5:30 or 6:00 pm. After school care programs provide supervised
environments in which children complete their homework or read silently. Thus, in
addition to the 6 hours spent in school, the current sample of boys probably spent an
additional 2 hours a day in sedentary activity. This accounted for approximately 18
hours of the day since an average of 10 hours a day were spent sleeping.

In the present study, time spent in television viewing and computer games
combined were 2.3, 2.7, and 2.0 hours for African, Indian and European boys,
respectively, and surpass the 2 hour maximum advocated by Bungum and Vincent
(1997). Bungum and Vincent (1997) found that television viewing of more than 2 hours
a day was negatively associated with physical activity. In the present study, there were

moderate negative correlations (-0.38, -O.40 and -0.30) between time spent on TV and

206

games, with sedentary, light and moderate physical activity, respectively. There is a high
moderate (r = -O.65) negative correlation between time spent on TV and games and total
energy expenditure, which does not change when the effects of race, SES and nutritional
status are partialled out. Additionally, there appears to be no signiﬁcant correlation
between the sum of skinfolds and BMI with television viewing (r = -0.01; and r = 0.00;
p>0.10), even when the effects of SES, race and nutritional status are removed
statistically.

The correlations between television viewing and estimated energy expenditure
and fatness are lower than those for a sample of 9-12 year old Canadian boys from
Quebec (Katzmarzyk, et al., 1998). The correlation between TV and MVPA and total
energy expenditure were -0.13, and -0. 12, respectively in Quebec boys. The correlations
between the sum of skinfolds and BMI were also low, -0.17 and 0.17, but signiﬁcant.
The amount of time spent in television viewing for Canadian boys thus has a weak
association with indicators of physical activity and health-related ﬁtness.

The difference in the association between television viewing and energy
expenditure and health-related ﬁtness of the Canadian and South African samples may be
due to lifestyle factors. South African boys watched more television (2.7 : 1.6 hours per
day) than the Canadian boys (2.0 : 0.9 hours per day). Furthermore, the negative
association between television viewing and estimated energy expenditure may have a
threshold effect, with a greater association after 2 hours or more of television viewing.

The limitations in the methodology employed to estimate energy expenditure and the

207

time spent watching television may also affect the association between television viewing
and energy expenditure, and health-related ﬁtness.

Attitudes surrounding sport and sport participation arellargely socialized.

Leonard (1982) identiﬁes the school, peer group, community and mass media as some of
the external agents of sports socialization. The mother, however, provides a greater
inﬂuence on the sport socialization of children (Baranowski et al., 1992). The role of the
school as a socializing inﬂuence is of particular importance in South Africa since there is
an absence of organized sport most schools. Also, some of the schools do not have PE
teachers and this reduces the opportunities for the school to provide the basic physical
activity skills and the development of physically active lifestyles. There is a need to
better understand the relationship between the environmental factors on the physical
activity levels of individuals. The effects of crime, crowding and lack of public spaces
on the physical activity patterns of children needs to be identiﬁed before plans to increase
activity levels can be implemented.

Habitual physical activity patterns of the subjects were also assessed using the
Physical Activity Questionnaire for Children (PAQ-C). There are no signiﬁcant race,
SES or nutritional status differences in the PAQ-C scores. Multiple regression analyses
of the PAQ-C score on the physical ﬁtness variables were also done (Table 73).
Additionally, to assess the effects of self-rated habitual physical activity during particular
segments of the day, multiple regression analyses of the habitual physical activity
variables of self-rated levels of activity during the physical education class, recess, lunch,

evenings and afterschool on physical ﬁtness were done (Table 74).

208

Table 73. Regression analyses of reported 7-day habitual physical activity (PAQ-C
score) on selected physical ﬁtness variables by race.

 

 

 

 

Dependent variable R2 B-valuell 13'” F
African
Sit-ups 0.05 2.53 0.23 6.60“
Indians
Push-ups 0.06 -4.0 0.25 4.50"I
'p<0.05 ”p< 0.01
“regression coefﬁcient b standardized regression coefﬁcient

° the PAQ-C represents the overall habitual physical activity score.

Habitual physical activity accounts for very little of the variation in physical ﬁtness,

explaining only 5% of the variance in the sit-ups of African boys and the 6% of the

variance in the push-ups of Indian boys (Table 73). The PAQ-c does not explain any of

the variance in other physical ﬁtness items for African and Indian boys. The PAQ-C also

does not contribute signiﬁcantly to the variance in the physical ﬁtness of European boys

and the total sample combined.

The self-reported habitual physical activity levels during speciﬁc segments of the

day has a slight effect on the physical ﬁtness of the present sample of boys. There is no

signiﬁcant association between physical ﬁtness and reported habitual physical activity of

the total sample (Table 74). However, the relationship between reported habitual

physical activity and physical ﬁtness is slightly higher for Indian compared to African

boys. Habitual physical activity in the evening accounts for 26%, with habitual physical

activity during the recess and lunch accounting for an additional 26% and 10%,

respectively of the variation in the sit-ups. performance of European boys. Habitual

physical activity accounts for a total of 52% of the variation in the sit-ups performance

209

Table 74. Regression analyses of reported habitual physical activity during speciﬁc
times of the day on selected physical ﬁtness variables by race.

 

 

 

 

Dependent variable Independent R2 B-value' [3" F
variable
Physical Activity
Mannie
dash, sec PE class 0.04 0.18 0.20 12.32"
Grip, kg Lunch 0.04 -0.63 -0.20 12.84"
Lunch and 0.06 -0.60 -0.19 9.35"
Evening -0.38 -0.13
Grip, kg/kg Evenings 0.01 0.01 -0.11 409*
Sit and reach, cm PE class 0.01 -0.55 -0. 12 4.15"
m
Dash, sec PE Class 0.06 0.21 -0.24 10.04“
Shuttle nm, sec Lunch 0.04 0.40 0.20 6.65""
Sit and reach, cm PE Class 0.03 -0.81 -0.17 4.76*
Push-ups, #/30 sec meh 0.04 -I.34 -0.20 6.55""
Nine minute run, In Lunch 0.03 -51.27 -0.18 5.14*
1.119399
Dash Recess 0.04 0.22 0.20 4..66*
Grip, kg Lunch 0.07 -0.88 -0.26 8.37"
meh and 0.11 -0.81 -0.24 6.71"
Evening -0.48 -0.20 6.71"
Shuttle run, sec PE class 0.11 -0.92 -0.33 14.00"
Standing long jump, cm Recess 0.10 -5.84 -0.31 12.26"
Emean
Sit-ups, #/30 sec Evening 0.26 -2.60 -0.51 8.66"
Evening and 0.42 -3.52 -0.69 8.23"
Recess 2.18 0.45
Evening, and 0.52 -3.77 -0.73 8.26"
Recess and 2.87 0.59
Lunch -1.59 -0.34
‘p<0.05 “p< 0.01
'regression coefficient b standardized regression coefficient

210

of European boys. The performance on some items of physical ﬁtness is thus associated
with the estimated level of habitual physical activity at speciﬁc times of the day.

Performance on tests of health- and performance-related ﬁtness may be affected
by activity, variety in the physical education curriculum, weekly activity time in physical
education, involvement in other physical activities, involvement in community
organizations (Dotson and Ross, 1985), and present level of physical activity (Dotson
and Ross, 1985; Shephard, 1982; Kemper and van Mechelen, 1995). Most schools in
South Africa have two 30-minute sessions of physical education per week. However, a
large number of schools do not have physical education teachers, and the activity during
physical education periods is unstructured and often includes supervised time in class.
Furthermore, most Government schools do not have gymnasiums, and this further
reduces the opportunities to participate in physical activity in inclement weather.

In the present sample ethnicity appears to play a major role in habitual physical
activity and anthr0pometry. Indian boys are the most sedentary, fattest and have the
poorest physical performance on most items compared to African and European boys,
whereas European boys are the least sedentary, taller and heavier, and have superior
performances on most of the physical ﬁtness tests. While there is no difference in the
PAQ-C scores of Africans, Indians and Europeans, Indian boys have the lowest
frequency of participation in most of the top ranked physical activities.

The poor physical ﬁtness of Indian boys in the present sample appears to be an
artifact of the low levels of physical activity of Indian boys. Indians have the highest

sedentary activity, the least light activity, and the lowest estimated total energy

211

expenditure compared to the African and European males. Indian boys engage in less
MVPA than African and European boys.

The correlations between estimated energy expenditure and physical ﬁtness,
however, are low and explain a minimal percentage of the variance (Table 75). Race-
speciﬁc regression analyses of estimated energy expenditure in sedentary, light, MVPA
and mean daily total estimated energy expenditure were also done (Table 76). Overall,
energy expenditure accounts for very little (1% to 2%) of the variation in the physical
ﬁtness scores of the total sample.

Estimated energy expenditure accounts for 4% to 6% of the variation in the
performance of Indian boys in some physical ﬁtness items. Thus, while Indian boys are
more sedentary and have lower estimated energy expenditure in MVPA, estimated
energy expenditure explains less of the difference in physical ﬁtness compared to
European boys. Indian boys who are more active at higher levels complete more sit-ups
and push-ups in 30 seconds compared to more sedentary boys. However, more sedentary
boys have better performances on the distance run. Differences in the distance run are
not associated with fatness since there is no signiﬁcant correlation (r = 0.08) between
estimated energy expenditure in sedentary activity and fatness (sum of ﬁve skinfolds).
Despite the higher estimated energy expenditure (mean daily total and MVPA) of
African boys, estimated energy expenditure differences do not account for variation in
physical ﬁtness, except on the dash.

Estimated energy expenditure accounts for minimal variance in physical ﬁtness of

African and Indian boys. Modiﬁcation of behavior patterns during the period of survey

212

Table 75. Correlations and partial correlations of health-related and performance-
related ﬁtness and indicators of activity.

 

 

 

Estimated energy expenditure (kcal/kg/day) Partial Correlations: Controlling for Race
Sedentary Light MVPA Total Sedentary Light MVPA Total
Health-related ﬁtness
Sit-ups 0.07 -0.07 0.12‘ 0.09 0.07 -0.07 0.12‘ 0.1 l
Push-ups -0. 11 -0.01 0.15‘ 0.09 -0.10 0.00 0.17“ 0.13‘
9 minute run 0.06 0.03 0.00 0.04 0.06 0.04 0.17 0.06
Sit and reach -0.1 1‘ 0.03 0.13‘ 0.10 -0.12‘ 0.00 0.10 0.06
2‘. of6 skinfolds -0.03 -0.15” -0.1 1‘ -0.1 1' -0.02 -0.l4‘ -0.10 -0.11
Performance related ﬁtness
Dash -0.11 -0.00 0.14“ 0.09 -0.11 -0.01 0.13“ 0.08
Grip -0.06 -0.01 -0.07 -0.10 -0.06 -0.01 -0.08 -0.1 1
Grip, kg/kg -0.09 -0.02 0.09 0.04 -0.09 -0.02 0.09 0.04
Shuttle run -0.03 -0.07 -0.07 -0.11‘ 0.03 -0.06 -0.06 -0.11
Standing long jump —0. 12 -0.02 0.04 ~0.01 -0.12‘ -0.01 0.05 0.00
FAQ-C 0.04 -0.02 0.1 l -0.10 0.04 -0.02 -0.1 1 -0.10

 

Signs for the dash and shuttle run are inverted since a lower score means a better performance.

Table 76. Regression analyses of estimated energy expenditure (kcal/kg/day) on
selected physical ﬁtness variables for the total sample, and by race.

 

 

 

Dependent variable Independent rtz a-vnlue‘ (1" r
var—table
M
Dash MVPA 0.02 -0.03 -0.13 4.88‘
Push-ups MVPA 0.02 0.24 0.16 7.57”
Shuttle nm MIDI-3‘ 0.01 0.05 0.11 4.05.
Sit and reach MVPA 0.02 0.15 0.13 5.41.
Sit-ups MVPA 0.02 o. 13 0.13 509°
Standing long jump sedentary 0.01 -1.24 -0.17 4.19‘
c m
Grip MTDE 0.17 .015 .042 5.40-
Grip, ltg/ltg sedentary 0.17 -0.01 .014 5.330
Pushups MTDI-z‘ 0.17 0.52 0.41 5.25-
Indian
Nine minute run sedentary 0.04 34.86 0.20 4.53-
Pushups MVPA 0.05 0.41 0.22 5.45.
Shuttle run MIDI-3‘ 0.05 -18.72 .022 5.78‘
Sit-ups MVPA 0.06 0.33 0.25 7.76“
African
Dash sedentary 0.05 .013 -022 8.34“
‘p<0.05 "p< 0.01
Note: signs for dash and shuttle run are inverted.
'regression coefﬁcient b standardized regression coefficient

° MTDE -- mean daily total estimated energy expenditure

213

and selective recall are other factors that could potentially affect accuracy. Accuracy of

the methodology is a possible confounder in this analysis.

214

Chapter 6

SUMNIARY AND CONCLUSIONS

This study considered the interrelationships am ong growth, physical ﬁtness,
habitual physical activity, SES, race and nutritional status of 389 ten year old South
African boys attending 18 schools in the greater Johannesburg area. Racial classiﬁcation
of subjects was done using anthroscopic appraisal. “Colored” boys were not included in
the sample. Boys of African (n = 208) and Indian (n = 137) ancestry were stratiﬁed into
higher (11 = 107, n = 99, respectively) and lower (n = 101, n = 38, respectively) SES
groups based on a composite SES score. There were proportionally more lower SES
African than Indian boys, and conversely more higher SES Indian than African boys. All
boys of European ancestry (n=42) were classiﬁed as higher SES. Nutritional status
classiﬁcation of undernourished (BMI-for-age s 15th percentile), nutritionally normal
(BMI-for-age > 15th but < 85th percentiles), and overweight (BMI-for-age .>. 85th
percentiles) was assigned based on the WHO (1995) criterion. Proportionally more
Indian boys were classiﬁed as undernourished and overweight compared to African boys.
Additionally, proportionally more African boys were classiﬁed as nutritionally normal
compared to Indian boys. A small number of European boys were classiﬁed as either

undernourished (n=3) or overweight (n=7).

215

Data for each subject included anthropometry (stature, mass, limb
circumferences, skinfold thicknesses, and skeletal breadths), physical ﬁtness (dash, grip,
shuttle run, standing long jump, nine minute run, push-ups, sit and reach, and sit-ups),
and physical activity (3-day physical activity records [Bouchard et al., 1983] and habitual
physical activity patterns [Physical Activity Questionnaire for Children, Kowalski et al.,
1997]). Sociodemographic data included demographic statistics, household size, and
income, occupation and education of parent/s or guardian/s.

The following research question was addressed: What is the relationship of
socioeconomic status, race, and nutritional status to growth status, physical ﬁtness and
habitual physical activity of 10-year-old South African boys?.

The growth status of 10 year old South African boys varies with SES, race and
nutritional status. Race accounts for a signiﬁcant but small proportion of the variation in
body proportions (sitting height/stature and bicristal/biacromial ratios), subcutaneous
fatness (sum of ﬁve skinfolds) and relative subcutaneous fat distribution (trunk-extremity
skinfold ratio, TER). Race differences remain signiﬁcant when the effects of SES and
nutritional status are controlled as covariates. European boys have proportionally shorter
legs relative to stature, and proportionally broader hips relative to the shoulders than
African or Indian boys. Further, Indian boys have proportionally broader hips relative to
the shoulders compared to African boys. Indian boys are fatter and have proportionally
more subcutaneous fat on the trunk compared to the extremities than African and
European boys. Additionally, Indian boys have larger estimated arm muscle areas

compared to African boys.

216

Body mass, the sitting height/stature ratio, fatness, relative subcutaneous fat
distribution, and estimated arm muscle vary signiﬁcantly with nutritional status, when
race and SES are used as covariates. Nutritional status may be confounded by race since
there are proportionally more overweight and undernourished Indian than African or
European boys. Overweight boys are obviously heavier than undernourished and
nutritionally normal boys, and nutritionally normal boys are heavier than undernourished
boys. There are no signiﬁcant SES, race, and/or nutritional status interactions for body
size. Overweight boys have larger accumulations of subcutaneous adipose tissue on the
trunk relative to the extremities and larger estimated arm muscle areas than
undernourished and nutritionally normal boys, while nutritionally normal boys have
more fat and estimated arm muscle than undernourished boys.

Variation in the bicristal/biacromial ratio, body mass and stature with SES are
signiﬁcant when the effects of race and nutritional status are statistically controlled.

This trend may also be confounded by race, since all European boys are higher SES.
Higher SES boys are taller and heavier, and have proportionally broader hips relative to
the shoulders compared to lower SES boys.

Variation associated with SES, race and nutritional status inﬂuences several of
the health- and performance-related physical ﬁtness tests of 10 year old South African
boys. Health-related ﬁtness (sit-ups, push-ups, 9 minute run, sit and reach, and the sum
of 5 skinfolds) and performance-related ﬁtness (dash, shuttle run, grip strength, and
standing long jump) were considered separately. SES and race do not contribute
signiﬁcantly to the variation in sit-ups, while nutritional status and SES do not contribute

signiﬁcantly to variation in the push-ups, distance run and sit and reach.

217

There is signiﬁcant racial variation in health-related ﬁtness (push-ups, distance
run, sit and reach, sum of skinfolds) when the effects of SES and nutritional status are
controlled. European boys have better ﬁtness on the push-ups and 9 minute run
compared to African and Indian boys, who do not differ. However, African boys have
greater ﬂexibility of the lower back on the sit and reach compared to Indian boys.

There is no signiﬁcant SES variation in health-related ﬁtness, but health-related
ﬁtness varies signiﬁcantly with nutritional status. Nutritionally normal boys have better
ﬁtness on the sit-ups and sit and reach compared to undernourished boys. Overweight
boys are less ﬁt on the sit-ups and 9 minute run compared to undernourished or
nutritionally normal boys. Additionally, overweight boys have less muscular strength
and endurance of the arms and shoulder girdle as measured by push-ups compared to
nutritionally normal boys.

SES, race and nutritional status do not contribute signiﬁcantly to the variation in
two performance-related ﬁtness items, the shuttle run and standing long jump, and SES
and race do not contribute signiﬁcantly to variation in the dash. Race-speciﬁc variation
in the dash and grip strength per unit mass are signiﬁcant. Indian boys are the most unﬁt
on the dash and grip strength per unit mass. African and European boys have greater
speed in the dash compared to Indian boys, and African boys have greater grip strength
per kilogram body mass compared to Indian boys.

Higher SES boys have greater absolute grip strength compared to lower SES
boys. This, however, is an artifact of the larger body size of higher SES boys, since the

difference in grip strength per unit body mass is not signiﬁcant.

218

Overweight boys have poorer performances on the 50 meter dash compared to
undernourished and nutritionally normal boys, but have greater absolute grip strength
compared to undernourished and nutritionally normal boys. Nutritionally normal boys
have better absolute grip strength than undernourished boys. However, the sequence is
reversed for grip strength per unit body mass: undernourished are greater than
nutritionally normal who are greater than overweight.

SES, race and nutritional status account for a small portion of the variance in the
estimated habitual physical activity of 10 year old South African boys. Habitual physical
activity was assessed as the time spent on light activity (3 to 5 METS) and MVPA (6 to 9
METS), and mean estimated total daily energy expenditure was also considered.
Nutritional status accounts for 41% of the variation in relative estimated energy
expenditure (kcal/kg/day) in sedentary activity. SES accounts for an additional 2% of
the variation in sedentary activity, while race accounts for 2% of the variation in MVPA
(kcal/kg/day).

Racial variation in estimated energy expenditure in sedentary and light activity
(kcal/day) is confounded by SES and nutritional status. Racial variation in estimated
sedentary and light energy expenditure (kcal/day) is not signiﬁcant when the effects of
SES and nutritional status are controlled. However, Indian boys are signiﬁcantly less
active in MVPA (kcal/day and kcal/kg/day) compared to African and European boys.

SES variation in estimated sedentary energy expenditure (kcal/day) is confounded
by body size. The differences between SES groups are not signiﬁcant when expressed
per unit body mass. Similarly, estimated sedentary, light and total energy expenditure

(kcal/day) do not differ among nutritional status groups when expressed per unit mass.

219

There are few differences in the growth status and physical ﬁtness of 10 year old
South African boys with different levels of energy expenditure. There are signiﬁcantly
more African boys in the highest quartiles of estimated energy expenditure than Indian
boys. Conversely, there are signiﬁcantly more Indian boys in the lowest quartiles of
MVPA and mean daily total estimated energy expenditure. Thus, some of the variation
in growth status and physical ﬁtness between boys in the highest and lowest quartiles of
estimated energy expenditure probably reﬂects the racial composition of the quartiles.

The least active African boys are taller and heavier with a higher BMI , and have
a higher absolute grip strength than the most active African boys. Additionally, African
boys who engage in the most MVPA are shorter and complete more sit-ups in 30 seconds
compared to African boys who engage in the least MVPA. This, however, is a function
of their larger body since grip strength per unit mass does not differ between boys at the
extremes of mean daily total estimated energy expenditure (kcal/kg/day).

Similarly, the least active Indian boys are taller and heavier, and have
proportionally longer legs relative to stature compared to the most active boys. There is
no difference in the growth and physical ﬁtness of European boys at the extremes of
mean daily total estimated energy expenditure, but European boys in the lowest quartile

of MVPA are taller than European boys in the highest quartile of MVPA.

mam-n

Race, SES and nutritional status account for generally small but signiﬁcant
proportions of the variance in measures of the growth status of 10 year old South African

boys. The following are the more consistent differences observed in this study:

220

0 Higher SES boys are taller and heavier than lower SES boys when the effects of

race are controlled.

0 European boys are taller and heavier than African and Indian boys when the

effects of SES are controlled.

0 Overweight boys are taller and heavier than nutritionally normal or

undernourished boys.

0 Indian boys have more subcutaneous fat but have a lower BMI than African and
European boys.
Race, SES and nutritional status account, in general, for small but signiﬁcant
proportions of the variances in measures of health- and performance-related physical
ﬁtness and estimated energy expenditure of 10 year old South African boys. The

following trends are suggested by the results of the analyses:

0 African boys are more ﬂexible on the sit and reach, while European boys
complete more push-ups. Indian boys have the poorest performance on most

physical ﬁtness items.

0 Undemourished boys have better physical ﬁtness on most test items compared to

overweight boys.

0 Indian boys engage in the most sedentary activity and have the lowest estimated
light and MVPA energy expenditure (kcal/kg/day). Estimated energy expenditure

differences between African and European boys are not signiﬁcant.

0 The most physically active African and Indian boys are taller and heavier than the

least active African and Indian boys, respectively.

221

Recommendations.

People of African, Indian and European ancestry in South Africa are not
homogenous groups. Thus language and cultural differences that are prevalent within
racial groups may have inﬂuenced the growth and physical ﬁtness of individuals. The
ethnicity (linguistic and cultural groupings) of subjects was not recorded in the present
study. The multiethnic nature of South African society must be taken into account for
future studies. Additionally, the cultural practices and speciﬁcally their signiﬁcance for
physical activity and physical ﬁtness need to be documented.

A more objective method of estimating energy expenditure, such as heart rate
recordings or motion sensors, may provide more accurate assessments. These should be
complemented by more culturally sensitive observational studies. '

There is a paucity of information on the growth, physical ﬁtness, habitual
physical activity, and the cultural context of activity and ﬁtness of South Africa Indians.
Research should be directed towards building up a database of information on these
growth, physical ﬁtness and habitual physical activity, and in particular to study the
signiﬁcance of high subcutaneous fatness in association with a low BMI in individuals of

Indian ancestry.

222

APPENDICES

223

APPENDIX A

University Committee on Research Involving Human Subjects

224

 

OFFICE OF
RESEARCH
AND
GRADUATE
STUDIES

riversity Committee on
Research Involving
Human Sublects
(UCRIHS)

Mnchigan State University
t6 Administration Building
East Lansing. Michigan
48824-1046

517/355-2180
FAX1517/432-1171

‘t‘e Mtrn-gan State Un/veISIty
DEA IS Institutional DwerSIty'
Excellence In ACIIOII

MSU Is an aﬁiImaIIl/e-BCIIOII,
mt-mﬁrrmw mthh mm

MICHIGAN STATE

 

UNIV

ERSITY

March 11, 1997

TO: Robert M. Malina
213 IM Sports Circle

RE: IRB#: 97-094

TITLE: THE EFFECTS OF SOCIOECONOMIC STATUS, ETHNICITY
AND NUTRITIONAL STATUS ON THE GROWTH AND
PHYSICAL FITNESS OF 10 YEAR OLD SOUTH AFRICAN
BOYS

REVISION REQUESTED: N/A

CATEGORY: FULL REVIEW

APPROVAL DATE: 03/10/97

The University Committee on Research Involving Human Subjects'tUCRIHS)
review of this project is complete. I am pleased to adVise that the

rights and

welfare of the human subjects appear to be adequately

rotected and methods to obtain informed consent are appropriate.

herefore,
above.

RENEWAL:

REVISIONS:

PROBLEMS/
CHANGES:

the UCRIHS approved this project and any reVisions listed

UCRIHS approval is valid for one calendar year, beginning with
the approval date shown above. Investigators planning to
continue a project beyond one year must use the green renewal
form (enclosed with t e original agproval letter or when a.
project is renewed) to seek u date certification. There is a
maXimum of four such expedite renewals ossible. Investigators
wishing to continue a prOject beyond the time need to submit it
again or complete reView.

UCRIHS must review any changes in procedures involving human
subjects, rior to initiation of t e change. If this is done at
the time o renewal, please use the green renewal form. To
revise an approved protocol at an other time during the year,
send your written request to the CRIHS Chair, requesting revised
approval and referencing the project's IRB # and title. Include
in your request a description of the change and any revised
instruments, consent forms or advertisements that are applicable.

Should either of the following arise during the course of the
work, investigators must noti UCRIHS promptly: (1) roblems
(unexpected Side effects, comp aints, etc.) involving uman
subjects or (2) changes in the research environment or new
information indicating greater risk to the human sub ects than
existed when the protocol was previously reviewed an approved.

If we can be of any future helpé please do not hesitate to contact us

at (517)355-2180 or FAX (517)4

Sincerely,

   
 

DEW:bed

cc: Reshma

1171.

vid E. Wright, Ph
CRIHS Chair

B. Naidoo

225

APPENDIX B

Consent forms (parents)

226

The Eﬂects of Socioeconomic Status, Ethnicity, and Nutritional status on the Growth and
Physical Fitness of 10 year old South African Boys

CONSENT FORMS (PARENTS)

We are undertaking a study aimed at identifying the environmental factors that
affect the growth and physical ﬁtness of children in the Johannesburg area. This study
will form the dissertation component of Reshma Naidoo's doctoral program at the
Michigan State University, East Lansing, Michigan, USA.

Your son/ward has been chosen as a subject for this study as he is a lO-year-old
male attending school in the Johannesburg area. Five hundred lO-year-old boys
attending schools in the Johannesburg area will be used as subjects for this study. The
participation of your son/ward is voluntary and your son/ward has the right to decide
whether to participate or not.

If you agree to allow your son/ward to participate in this study, he will have to
complete the following activities. The researcher will interview your son/ward to list the
physical activity performed that your son/ward for the past 24 hours. This will be done
on three successive days, taking about 10 minutes each day. Additionally, he will have
to complete a physical activity questionnaire in which he will record his physical activity
patterns for the last week (7 days).

The following tests will be conducted during school time over two days, and will
involve a total time of approximately 90 minutes. Some of the boys will be asked to take
these tests for a second time, a week later. If your son/ward is selected to take the retest,
it will take another 90 minutes (over two days) to complete. His weight, height, breadth

of his shoulders and hips will be measured. These measurements will be taken using

227

special instruments. This is completely painless and will not hurt your son/ward in any
way. The amount of muscle and fat in the body will be measured using skinfolds and
circumference. A skinfold is taken by pinching the Skin and measuring the thickness of
the skin that is lifted away from the body. This is completely harmless and will cause
little pain to your son/ward. Skinfold measurements will be taken at the upper arm,
shoulder, stomach, waist, hip and calf.

He will have to perform tests that will determine his physical ﬁtness. These tests
will not indicate how good he is at sports. The test results will only be meaningful to the
tester when used with the results of all other subjects. He will have to run as far and as
fast as he can, continuously, for 9 minutes. He will do as many sit-ups and push-ups as
he can for 30 seconds each. He will run 50 meters as fast as he can, and run up and down
a lO-meter area around two cones 5 times. His left and right grip strength will be
assessed with a handgrip dynamometer (a machine). His sit-and-reach ﬂexibility will be
measured.

Although your son/ward will be expected to work very hard, none of these tests
are dangerous and every effort will be taken to ensure that no harm comes to him.
However, he may be sore and stiff after the physical performance tests if he does not
usually exercise this hard. In order to minimize the discomfort, he will be thoroughly
warmed-up before and cooled down after the physical performance testing.

Since these tests will cut into the school day, every effort will be made to
minimize the loss of academic teaching time, and children will be scheduled at times that

are convenient for their teachers and academic schedule. Additionally, a researcher may

228

visit your home and conduct an interview with you. The interview will be used to gain
information about your family.

All information and data gathered will be treated with the strictest of conﬁdence
and will not be used except for the purposes stated above. The conﬁdence of subjects
will be maintained at all times, but may be Shared with your, the parent or guardian
Should your child wish. All participants will remain anonymous in any report of the
research ﬁndings.

These tests involve no ﬁnancial costs to you or your family. All materials needed
for the testing will be provided. However, on the days of testing, your son/ward will
need to be dressed in his physical education outfit that will allow him to perform the tests
that will be conducted.

If you have any questions regarding this project, you can contact Reshma N aidoo
at 67 Somerset Street, Kensington, 2001 (Phone: 011-614-8610), or Dr. Robert M.
Malina at Institute for the Study of Youth Sports, Michigan State University, East
Lansing, Michigan 48825 (Phone: 517-355-7620).

You may keep the above portion of the form and return only the bottom portion.

Please Sign and return this form as soon as possible. Thank you very much.

Yours Sincerely

 

Reshma Naidoo

229

The Eﬂects of Socioeconomic Status, Ethnicity, and Nutritional status on the Growth and
Physical Fitness of 10 year old South African Boys

CONSENT FORM

I have decided to allow my son/ward

 

to participate in a study of the factors affecting grth and physical ﬁtness of children
that will be conducted in his regularly scheduled school day. My signature indicates that
I have read the above information, or have had someone read the above information to
me, and have given permission for my son/ward to participate in the study. My
son's/ward's signature indicates that he understands that a study will be conducted at
school and agrees to participate in it. I understand that my son/ward may withdraw at
any time (I can choose to discontinue my son/ward's participation in the study) without
any negative consequence to me or my family in my present or future relations with
Michigan State University.

I understand that if my child is injured as a result of my participation in this
research project, Michigan State University will provide emergency medical care if
necessary. I further understand that if the injury is not caused by the negligence of MSU
I (the parent/ward) am personally responsible for the expense of this emergency care ad

any other medical expenses incurred as a result of this injury.

 

 

 

 

 

 

Signature of parent! guardian Date
Signature of child Date
Signature of investigator Date
Please check here if you would like to keep a copy of this form for your records

Please sign and return this page.

230

APPENDIX C

Assent form for children

231

The Effects of Socioeconomic Status, Ethnicity, and Nutritional status on the Growth and
Physical Fitness of 10 year old South A ﬁican Boys

ASSENT FORM FOR CHILDREN

I agree to take part in a study that will address the factors that affect the growth
and physical ﬁtness of children. This study has been explained to my mother/father/
guardian and that she/he has given his/her permission for me to participate in this study.
I understand that I may decide that I do not want tocontinue with this study at any time
and I can stop at any time I want to. Any information about what I say or do will not be
given to anybody else, and will only be used for the purpose of this study.

I understand that I will be asked to complete many physical ﬁtness tests to see
how strong and fast I am and that these tests require me to make my best efforts. I also
will have to report my physical activity for three days, which includes all the Sports and
games I play as well as all the running that I do as well as ﬁll in questionnaire about my
physical activity over the last 7 days. The tester will take measurement of the length,
weight, skinfolds and breadths of my body. I understand that nothing bad or wrong will
happen to me if I decide to stop my participation in this study at any time.

When I Sign my name to this page, I am indicating that this page was read to me
and that I am agreeing to participate in this study. I am indicating that I understand what

will be required of me and that I may stop participating in the Study at any time.

 

 

Child's Signature Date

 

 

Signature of principal investigator Date

232

APPENDIX D

Sociodemographic information questionnaire (parents)

233

The Effects of Socioeconomic Status, Ethnicity, and Nutritional status on the Growth and
Physical Fitness of 1 0 year old South African Boys

SOCIODEMOGRAPHIC INFORMATION QUESTIONNAIRE (PARENTS)

l. Subject’s name:

 

 

 

2. Name:
3. Relationship to child: Please check one of the followin columns.
Mother father aunt uncle brother sister grandmother grandfather other

 

 

 

 

 

 

 

 

 

 

 

If you fall into the “other” category, please ﬁll

in your relationship to the child

 

4. Home Address:

 

 

5. Which of the following best describes the educational level of the head of the household

(male and female ?

Male father Female

1.
2.
3.
4.
5.
6.
7.
8.
9.

6. I have (ﬁll in the number)

 

Cannot read and write
Can read and write but never went to school
Some school
school
school
school

Some

Technical
or
Master or doctoral

Ol'

 

 

 

 

 

 

 

sons daughters
7. How many people live in the same house with you?
8. How many rooms are there is your house?
9. How many adult live in the same house as you do?

10. How many children live in the same house as you?

 

 

 

 

234

 

l 1. Some examples of the different occupations given below. Please indicate which of the
following best indicates your occupation?

Mother (female head of household):

Father (male head of household):

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

College/ President] City council or Medical doctor Radio-television
University Director/ local authority and ﬁlm
principal Chairperson of member industry actor
company
College/ Manager Bank - clerical or Nurse Chef
University lower
teacher management
School principal/ Sales F ireman/ Allied health Janitorial/
Deputy or Vice policeperson/ worker custodial
Principal military services
School teacher Clerical/ Technician Postal service Hairdresser/
Secretarial barber/
beautician
Artist/actor/ Cashier Trader/ Taxi owner Nanny/
Musician merchant caregiver
Engineer/ Factory worker Apprentice Driver/delivery Domestic
architect person laborer
Lawyer/judge Cook/baker Laborer Farmer Waiter/waitress/
bartender
Miner Housekeeper/ Gatekeeper/ Other Gardener
Maidservice Doorman/ Butler

 

12. Please indicate how much you earn each month. Circle the number that reﬂects your

income.

 

1 less than R417 a month

4 R 1,668 to R3,334Jer month

 

2 R417 to R833 per month

5 R3,335 or R6,667 per month

 

 

 

3 R834 to R1,667 a month

 

 

6 R 6,668 per month and more

 

 

13. How much does your spouse (or other contributing adult) earn per month? Please circle the

number that reﬂects his/her income.

 

1 less than R417 a month

4 R 1,668 to R3,334 per month

 

2 R417 to R833 per month

5 R3,335 or R6,667 per month

 

 

 

3 R834 to RL667 a month;

 

 

6 R 6,668 per month and more

 

 

15. How many times has your child been sick in the past year/

 

16. Does your child have intestinal worms (such as round worms or tape worms)?

 

17. Has your child been treated for intestinal worms in the past year?

 

18. Did your child have any intestinal parasites in the past year?

 

Thank you for your assistance in this project

235

 

APPENDIX E

Physical Activity Record

236

The Effects of Socioeconomic Status. Ethnicity, and Nutritional status on the Growth and
Physical Fitness of l 0 year old South African Boys

PHYSICAL ACTIVITY RECORD

Name:

 

II) Number:

 

Date and day of activity record:

 

This part of the testing process is going to be used to account for your daily
activity level. You are going to have to ﬁll in Similar forms for three days in a row
(Monday, Tuesday, and Wednesday).

For the purposes of recording, each day is divided into before and after breakfast,
school, afterschool, before and after supper and before bed. You have to ﬁrst recall what
you did for that segment of time. Write it down and indicate approximately how long
you spent on the activity.

When you have performed more than one activity during that time period, use the
activity that was most performed during that time. Also indicate if you hyperventilated
(breathed heavily) and/or sweated during the activity. Indicate how hard you think that
you were working for each of the activities.

If there are any doubts or questions, you can direct them to me (the interviewer)
when you are ﬁlling in the form.

Thank you very much for helping with this project.

237

EXAMPLE

For each activity indicate if you sweated, and breathed hard. For each note list a
Leave the Code column blank

little, not too much, a lot.

 

 

 

In; Mm lies sternum
got up 6.30 am
before showered and dressed 6.30 - 7.00 easy
breakfast.
breakfast 7 - 7.15 am. light
after breakfast. helped to clean up 7.15-7.30 am sweated
walking to school 7 .30-7.50 sweat! breathing hard
before school played soccer with friends 7.50-8.05 sweating lbreathing
hard
ﬁrst period sat - English 8,05 - 8.35
second math 8.35-9.05
third science 9.05-9.35
ﬁrst break played soccer 9.35-9.50 sweated/breathing hard
fourth history 950-1020
ﬁfth gymnastics in PE class 10.20-10.50 breathing hard
sixth period geography 10.50-11.20
second break lunch: played soccer 11.20-11.50 sweated/breathing hard
seventh period sat in library class 11.50-12.20
eight period stood in music class 12.20-12.50
ninth period drew in art class 1250-1.20
right living 1.20-1.30
after school walked home 1.30-1.45 sweated/ breathing very
watched TV 1.45-3.30 hard
played ball with friends 3.30 -5.30 sweated/ breathing hard
before supper cleaned up the house 5.30 - 6.30 sweated! breathing hard
helped with the laundry
supper had dinner 6.30 - 7.00
after supper cleared up the kitchen 7.00-7.30 sweated
did homework 7.30 - 8.30
before bedtime bath 8.30-10.00
dressed
read
bedtime slept 10.00

 

238

bhune: II)# IDay:

 

 

 

 

 

Monday
For each activity indicate if you sweated, and breathed hard. For each note list a

little, not too much, a lot. Leave the Code column

blank
Activity Amount of time Effort Code
got up at

Before breakfast

Breakfast

After breakfast

 

Before lunch

 

During lunch

 

After lunch

 

Before supper

 

During supper

 

After supper

 

Before bedtime

 

 

Bedtime

 

 

 

 

 

239

 

Name: ID # Day: Tue/W ed

For each activity indicate if you sweated, and breathed hard. For each note list a

 

 

 

 

 

little, not too much, a lot. Leave the Code column blank

Activity Amount of Effort Code
time

Got up at

Before

breakfast

Breakfast

After breakfast

 

Before school

 

First period

 

Second period

 

Third period

 

First break

 

Fourth period

 

Fiﬁh period

 

Sixth period

 

Second break

 

Seventh period

 

Eight period

 

Ninth period

 

After school

 

Before supper

 

During supper

 

After supper

 

 

 

 

 

 

Bed

 

Leave this section blank. It is for use by the investigator.

Summary:

 

1 2 3 4 5 6 7 8 9

 

 

 

 

 

 

 

 

 

 

 

 

 

APPENDIX F

Habitual Physical Activity Questionnaire

241

The Effects of Socioeconomic Status, Ethnicity, and Nutritional status on the Growth and
Physical Fitness of 10 year old South African Boys

HABITUAL PHYSICAL ACTIVITY QUESTIONNAIRE

Name: ID #:

Schook

 

We are trying to ﬁnd out about your physical activity levels that you have done in the
last 7 days (in the last week). This includes sports or dance that make you sweat or make your
legs feel tired, or games tat make you breathe harder, like tag, skipping, running, climbing and
others.

Remember: There are no right or wrong answers - this is not a test.
Please answer all the question as honestly and accurately as you can - this is very
important.
1. In the last 7 days, during your physical education (PE) classes, how often were you very
active (playing hard,
I don t do PE
ever

Sometimes
often

 

2. In the last 7 days, what did you do most of the time AT RECESS?
Sat down (Talking, reading, doing school work)
Stood around or walked around

Ran or played a little bit

Ran and played hard most of the time

 

 

 

 

 

 

 

 

3. In the last 7 daysghat did you do normally do AT LUNCH (besides eating lunch)?
Sat down (T alkingggeading, doing school work)
Stood around or walked around

Ran or played a little bit

Ran and played hard most of the time

 

 

 

 

 

 

 

 

4. In the last 7 days, on how many days RIGHT AFTER SCHOOL, did you do sports, dance,
playl games, or hard work in which you were very active?

None

1 time last week

2 or 3 times last week

4 times last week

5 times last week

 

 

 

 

 

 

 

 

 

242

5. In the last 7 days, on how many days EVENINGS, did you do sports, dance, play games, or
do hard work in which you were very active?

None

1 time last week

2 or 3 times last week

4 or 5 times last week

6 or 7 times last week

 

 

 

 

 

 

 

 

 

6. ON THE LAST WEEKEND, how many times did you do sports, dance, play games, or do
hard work in which you were very active

None

1 time last week

2 or 3 times last week

4 or 5 times last week

6 or 7 times last week

 

 

 

 

 

 

 

 

 

Which ONE of the following describes you best for the last 7 days?

 

>9

All or most of in the time that l was not at school was spent doing things that
involve little physical eﬂ‘ort (e.g. watching TV, doing homework, playing
computer games)

 

I sometimes (1 - 2 times last week) did physical things in the time that I was not
at school (C HIayed sports, went running, swimminngardeninngashed the car)

 

 

I often (3 - 4 times last week) did physical things in the time that I was not at
school

 

I quite often (5 -6 times last week) did physical things in the time that I was not at
school

 

 

mpow

I often ( 7 or more times last week) did physical things in the time that I was not at
school

 

 

 

8. How fit do think are to other of
V ﬁt
Fitter than most

About

              

 

Less ﬁt than most
V unfit

9. Were you sick last week, or did anything prevent you from doing your normal physical
activities?
Yes
No

 

 

 

 

 

 

If yes, what prevented you?

 

243

 

BIBLIOGRAPHY

244

 

Reference List

Adams EJ, Stephenson LS, Latharn MC and Kinoti SN (1994) Physical activity and
growth of Kenyan school children with hookworm, Trichuris trichiura and Ascaris
lumbricoides infections are improved after treatment with albendazole. Journal
of Nutrition 124: 1199-1206.

Allen LH (1990) Functional indicators and outcomes of undernutrition. Journal of
Nutrition 120: 924-932.

Allen LH (1993) The Nutrition CRSP: what is marginal malnutrition, and does it affect
human function? Nutrition Reviews 51: 255-267.

Allen LH (1994) Nutritional inﬂuences on linear growth: a general review. European
Journal of Clinical Nutrition 48: S75-S89.

Allen LH and Uauy R (1994) Guidelines for the study of mechanisms involved in the
prevention or reversal of linear growth retardation in developing countries.
European Journal of Clinical Nutrition. 48: S212-216.

American Alliance for Health PE, Recreation and Dance (1984) Technical manual, health
related physical ﬁtness. Reston, VA: American Association for Health, Physical
Education, Recreation and Dance.

Appleton CC and Bailey IW (1990) Canoeists and waterborne diseases in South Africa.
South African Medical Journal 78: 323-326.

Archer R and Bouillon A (1982) The South African Game. London: Zed Press.

Ashcroﬁ MT and Desai P (1976) Ethnic differences in growth potential of children of
African, Indian, Chinese and European origin. Transactions of the Royal Society of
Tropical Medicine and Hygiene 70: 433-438.

Astrand PO (1952) Experimental studies of working capacity in relation to sex and age.

Copenhagen: Munksgard.

Badenhorst ML, Peters EM and Ash J (1992) Maximal work capacity and dietary status
of rural black South African schoolboys ages 10-14 years. South African Medical
Journal 18: 504-508.

Barac-Nieto M, Spurr GB, Maksud MG and Lotero H (1978) Aerobic work capacity in
chronically undernourished adult males. Journal of Applied Physiology:
Respiration, Environment and Exercise Physiology 44: 209-215.

Baranowski T, Bouchard C, Bar-Or O, Bricker T, Heath G, Kimm SYS, Malina R,
Obarzanek E, Pate R, Strong WB, Truman B and Washington R (1992)
Assessment, prevalence, and cardiovascular beneﬁts of physical activity and
ﬁtness in youth. Medicine and Science in Sport and Exercise 24: $237-$247.

Baranowski T, Dworkin RJ, Cieslik CJ, Hooks P, Clearman DR, Ray L, Dunn JK and
Nader PR (1984) Reliability and validity of self reports of aerobic activity:
Family Health Project. Research Quarterly for Exercise and Sport 55: 309-317.

245

e'ﬁ

Baranowski T, Thompson WO, DuRant RH, Baranowski J and Puhl J (1993)
Observations on physical activity in physical locations: age, gender, ethnicity, and
month effects. Research Quarterly for Exercise and Sport 64: 127-133.

Bar-Or O (1983) Pediatric Sports Medicine for the Practitioner. NY: Springer-Verlag.

Beneﬁce E (1992) Physical activity and anthropometric and ﬁrnctional characteristics of
mild malnourished Senegalese children. Annals of Tropical Paediatrics 12: 55-66.

Beneﬁce E (1993) Physical activity, cardiorespiratory ﬁtness, motor performance, and
growth of Senegalese pre-adolescents. American Journal of Human Biology 5:
653-667.

Beneﬁce E and Malina RM (1996) Body size, body composition and motor
performances of mild-to—moderately undernourished Senegalese children. Annals
of Human Biology 23: 307-321.

Berg A (1973) The Nutrition Factor: Its Role in Development. Washington DC.: The
Brookings Institute.

Berkey CS, Dockery DW and Ferris BG (1994) Adolescent height growth in US
children. Annals of Human Biology 21: 435-422.

Berkowitz FE, Vallabh W, Bquana A and Heney C (1989) Cryptosporidiosis in black
South African children. South African Medical Journal 74: 272-273.

Beunen G, Malina RM, Ostyn M, Renson R, Simons J and van Gerven D (1983)
Fatness, growth and motor ﬁtness of Belgian boys 12 through 20 years of age.
Human Biology 55: 599-613.

Beunen GP, Malina RM, Van't Hof MA, Simons J, Ostyn M, Renson R and Van
Gerven DC (1988) Adolescent Growth and Motor Performance: A Longitudinal
Study of Belgian Boys. Champaign, IL: Human Kinetics.

Blair SN (1984) How to assess exercise habits and physical ﬁtness. In J Matarazzo, S
Weiss, JA Herd, N Miller and S Weiss (eds): Behavioral Health: A Handbook
of Health Enhancement and Disease Prevention. NY: John Wiley and Sons, pp.
424-447.

Bogin B, Wall M and MacVean RB (1992) Longitudinal analysis of adolescent growth
of ladino and Mayan school children in Guatemala: Effects of environment and
sex. American Journal of Physical Anthropology 89: 447-457.

Botha JF, Clarke D and Jokl E (1945) Medical research in Physical Education: health
and efficiency. South Aﬁican Medical Journal 19: 381-383.

Bouchard C and Shephard RJ (1994) Physical Activity, Fitness and Health: The model
and key concepts. In C Bouchard, RJ Shephard and T Stephens (eds.): Physical
Activity, Fitness and Health: International Proceedings and Consensus Statement.
Champaign, IL: Human Kinetics, pp. 77-88.

Bouchard C, Tremblay A, Leblanc D, Lortie G, Savard R and Theriault G (1983) A
method to assess energy expenditure in children and adults. American Journal of
Clinical Nutrition 37: 461-467.

Branta CF and Goodway JD (1996) Facilitating social skills in urban school children
through physical education. peace and conﬂict: Journal of Peace Psychology 2:
305-319.

Branta CF, Haubenstricker J and Seefeldt V (1984) Age changes in motor Skills during
childhood and adolescence. Exercise and Sport Science Reviews, pp.267-320

246

.-l
,‘J

i'.

Bungum TJ and Vincent ML (1997) Determinants of physical activity among female
adolescents. American Journal of Preventive Medicine 13: 115-122.

Burger PJ (1968) Die voorkoms van intestinale parasiete by skoolkinders in die
Tygerbergse omgewing. South Aﬁican Medical Journal 42: 811-812.

Cameron N (1991) Human growth, nutrition, and health status in sub-Saharan Africa.
Yearbook of Physical Anthropology 34: 211-250.

Cameron N, Grieve CA, Kruger A and Leschner KF (1993) Secondary sexual
development in rural and urban South African black children. Annals of Human
Biology 20: 583-593.

Cameron N and Kgarnphe J S (1993) The growth of South African rural black children.
South African Medical Journal 83: 184-190.

Cameron N, Kgamphe J S, Leschner KF and Farrant PJ (1992) Urban-rural differences
in the growth of South African black children. Annals of Human Biology 19: 23-
33.

Cameron N, Kgarnphe J S, Lunz R and Knight S (1988) Single year height and weight
velocities of South African rural black children. South African Journal of Science
84: 5 1 9-5 20.

Cameron N, Mitchell J, Meyer D, Moodie A, Bowie MD, Mann M D and Hansen JDL
(1988) Secondary sexual development of 'Cape Coloured' girls following
kwashiorkor. Annals of Human Biology 15: 57-60.

Cameron N, Mitchell J, Meyer D, Moodie A, Bowie MD, Mann MD and Hansen JDL
(1990) Secondary sexual development of Cape Coloured boys following
kwashiorkor. Annals of Human Biology 17: 217-228.

Carey SM (19 80) Aspects of the anthropometric growth and development of
Johannesburg children in two racial groups. Anthropology 264.

Cemerud L (1993) The association between height and some structural social variables:
a study of lO-year-old children in Stockholm during 40 years. Annals of Human
Biology. 20: 469—476.

Charting-Pearce SM and Solomon L (1987) Pubertal development in black and white
Johannesburg girls. South African Medical Journal 71: 22-24.

Charting-Pearce SM and Solomon L (1986) A longitudinal study of height and weight in
Black and White Johannesburg children. South African Medical Journal 70: 743-
746.

Chase HP and Martin HP (1970) Undemutrition and child development. New England
Journal of Medicine 282: 933-939.

Chavez A and Martinez C (1979) Consequences of insufﬁcient nutrition on child
character and behavior. In DA Levitsky (ed.): Malnutrition Environment and
Behavior. Ithaca, NY: Cornell University Press, pp. 23 8-355.

Chavez A and Martinez C (1984) Behavioral measurements of activity in children and the
relationship to food intake in a poor community. In E Pollit and P Amante (edS.):
Energy intake and activity. NY.: Liss Inc, pp. 303-321.

Chrysler F und/Amateur Athletic Union (1993-1994) Testing Packet: Amateur Athletic
Union Physical Fitness Program. Bloomington IN: Amateur Athletic Union .

Clarke HH (1971) Physical and Motor Tests in the Medford Boy‘s Growth Study.
Englewood Cliffs, NJ: Prentice-Hall.

247

Clarke HH (1976) Application of Measurement to Health and Physical Education. NJ:
Prentice-Hall.

Cohen J (1988) Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ:
Lawerence Erlbaurn Associates.

Connolly KI and Kvalsvig JD (1993) Infection, nutrition and cognitive performance in
children. Parasitology 107: 8187-8200.

Corlett JT (1986) Growth of urban schoolchildren in Botswana. Annals of Human
Biology 13: 73-82.

Corlett JT and Woollard E (1988) Growth patterns of rural children in the Kgalagadi
region of Botswana. Annals of Human Biology 15: 153-159.

Cronk CE and Roche AF (1982) Race- and sex-speciﬁc reference data for triceps and
subscapular skinfolds and weight/stature? American Journal of Clinical Nutrition
35: 347-3 54.

Culver EH, de Jongh TW and Jokl E (1942) Comparison between physical efﬁciency
standards of Bantu, Chinese, Coloured, European and Indian school children.
Manpower (South Aﬁica) 1: 39-54.

Daniels J, Oldridge N, Nagle F and White B (1978) Differences and changes in V02
among young rrmners 10 to 18 years old. Medicine and Science in Sports 10: 200-
203.

de Villiers FPR (1987) The growth pattern of adolescent Tswana schoolchildren. Journal
of Tropical Pediatrics 33: 143-152.

Department of National Health and Population Development (1990) Health Trends in
South Aﬁica. Department of National Health and Population Development,
Pretoria.

Desai 1D, Waddell C, Dutra S, Dutra de Oliveira S, Duarte E, Robazzi ML, Cevallos
Romero LS, Desai MI, Vichi FL, Bradﬁeld RB and Dutra de Oliviera JE (1984)
Marginal malnutrition and reduced physical work capacity of migrant adolescent
boys in Southern Brazil. American Journal of Clinical Nutrition 40: 135-145.

Dietz WH and Gortrnaker SL (1985) Do we fatten our child at the TV set? Obesity and
television viewing in children and adolescents. Pediatrics 75: 807-812.

Dotson CO and Ross JG (1985) Relationships between activity patterns and ﬁtness.
Journal of Physical Education, Recreation and Dance 56: 86-90, January.

Espenschade AS (1963) Restudy of relationships between physical performances of

school children and age, height and weight. Research Quarterly for Exercise and
Sport 34: 144-153.

Eveleth PB and Tanner JM (1976) Worldwide Variation in Human Growth. . Cambridge:
Cambridge University Press.

Eveleth PB and Tanner JM (1990) Worldwide Variation in Human Growth, 2nd edition.
Cambridge: Cambridge University Press.

Falls HB (1980) Modern concepts of physical ﬁtness. Journal of Health, Physical
Education and Recreation 51: 25-27.

Fincham JE, Evans AC and Schoeman S (1996) Interaction between nutrition of
children and helminth infections. South Aﬁican Journal of Food Science and
Nutrition 8: pp. 13-15.

248

 

r'r

Food and Agriculture Organization (1985) Energy and Protein Requirements. Geneva:
World Health Organization. Technical Report #724.

Food and Agriculture Organization (1987) The Fifth World Food Survey. Rome: Food
and Agriculture Organization.

Food Nutrition and Agriculture (1992) International Conference on Nutrition. Food,
Nutrition and Agriculture #5/6. United Nations Food and Agriculture
Oraganization.

Frank DA and Zeisel SH (1988) Failure to thrive. Pediatric Clinics of North America
35: 1187-1207.

Freeman A and Grunewald JW (1980) Incidence of parasitic infestation in black children
at Livingstone Hospital. South African Medical Journal 57: 358-360.

Gear GHS and Pitchford RJ (1985) Bilharzia in South Africa. Pretoria: Department of
Health.

Ghesquiere J and Eeckels R (1984) Health, physical development and ﬁtness of primary
school children in Kinshasa. In J Ihnarien and I Valirnaki (eds): Children and
Sport. Berlin/Heidelberg: Springer, pp. 18-30.

Giay T and Khoi HH (1991) Use of body mass index in the assessment of adult
nutritional status in Vietnam. European Journal of Clinical Nutrition 48: S124-
S130.

Goodway JD (1994) The effect of a motor Skill intervention on the fundamental motor
skills and sustained activity of African-American preschoolers who are at-risk.
Doctoral Dissertation, Michigan State University, East Lansing.

Gopalan C (1992) Undemutrition: Measurement and Implications. Oxford: Clarendon
Press.

Goslin BR and Burden SB (1986) Physical ﬁtness of South Aﬁican school children.
Journal of Sports Medicine and Exercise 26: 128-136.

Haas JD, Martinez EJ, Murdoch S, Conlisk E, Rivera JA and Martorell R (1995)
Nutritional supplementation during preschool years and physical work capacity in
adolescent and young adult Guatemalans. Journal of Nutrition, Supplement, 125:
1078S-1089S.

Hammer LD, Kraemer HC, Wilson DM, Ritter PL and Dombusch SM (1991)
Standardized percentile curves of body-mass index for children and adolescents.
American Journal of Diseases of Childhood 145: 259-263.

Haubenstriker JL and Seefeldt VD (1986) Acquisition of motor skills during childhood.
In V Seefeldt (ed): Physical Activity and Well-Being. Reston, VA: American
Alliance for Health, Physical Education, Recreation and Dance, pp. 41-102.

Heinz HK and McNab GM (1965) Cysticercosis in the Bantu of Southern Africa. South
Aﬁican Medical Journal 30: 19-31.

Henneberg M, Warton C and Hollinshead E (1987) Physical Fitness and growth status
of street boys in Cape Town. Abstract. Paper presented at the 17th Annual
Congress of the Anatomical Society of Southern Aﬁica, 22-24 April, 1987.

Henry J. Kaiser Family Foundation (1991) Changing Health in South Africa: Towards
New Perspectives in Research. Henry J. Kaiser Family Foundation, NY.

Himes JH (1978) Bone growth and development in protein-calorie malnutrition. World
Review of Nutrition and Dietetics 28: 143-187.

249

Himes JH and Bouchard C (1989) Validity of anthropometry in classifying youth as
obese. International Journal of Obesity 12: 183-193.

Hindson D, Byerley M and Morris M (1994) From violence to reconstruction: The
making, disintegration and remaking of an apartheid city. Antipode 26: 323-350.

Huang Y-C (1994) Relationship of sociodemographic and physical activity variables of
physical ﬁtness of Taiwanese junior high school students. Doctoral Dissertation,
University of Texas at Austin.

Hutinger PW (1959) Differences in speed between American Negro and White children
in performance of the 35-yard dash. Research Quarterly for Exercise and Sport 30:
366-368.

Ij sselmuiden CB (1984) Nutritional status of children under the age of 5 years in

Northern Gazankulu. South African Medical Journal 65: 346-347.

International Defense and Aid Fund (1991) Apartheid: The F acts. London: International
Defense and Aid Fund Publications Ltd.

Irwig LM and Ingle RF (1984) Childhood mortality rates, infant feeding and use of
health services in rural Transkei. South Aﬁican Journal of Medicine 66: 608-613.

Joe T (1987) Economic inequality: The picture in Black and White. Crime and
Delinquency 33: 287-299.

Johnston FE, Hamill PW and Lemeshow S (1972) Skinfold thickness of children 6-11
years, United States. Vital Health Statistics, Series 11, No. 120.

Johnston FE, Low SM, Baessa Y and MacVean RB (1985) Growth status of
disadvantaged urban Guatemalan children of a resettled community. American
Journal of Physical Anthropology 68: 215-224.

Jooste PL, Badenhorst CJ, Schutte CHJ, Faber M, var.- Staden E, Oelofse A and
Aalbers C (1992) Endemic goitre among undernourished schoolchildren in
eastern Caprivi, Namibia. South Aﬁican Medical Journal 81: 571-574.

Katzmarzyk PT (1997) A familial study of growth and health-related ﬁtness among
Canadians of aboriginal and European ancestry. Doctoral dissertation, Michigan
State University.

Katzrnarzyk PT, Malina RM, Song TMK and Bouchard C (1998) Television viewing,
physical activity, and health-related ﬁtness of youth in the Quebec family study.
Journal of Adolescent Health 23: 318-325.

Kemper HCG and van Mechelen W (1995) Physical fitness and the relationship to
physical activity. In H Kemper (ed) The Amsterdam Growth Study. A longitudinal
analysis of health, ﬁtness and lifestyle. Illinois: Human Kinetics, pp. 174-188.

Kelder RJ (1984) A comparative study of the physical ﬁtness levels of Indian and White
pupils between the ages of 10 and 14 years. Unpublished Masters Thesis.
Stellenbosch: Universiteit van Stellenbosch.

King AC, Jeffery RW, Fridinger F, Dusenbury L, Provence S, Hedlund SA and
Spengler K (1995) Environmental and policy approaches to cardiovascular
disease prevention through physical activity: Issues and opportunities. Health
Education Quarterly 22: 499-511.

Klika RJ (1995) Growth, maturation and motor characteristics of competitive Alpine
Skiers aged 8-18 years. Doctoral Dissertation, University of Texas at Austin.

250

d to“! La

' '1 gr

Kotz, S and Johnson, NL (1985) Encyclopedia of Statistical Science. Lindeberg
condition to multitrait-multirnethod matrices. Volume 5. NY: John Wiley and
Sons.

Kotze JP, Williams WN, Maclntyre U and De Hoop ME (1986) Anthropometric and
dental data on different population groups in South African Primary Schools.
South African Journal of Science 82: 329-334.

Koumans EHA (1992) Infant and child mortality in the Elim District, northern
Transvaal, 1976 - 1986, and a comparison of trends. South African Medical
Journal 81:202-205.

Kowalski KC, Crocker PRE and Kowalski NP (1997) Convergent validity of the
physical activity questionnaire for adolescents. Pediatric Exercise Science 9: 342-
352.

Krahenbulrl GS, Skinner J S and Kohrt WM (1985) Developmental aspects of maximal
aerobic power in children. Exercise and Sport Science Reviews, 13: 503-538.

Krogman WM (1970) Growth of the head, face, trunk and limbs in Philadelphia white
and Negro children of elementary and high school age. Monographs of the Society
for Research in Child Development 35: 1-80.

Kustrrer HGV and Du Plessis G (1991) The cholera epidemic in South Africa, 1980-

1987. South African Medical Journal 79: 539-544.

Kvalsvig JD, Cooppan RM and Connolly KJ (1991) The effects of parasite infections on
cognitive processes in children. Annals of Tropical Medicine and Parasitology 85:
551-568.

Kvalsvig JD (1981) The effects of schistosomiasis on spontaneous play activity in black
school children in endemic areas. South African Medical Journal 59: 260-263.

Larnpl M, Johnston FE and Malcolm LA (1978) The effects of protein supplementation
on the growth and skeletal maturation of new Guinean school children. Annals of
Human Biology 5: 219-227.

Largo RH and Prader A (1983) Pubertal development in Swiss boys. Helvetica
Paediatrica Acta 38: 21 1-228.

Leary PM (1968) The body measurements of Pedi schoolchildren. South African
Medical Journal 81: 1314-1322.

Little BB, Malina RM, Buschang PH and Little LR (1989) Natural selection is not
related to reduced body size in a rural subsistence agricultural community in
Southern Mexico. Human Biology 61:287-296.

Livingstone MBE, Coward WA, Prentice AM, Davies PSW, Strain JJ, McKenna PG,
Mahoney CA, White JA, Stewart CM and Kerr M-J (1992) Daily energy
expenditure in free-living children: comparison of heart-rate monitoring with the
doubly labeled water (2H2180) method. American Journal of Clinical Nutrition
56: 343-352.

Lohman TG, Roche AF and Martorell R (eds.) (1988) Anthropometric Standardization

Reference Manual. Champaign, IL: Human Kinetics.

Lurie GM and Ford FJ (1958) Heights and weights of European and Coloured

schoolchildren in Cape Town. South African Medical Journal 32: 1017-1025.

251

Malcolm LA (1973) Ecological factors relating to child growth and nutritional status. In
AF Roche and F F alker (eds): Nutrition and malnutrition, Identiﬁcation and
measurement. NY: Plenum Press, pp. 329-353.

Malina RM (1975) Anthropometric correlates of strength and motor performance.
Exercise and Sport Science Review, pp. 249-274.

Malina RM (1985) Growth and physical performance of Latin American children and
youth: Socioeconomic nutritional contrasts. Collegium Anthropologicurn. 9: 9-31.

Malina RM (1988) Racial/ethnic variation in the motor development and performance of
American children. Canadian Journal of Sport Science 13: 136-143.

Malina RM (1992) Physical activity and behavioral development. In NG Norgan (ed):
Physical Activity and Health. NY: Cambridge University Press, pp. 101-119.

Malina RM (1994) Anthropometry, strength and motor ﬁtness. In SJ Ulijaszek and CGN
Mascie-Taylor (eds): Anthropometry: the Individual and the Population.
Cambridge: Cambridge University Press, pp. 160-177.

Malina RM (1995a) Physical activity and ﬁtness of children and youth: Questions and
implications. Medicine, Exercise, Nutrition and Health 4: 125-137.

Malina RM (1995b) Anthropometry. In P Maud and C Foster (eds): Physiological
Assessment of Human Fitness. Champaign, IL: Human Kinetics, pp. 205-219.

Malina RM, Beunen G, Claessens A, Lefevre J, Vanden Eynde B, Renson R,
Vanreusel B and Simons J (1995) Fatness and physical ﬁtness of girls 7 to 17
years. Obesity Research 3: 221-231.

Malina RM and Bouchard C (1991) Growth, Maturation, and Physical Activity.
Champaign, IL: Human Kinetics.

Malina RM and Bouchard C (1988) Subcutaneous fat distribution during growth. In C
Bouchard and FE Johnston (eds): Fat Distribution During Growth and Later
Health Outcomes. NY, NY: Alan R Liss, pp. 63-84.

Malina RM and Buschang PH (1985) Growth, strength and motor performance of
Zapotec children, Oaxaca, Mexico. Human Biology 7: 163-181.

Malina RM, Hamill PVV and Lemeshow S (1973) Selected body measurements of
children 6-11 years, United States. Vital and Health Statistics, Series 11, No. 143.

Malina RM and Little BB (1985) Body composition, strength, and motor performance in
undernourished boy. In RA Binkhorst, HCG Kemper and WHM Saris (eds):
Children and Exercise XI. Champaign, IL: Human Kinetics, pp. 293-300.

Malina RM, Little BB and Buschang P (1991) Estimated body composition and strength
of chronically mild-to-moderately undernourished rural boys in Southern Mexico.
In RJ Shephard and J Parizkova (eds): Human growth, Physical Fitness and
Nutrition. Basel: Karger, pp. 119-132.

Malina RM, Little BB, Shoup RF and Buschang PH (1987) Adaptive signiﬁcance of
small body size: strength and motor performance of school children in Mexico
and Papua New Guinea. American Journal of Physical Anthropology 73: 489-499.

Malina RM and Moriyama M (1991) Growth and motor performance of Black and White
children 6-10 years of age: Multivariate analysis. American Journal of Human
Biology 53: 163-179.

Malina RM and Roche AF (1983) Manual of physical status and performance in

childhood: Vol.2. Physical performance. New York: Plenum.

252

Marriott F HC (1990) A Dictionary of Statistical Terms. 5th ed. NY: John Wiley and
Sons.

Marshall WA and Tanner JM (1970) Variations in the pattern of pubertal changes in
boys. Archives of Disease in Childhood 45: 13-23.

Martorell R (1982) Nutrition and Health Status Indicators: Suggestions for Surveys of the
Standard of Living in Developing Countries. Washington, DC.: World Bank
Research Center.

Martorell R (1993) Enhancing human potential in Guatemalan adults through improved
nutrition in early childhood. Nutrition Today 28: 6-13.

Martorell R, Rivera J and Kaplowitz H (1990) Consequences of strmting in early
childhood for adult body size in rural Guatemala. Annals Nestle 48: 85-92.

Mata LJ (1972) Inﬂuence of recurrent infections on nutrition and growth of children in
Guatemala. American Journal of Clinical Nutrition 25: 1267-1275.

McGarvey ST, Bindon JR, Crews CE and Schendel DE (1989) Modernization and
adiposity causes and consequences. In M Little and J Haas (eds): Human
Population Biology. New York: Oxford University Press, pp. 263-279.

McIntyre J (1996) HIV/AIDS in South Aﬁica - a relentless progression. South African

Medical Journal 86: 27-28.

McKay JE, Sinisterra L, McKay A, Gomez H and Lloreda P (1978) Improving
cognitive ability in chronically deprived children. Science 200: 270-278.
McLeroy KR, Bibeau D, Steckler A and Glanz K (1988) An ecological perspective on

health promotion programs. Health Education Quarterly 15: 351-377.

Millar AJW, Bass DH and van der Merwe P (1989) Parasitic infestation 1n Cape Town
children: A random study of 101 patients. South African Medical Journal 76: 197-
1 98.

Milne C, Sefeldt V and Reuschlein P (1976) Relationship between grade, sex, race and

motor performance in young children. Research Quarterly for Exercise and Sport
47: 726-730.

Ministry for Welfare and Population Development (1995) A green paper for public
discussion. Population policy for South Africa. [Online] Available
Http://www.polity.org.za/govdocS/green_papers/popn.html, April 20, 1995.

Mirwald RL and Bailey DL (1986) Maximal Aerobic Power. London, Ontario: Sports
Dynamics.

Moodley D, Jackson TFHG, Gathir'am V and Van Den Ende J (1991) Cyrptosporidium
infections in children in Durban. South African Medical Journal 79: 295-297.

Must A, Dallal GE and Dietz WH (1991) Reference data for obesity. 85th and 95th
percentiles of body mass index (wt/ht2 ) and triceps skinfold thickness. American
Journal of Clinical Nutrition 53:839-846.

Must A, Dallal GE and Dietz WH (1991) Reference data for obesity: 85th and 95th
percentiles of body mass index (wt/ht2 )- a correction. American Journal of
Clinical Nutrition 54:773.

Must A, Jacques P, Dallal G, Bajema CJ and Dietz WH (1992) Long term morbidity
and mortality of overweight adolescents. New England Journal of Medicine 327:
1 3 50-1 3 55.

253

0!-

1"

Naidu AN and Rao NP (1994) Body mass index: a measure of the nutritional status in
Indian populations. European Journal of Clinical Nutrition 48: $131-$140.

Najjar MF and Rowland M (1987) Anthropometric Reference Data and Prevalence of
Overweight, U.S. (1976-1980). U.S. Department of Health and Human Services:
Public Health Service. National Center for Health Statistics. Series 11, # 238

Parizkova J (1987) Growth, functional capacity and physical ﬁtness in normal and
malnourished children. In GH Bourne (ed): World Review of Nutrition and
Dietetics: NY: Karger, pp. 1-44.

Pate RR (1988) The evolving deﬁnition of physical ﬁtness. Quest 40: 174-179.

Pate RR, Dowda M and Ross JG (1990) Associations between physical activity and
physical ﬁtness in American children. American Journal of Diseases of Children
144: 1123-1129.

Peters J and Ulij aszek SJ (1992) Population and sex differences in arm circumference

and skinfold thickness among Indo-Pakistani children living in the East Midlands
of Britain. Annals of Human Biology 19: 17-22.

Perez GEN, Angelo CPD and Zabala RD (1991) Physical ﬁtness in children and
adolescents from differing SES strata. In RJ Shephard and J Parizkova (eds):
Human Growth, Physical Fitness and Nutrition. Basel: Karger, pp. 80-98.

Perez-Escamilla R and Pollitt E (1995) Growth improvements in children above 3 years
of age. Journal of Nutrition 125: 885-893.

Perusse L, Tremblay A, Leblanc C and Bouchard C (1989) Genetic and environmental
inﬂuences on level of habitual physical activity and exercise participation.
American Journal of Epidemiology 129: 1012-1022.

Pollitt E, Gonnan KS, Engle PL, Marotrell R and Rivera JA (1993) Early
supplementary feeding and cognition: effects over two decades. Monograph of
Society for Research in Child Development #7558: 1-99.

Pollitt E, Gorman KS, Engle PL, Rivera JA and Martorell R (1995) Nutrition in early
life and fulﬁllment of intellectual potential. Journal of Nutrition, Supplement 125:
1111S - 11183.

Pontheiux NA and Barker JE (1965) Relationship between race and physical ﬁtness.
Research Quarterly for Sport and Exercise 36: 468-472.

Post GB, Kemper HCG, Lujan C, Arze RM, Parent G, Speilvogel H and Coudert J
(1992) Comparison of 10-12 years old school boys living at high (3700m) and at
low (420m) altitude in Bolivia. International Journal of Sport Medicine 13: 88,
Abstract. ’

Reaven GM (1988) Role of insulin resistance in human disease. Diabetes 37: 1595-1607.

Richardson BD (1986) Changes in anthropometric measurements of South African
children - a cause for concern. South African Medical Journal 69: 11-12.

Richardson BD (1977) Underweight - a nutritional risk? South African Medical Journal
5 1 : 42-48.

Richardson BD (1978) Growth patterns of South African children: An overview. South
African Medical Journal 74: 246-249.

Richardson BD and Sinwell RE (1984) Infant growth in a rural Tswana community.
South African Medical Journal 65: 775-558.

Riley E (1991) Major political events in South Africa: 1948-1990. Oxford: Facts on File.

254

- “31‘

can.

In.
H. 4
‘ D

I

Rivera JA, Martorell R, Rue] MT, Habicht J -P and Haas JD (1995) Nutritional
supplementation during the preschool years inﬂuences body Size and composition
of Guatemalan adolescents. Journal of Nutrition Supplement 125: 1068s - 1077S.

Robson JRK, Bazin DPH and Soderstrom R (1971) Ethnic differences in skinfold
thickness. The American Journal of Clinical Nutrition 24: 864-868.

Rocha Ferreira M B (1987) Growth, physical performance and psychological
characteristics of eight year old Brazilian school children from low socioeconomic
background. Doctoral Dissertation, UT-Austin.

Rocha Ferreira MB, Malina RM and Rocha LL (1991) Anthropometric, functional and
psychological characteristics of eight-year-old Brazilian children from low
socioeconomic status. In RJ Shephard and J Parizkova (eds): Human Growth,
Physical Fitness and Nutrition. Basel: Karger, pp. 109-228.

Roche AF, Siervogel RM, Chumlea C and Webb P (1981) Grading body fatness from
limited anthropometric data American Journal of Clinical Nutrition 34: 2831-
2838. ,

Rolland-Cachera MF (1993) Body composition during adolescence: methods, limitations
and determinants. Hormone Research 39: 25-40.

Rolland-Cachera MF, Cole TJ, Sempe M, Tichet J, Rossignol C and Charraud A
(1991) Body mass index variations: centiles from birth to 87 years. European
Journal of Clinical Nutrition 45: 13-21.

Rona RJ and Chinn S (1986) National Study of Health and Growth: social and biological
factors associated with height of children from ethnic groups living in England.
Annals of Human Biology 13: 453-471.

Ross JG, Dotson CO, Gilbert GG and Katz SJ (1985) New standards for ﬁtness
measurement. Journal of Physical Education, Recreation and Dance 56/1: 62-66.
January.

Sallis JF, Buono MJ and Freedson PS (1991) Bias in estimating caloric expenditure

ﬁ'om physical activity in children. Sports Medicine 11: 203-209.

Sallis JF, Johnson MP, Calfas KJ, Caparosa S and Nichols JF (1997) Assessing
perceived physical environmental variables that may inﬂuence physical activity.
Research Quarterly for Exercise and Sport 68: 345-351.

Sallis JF, McKenzie TL and Alcaraz JE (1993) Habitual physical activity and health-
related physical ﬁtness in fourth-grade children. American Journal of Diseases of
Children 147: 890-896.

Sallis JF, Nader PR, Broyles SL, Berry CC, Elder JP, McKenzie TL and Nelson JA
(1993) Correlates of physical activity at home in Mexican-American and Anglo-
American preschool children. Health Psychology 12: 390-398.

Sallis JF and Owen N (1997) Ecological models. In: K Glanz, FM Lewis, and BK
Rimer (Eds) Health behavior and health education; Theory, research and practice.
San Francisco: Jossey-Bass, pp. 403-424.

Saris WHM (1986) Habitual physical activity in children: Methodology and ﬁndings in
health and disease. Medicine and Science in Sport and Exercise 18: 253-263.

255

Saris WHM, Binkhorst RA, Cramwinckel AB, Waesberghe F and Veen-Hegemans AM
(1980) The relationship between working performance, daily physical activity,
ﬁtness, blood lipids and nutrition in school children. In K Berg and B0 Eriksson
(eds): Children and Exercise IX. Baltimore: University Park Press, pp. 166-174.

Satyanarayana K, Naidu AN and Rao BSN (1979) Nutritional deprivation in childhood
and the body Size, activity, and physical work capacity of young boys. American
Journal of Clinical Nutrition 32: 1769-1775.

Satyanarayana K, Radhaiah G, Murali Mohan KR, Thimmayamma BVS, Pralhad Rao
N and Narasinga Rao BS (1989) The adolescent growth spurt of height among
rural Indian boys in relation to childhood nutritional background: An 18 year
longitudinal study. Annals of Human Biology 16: 289-300.

Schlemmer L and Stopforth P (1979) A Guide to the Coding of Occupations in South
Aﬁica. Durban: Centre for Applied Social Sciences, University of Natal.

Schmucker B, Rigauer B, Hinrichs W and Trawinski J (1984) Motor abilities and
habitual physical activity in children. In J Ilrnarinen and I Valirnaki (eds):
Children in Sport. NY: Springer-Verlag, pp. 46-52.

Schoeller DA (1988) Measurement of energy expenditure in ﬁ'ee-living humans by
doubly labeled water. Journal of Nutrition 118: 1278-1289.

Schutte CHJ (1985) Combating parasitic diseases of man in South Africa. South African
Journal of Science 81: 451-452.

Schutte CHJ, Eriksson IM, Anderson CB and Lamprecht T (1981) Intestinal parasitic
infections in black scholars in northern Kwazulu. South African Medical Journal
60: 137-141.

Seliger V, Trefrry Z, Bartunkova S and Palver M (1974) The habitual activity and
physical ﬁtness of 12 year old boys. Acta Paediatrica Belgica Supplement 28: 54-
59.

Sharp BL, Ngxongo S, Botha MJ, Ridl F and 1e Sueur D (1988) An analysis of 10
years of retrospective malaria data from the KwaZulu areas of Natal. South
African Journal of Science 84: 102-105.

Shephard RJ (1982) Physical Activity and Growth. Chicago: Yearbook Medical
Publishers.

Shephard RJ, Jaqvier J C, LaBarre R and Rajic M (1980) Habitual physical activity:
Effects of sex, milieu, season and required activity. Journal of Sports Medicine
20: 55-65.

Shetty PS (1993) Chronic undemutrition and metabolic adaptation. Proceedings of the
Nutrition Society 52: 267-284.

Shuenyane E, Mashigo S, Eyberg C, Richardson BD, Buchanan N, Pettifor J,
MacDougall L and Hansen JDL (1977) A socioeconomic, health and cultural
survey in Soweto. South African Medical Journal 51: 495-500.

Sidiropoulos E, Jeffery A, MacKay S, Forgey H, Chipps C and Corrigan T (1997)
South Africa Survey 1996/7. Johannesburg: South African Institute of Race
Relations. .

Simons-Morton BG, 0' Hara NM, Parcel GS, Huang 1W, Baranowski T and Wilson B
(1990) Children's frequency of participation in moderate to vigorous physical
activities. Research Quarterly for Exercise and Sport 61: 307-314.

256

p\

Slaughter MH, Lohman TG and Boileau RH (1982) Relationship to anthropometric
dimensions to physical performance in children. Journal of Sports Medicine 22:
377-385.

Sloan AW (1966a) Physical ﬁtness of Cape Town high school children. South Aﬁican

Medical Journal 40: 682-687.

Sloan AW (1966b) Physical ﬁtness of South African compared with British and
American High School Children. South African Medical Journal 40: 688-690.

Sloan AW and Hansen JDL (1969) Nutrition and physical ﬁtness of White, Coloured,
and Bantu high-school children. South African Medical Journal 40: 508-511.

Slooten J, Kemper HCG, Post GB, Lujan C and Coudert J (1994) Habitual physical
activity in 10- to 12-year-old Bolivian boys: The relations between altitude and
socioeconomic status. International Journal of Sports Medicine 15: S106—S1 l l.

Smaje C (1996) The ethnic patterning of health: New directions for theory and research.
Sociology of Health and Illness 18: 139-171.

Smit PJ (1969) Anthropometric, motor performance and physiological studies on South
Aﬁican children involved in a nutritional status survey. Unpublished Doctoral .
Dissertation. Johannesburg: University of the Witwatersrand.

Srrrit PJ and Potgieter JF (1967) Body measurements of school children of four racial
groups in Pretoria. South African Medical Journal 50: 868-889.

Spurr GB (1983) Nutritional status and physical work capacity. Yearbook of Physical
Anthropology 26: 1-35.

Spurr GB (1987) The effects of chronic energy deﬁciencies on stature, work capacity and
productivity. In B Schurch and NS Schrimshaw (eds): Effects of Chronic Energy
Deﬁciencies on Stature, Work Capacity and Productivity. Lausanne: The
International Dietary Energy Consultancy Group, pp. 95-134.

Spurr GB, Barac Nieto M and Reina JC (1991) Growth, maturation, body composition
and maximal aerobic power of nutritionally normal and marginally malnourished
school-aged Colombian children. In RJ Shepard and J Parizkova (eds): Human
Growth, Physical Fitness and Nutrition. Basel: Karger, pp. 41-60.

Spurr GB, Barac—Nieto M and Maksud MG (1977) Efﬁciency and daily work effort in
sugar cane cutters. British Journal of Industrial Medicine 34: 137-141.

Spurr GB and Reina J C (1990) Daily pattern of %V02 m, and heart rates in normal and
undernourished school children. Medicine and Science in Sports and Exercise 22:
643-652.

Spurr GB and Reina J C (1988) Patterns of daily energy expenditure in normal and
marginally undernourished school-aged Colombian children. European Journal of
Clinical Nutrition 42: 819-834.

Spurr GB, Reina J C, Dahnners HW and Barac Nieto M (1983b) Marginal malnutrition
in school-aged Colombian boys: functional consequences in maximal exercise.
American Journal of Clinical Nutrition 37: 834-847.

Stephenson LS, Latlrarn MC, Adams EJ, Kinoti SN andPerte A (1991) Albendazole
treatment improves physical ﬁtness, growth and appetite of Kenyan school
children with hookworm, T. trichiura, and A. Lurnbricoides infections. Federation
of American Societies for Experimental Biology 5: A1081 (Abstract).

257

9
U
l"'!“
..\

Stephenson LS, Latham MC, Adams EJ, Kinoti SN and Perte A (1993) Weight gain in
Kenyan school children infected with hookworm. Trichuris trichuria and Ascaris
lumbricoides is improved following once- or twice-yearly treatment with
albendozole. Journal of Nutrition 123: 656-665. .

Stephenson LS, Latham MC, Kurz KM, Kinoti SN and Brigham H (1989) Treatment
with a single dose of albendazole improves growth in Kenyan school children
with hookworm, Trichuris trichiura and Ascaris lumbricoides infections.
American Journal of Tropical Medicine and Hygiene 41: 78-87.

Stevens JP (1980) Power of the multivariate analysis of variance tests. Psychological
Bulletin 68: 728-737.

Stevens JP (1992) Applied Multivariate Statistics for the Social Sciences. Hillsdale, NJ:
Lawerence Erlbaum Associates.

Sunnegardh J and Bratteby L (1987) Maximal oxygen uptake, anthropometry and
physical activity in a randomly selected sample of 8 and 13 year old children in
Sweden. European Journal of Applied Physiology 56: 266-272.

Suskind RM (1977) Malnutrition and the Immune Response. New York: Raven Press.

Tanner JM (1962) Growth at Adolescence. 2nd edition. Oxford: Blackwell Scientiﬁc.

Thomas JL, Lochbaum MR, LanderS DM and He C (1997) Planning signiﬁcant and
meaningful research in exercise science: Estimating sample size. Research
Quarterly for Exercise and Sport 68: 33-43. ,

Tobias PV (1986) Physical stature in disadvantaged communities: Johannesburg blacks
have not grown taller this century. South African Joumal of Science 83: 585-588.

Tobias PV (1990) Adult stature in southern African Negroes - further evidence on the
absence of a positive secular trend. South African Journal of Science 78: 97-101.

Torun B, Chew F and Mendoza RD (1983) Energy costs of activities of preschool
children. Nutrition Research 3: 401-406.

Torun B, Davies PSW, Livingstone MBE, Paolisso M, Sackett R and Spurr GB (1996)
Energy requirements and dietary energy recommendations for children and
adolescents 1 to 18 years old. European Journal of Clinical Nutrition 50: S37-S81.

UNESCO (1974) Racism and Apartheid in Southern Africa: South Africa and Namibia.
Paris: UNESCO Press.

UNICEF (1989) Children on the Front Line. Geneva: UNICEF.

Van Niekerk CH, Weinberg EG, Lorn Shore SC and Heese HV (1979) Intestinal
parasitic infestation in urban and rural Xhosa children. South Aﬁican Medical
Journal 55: 756-757.

Van Rooyen F (1994) Effects of an enriched physical activity programme on selected
anthropometric, physiological and performance characteristics of an
underprivileged ethnic group. Unpublished Masters Thesis. Rhodes University.

van Wieringen JC (1986) Secular Growth Changes. In F Falkner and JM Tanner (eds):

Human Growth 2. Postnatal Growth. London: Bailliere Tindall, pp. 307-331.

Wachs TD (1995) Relation of mild-to-moderate malnutrition to human development

correlational studies. Journal of Nutrition 125: 2245S-2254S.

258

Wagstaff L, Reinach SG, Richardson BD, Mkhasive C and de Vries G (1987)
Anthropometrically determined nutritional status and the school performance of
black urban primary school children. Human Nutrition: Clinical Nutrition 41C:
277-286.

Walker ARP, Walker BF and Richardson BD (1972) Running performance in South
African Bantu children with schistosomiasis. Tropical and Geographical Medicine
24: 347-352.

Walker ARP and Walker BF (1977) Studies on increases in growth rate of South African
Black schoolchildren and their Signiﬁcance to health. South Aﬁ'ican Medical
Journal 51: 707-712.

Washburn RA and Montoye HJ (1986) The assessment of physical activity by
questionnaire. American Journal of Epidemiology 123: 563-576.

Waterlow 1C, Buzina R, Keller W, Lane JM, Nichaman MZ and Tanner JM (1977)
The presentation and use of height and weight data for comparing the nutritional
status of groups of children under the age of 10 years. Bulletin of the World
Health Organization 55:489-498.

Waterlow J C (1994) Summary of causes and mechanisms of linear growth retardation.
European Journal of Clinical Nutrition 48: $210.

Weekly Mail (May 16, 1997) A human ﬂood is drowning Guateng. SAPA Reuter.

Wellens RE (1989) Activity as a temperamental trait: Relationships to physique, energy
expenditure and physical activity habits in young adults. Doctoral Dissertation,
University of Texas at Austin.

WHO Working Group (1986) Use and interpretation of anthropometric indicators of
nutritional status. Bulletin of the World Health Organization 64: 929-941.

Wolpe H (1987) Race, Class and the Apartheid state. Paris: UNESCO Press.

World Bank (1995) World Development Report 1995: Workers in an integrating world.
New York: Oxford University Press.

World Health Organization (1976) Methodology of Nutritional Surveillance. Report of a
Joint FAQ/UNICEF/WHO expert committee on the methodology of Nutrition
Surveillance. WHO Technical Report Series, 593. Geneva: World Health
Organization.

World Health Organization (1995) Physical Status: The use and interpretation of
anthropometry. WHO Technical Report Series, 854. Geneva: World Health
Organization.

Wyndham CH (1973) The work capacity of rural and urban Bantu in South Africa. South
African Medical Journal 7: 192-197.

Wyndham CH, Strydom NB, Morrison JP, Peter J, Williams CG, Bredell GAG and
Joffe A (1963) Differences between ethnic groups in physical working capacity.
Journal of Applied Physiology 18: 361-366.

Yach D (198 8) Urbanization in South Africa - consequences for health. South Aﬁ'ican
Medical Journal 74: 479-480.

Yarbrough C, Habicht J-P, Malina RM, Lechtig A and Klein RE (1975) Length and
weight in rural Guatemalan Ladino children: Birth to seven years of age.
American Journal of Physical Anthropology 42: 439-448.

259

 

Zavaleta AN and Malina RM (1982) Growth and body composition of Mexican-
American boys 9 through 14 years of age. American Journal of Physical
Anthropology 57: 261-271.

Zeitlin M (1991) Nutritional resilience in a hostile environment: Positive deviance in
child nutrition. Nutrition Reviews 49: 259-268.

260