“as? 5!
VA. . ’03:...“
.

I l
K... I:
a IR.
.5.
.1 y. . b .
7:\.....;‘
{.73}...
I. , z
11.; a“... l«.,
: . 91‘ 1:11 3. .7... 12:: .1; .1
.tis . . l V . ::I\l\y-‘ ~.
1. 1.9:. 1.1%. If} r
1.. 9 vkl% at l
I.
9‘
33‘}...
[:1 - N.A..d.u~i
It: ix! 5.
7%!
x19?

3
. . .sirn..7.l..
. n..x

 

,2:
...:
2.

ion?

THE-'33“

30m

 

LIBRARY
Michigan State:
University

~--'I'

 

 

This is to certify that the

dissertation entitled

RUNNING ECONOMY AND PERCEIVED EXERTION
IN ADOLESCENT GIRLS

presented by

Karin Allor Pfeiffer

has been accepted towards fulfillment
of the requirements for

Ph.D. degreein Kinesiol oqx

£wa Wfimj

a Major professor
l
Date g/1 3 /0_)
/ /

M5 U i: an Affirmative Action/Equal Opportunity 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

DATE DUE

DATE DUE

 

J91“ 9 :2 363.97.

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDue.p65-p.15

 

 

RUNNING ECONOMY AND PERCEIVED EXERTION
IN ADOLESCENT GIRLS

By

Karin Allor Pfeiffer

AN ABSTRACT OF A DISSERTATION
Submitted to

Michigan State University

In partial fulfillment of the requirements
For the degree of
DOCTOR OF PHILOSOPHY
Department of Kinesiology
2001

Professor James M. Pivarnik

ABSTRACT

RUNNING ECONOMY AND PERCEIVED EXERTION IN ADOLESCENT
GIRLS

By

Karin Allor Pfeiffer

This dissertation includes an introduction (Chapter 1), a review of literature
(Chapter 2) and two papers. In the first paper (Chapter 3), the relationship between age
and exercise economy in adolescent girls, ages 13-18, was examined. In the second

paper (Chapter 4), the reliability and validity of two methods for determining rating of
perceived exertion (RPE) in adolescent girls was examined. Chapter five includes
discussion of the relationships between exercise economy, RPE, and physical activity.
All data included the same group of adolescent girls.

Exercise economy generally improves as a function of age, but the
relationship was not found in this dissertation. Heart rate was found to be the best
predictor of oxygen consumption (V02) for three treadmill stages employed in the
exercise protocol. Ventilation, body surface area, stride frequency, and aerobic capacity
were other variables that helped predict V02 during exercise. Study participants were
fitter than average girls their age. The treadmill speeds chosen may have impacted the
results by causing inefficient running mechanics, thus masking any relationship that may
have existed.

The Borg category RPE scale and the OMNI RPE scale were tested for
reliability and validity. Results showed that the OMN I scale was both more reliable and

valid than the Borg scale for this sample of adolescent girls.

Study results infer a potential relationship between exercise economy, RPE, and
physical activity. Although the number of participants was not large enough to perform
inferential statistics, economy correlated fairly well with number of days of the week on
which girls exercised vigorously for 20 minutes or more. Running economy also
correlated well with OMN I RPE during stage 3 of the treadmill test. More research is
needed before any directionality or causality can be determined from these potential

relationships.

ACKNOWLEDGEMENTS

There are many individuals who have contributed to my success in completing
this dissertation and doctoral degree. My advisor, Dr. Jim Pivarnik, has provided me the
best mentoring I could have possibly received on both a professional and personal level.
His advice and support have been crucial and outstanding. Dr. Robert Malina has taught
me a great deal about understanding the role of growth and maturation in physiological
processes. He has also helped me to become a more critical thinker, and he has been a
supportive mentor. It has been great getting to know Dr. Chris Womack, who always
takes time out to brainstorm and go for “good job” runs. I appreciate the expertise he has
brought to this study and the humor he has brought to my life. Dr. Mat Reeves has been
great at challenging my thought processes and teaching me the principles of
Epidemiology. I have enjoyed all I’ve learned from him. My relationship with Dr.
Jeanne Foley has progressed from mere unknown sport psychology student to friend. I
appreciate all she has done for me and taught me on both a personal and professional
level. I certainly could not have made it without the assistance and friendship of Jo Ann
J anes. I could never appropriately thank her for all she has done for me.

I will never forget the members of the Human Energy Research Lab (HERL).
Thanks to all of them throughout my years at MSU. In particular, Dawn Coe, Cooker
Perkins, Patti Sawyer, and Evelyn Warner have all been a tremendous help in this
dissertation process. I hope all of them are able to encounter individuals as special as

they are when it comes time for them to accomplish important endeavors.

My road to success would have been much more difficult without my family and
friends. A HUGE thank you goes to my husband Matthew for the unconditional love and
support he has provided through both the completion of my masters degree and my entire
doctoral degree. Sometimes I have no idea how he survived the process, but he has been
steadfast as a rock. My parents have been the greatest throughout my life, and I am glad
to make them proud of the monster they have created (stated previously, but never said
enough). Thank goodness I was lucky enough to have my best friend living in the same
town. There were many times I needed Michelle’s perspective to guide me. She is the
better than the best friend for whom I could ask. Thank you also to my other friends and
family (including members of my new family), who have had to understand the times I
could not spend with them in lieu of getting work done.

Finally, I owe thanks to other people/entities:

0 The entire Kinesiology Department at MSU, whose members have become a

second family during my seven-year stint

0 Ms. Linda Quint, who has given up a lot of time with her significant other on

my behalf and provided friendship and support

0 Dr. Greg Fink for his study design and statistical advice

0 The Spencer Foundation, for financial support

TABLE OF CONTENTS

LIST OF TABLES viii
LIST OF FIGURES ix
CHAPTER 1 INTRODUCTION 1
CHAPTER 2 REVIEW OF LITERATURE 5
Introduction 5
Exercise Economy 5
Age and Economy 6
Gender and Economy 10
Body Size and Economy 11
Maturation 13
Pulmonary Parameters 14
Cardiac Function 15
Resting Energy Expenditure 16
Substrate Utilization 16
Running Mechanics 17
Training 19
Other Factors 19
Summary 21
Rating of Perceived Exertion 22
Exercise Mode and Intensity 23
Measurement Issues 26
Development of Child-Specific Scales 31
Summary 36
CHAPTER 3 RUNNING ECONOMY IN ADOLESCENT GIRLS 38
Abstract 38
Introduction 39
Methods 42
Results 47
Discussion 51
CHAPTER 4 RATING OF PERCEIVED EXERTION IN 59
ADOLESCENT GIRLS
Abstract 59
Introduction 59
Methods 61
Results 64

Discussion 65

vi

CHAPTER 5 RELATIONSHIP BETWEEN ECONOMY, RATING
OF PERCEIVED EXERTION, AND PHYSICAL
ACTIVITY
Methods
Results and Discussion
Relationship Between Economy and Physical Activity
Relationship Between RPE and Physical Activity
Relationship Between Economy and RPE
APPENDIX A BORG CATEGORY RPE SCALE
APPENDIX B CHILDREN’S EFFORT RATING TABLE
APPENDIX C CHILDREN’S OMNI SCALE
APPENDIX D CART AND LOAD EFFORT RATING SCALE
APPENDIX E CONSENT FORM
APPENDIX F HUMAN SUBJECTS APPROVAL
APPENDIX G CORRELATION MATRICES

APPENDIX H PREVIOUS DAY PHYSICAL ACTIVITY RECALL

APPENDIX I YOUTH RISK BEHAVIOR SURVEILLANCE SURVEY

REFERENCES

vii

69

70
72
72
74
76
79
81
83
85
87
90
92
97
100

103

LIST OF TABLES

Table 3.1. Age and anthropometric characteristics

Table 3.2. Physiological data at rest and during exercise

Table 3.3. Physiological differences in eight girls with poor economy
Table 3.4. Results of linear stepwise regression analysis

Table 4.1. RPE values for days one and two

Table 4.2. RPE level and %HRmax and %VOzmax by exercise stage
Table 5.1. Prevalence of behaviors in YRBSS questions

Table G.1. Correlation matrix for potential factors affecting economy
in the first treadmill stage

Table G.2. Correlation matrix for potential factors affecting economy
in the second treadmill stage

Table G.3. Correlation matrix for potential factors affecting economy
in the third treadmill stage

Table G.4. Correlation matrix for relationship of economy, RPE, and
physical activity

viii

48

49

51

64

65

73

93

94

95

96

LIST OF FIGURES

Figure 3.1. Relationship between age and V02max
Figure 3.2. Relationship between age and economy in stages 1, 2, and 3

Figure 5.1. Proposed model for relationship between economy, RPE, and
physical activity

ix

57

58

78

CHAPTER 1

INTRODUCTION

Physical activity, defined as any bodily movement produced by skeletal muscles
that results in caloric expenditure”, has been cited as an important variable affecting
chronic disease in adults“. It has been found to have an inverse association with
coronary heart disease incidence and morbidity and mortality of several chronic
ailments”. Recent data suggest that incorporating physical activity into one’s lifestyle
may improve the person's cardiovascular disease risk factor profile, particularly if this
behavior becomes habitual early in life“. Researchers have also found that physical
activity declines during adolescence, especially in females 48. Although there is only
moderate evidence that physical activity tracks (i.e., remains stable) from childhood to
adulthood 49, researchers generally agree that it is important for children to become and
stay physically active 6. The Surgeon General’s Report on Physical Activity and Health
indicates that children should participate in regular physical activity, and continue to do
so throughout their lives”. Two important questions for researchers are 1) how to get
children to be physically active and 2) how to get them to stay physically active
throughout adolescence, so that they might improve their chronic disease risk factor
profiles during adulthood.

Exercise economy, defined as the amount of energy required to perform work at a
submaximal intensity, may be an important factor affecting physical activity behavior.
Individuals with poor economy may have difficulty performing physical activities and

thus, be less likely to perform them on a routine basis. Studies have found that children

are less economical than adults during exercise 1’69'71’78. Possible factors cited include
body size (including obesity), maturation, pulmonary and cardiac characteristics, resting
energy expenditure, substrate utilization, running mechanics, and training 56‘ 66‘ 67. Of
these potential factors, body size has been commonly cited as a variable affecting running
economy 69‘ 71. However, the child study participants were always smaller than the adult
participants, leaving it questionable whether body size was the only variable related to the
poorer economy in children. In a recent study, we examined 13 year-old girls and 23
year-old women of similar heights and weights 1. Despite being the same size, the girls
were still less economical than the adults, although the difference in energy expenditure

65’ 71. Thus, the true nature of the relationship between

was less than found previously
economy and age in females remains unclear.

Exercise economy may affect ratings of perceived exertion (RPE). Perceived
exertion is defined as the ability to detect and respond to sensations that arise as a result
of physiological adaptations to exercise 57. The most common way to assess RPE is use
of the Borg category scale. The original Borg 15-point RPE scale was developed for use
with adult males. Some researchers have found that adults and children differ in their
ratings of perceived exertion at a given workload using this instrument 7. There is also
evidence suggesting that the scale is valid with children 22. Recently, researchers have
proposed a new instrument for measuring RPE in children, the OMNI Scale of Perceived
Exertion 60. It is not clear whether one scale may be more valid or reliable than the other

for any given activity session. Further, it is not known how exercise economy may

relate to either of these methods used for measuring perceived exertion in children.

This dissertation is divided into two studies with one comprehensive review of
literature (Chapter 2). The first study (Chapter 3) examines determinants of, and
differences in exercise economy in adolescent girls ages 13-18 years. The second study
(Chapter 4) focuses on RPE during submaximal exercise conditions using the Borg and
OMNI scales. Chapter five examines the relationships between economy, RPE and
physical activity (estimated by participant recall).

Overall, this dissertation includes four research questions and hypotheses:
Research Question 1: Do walking and running economy improve as a function of age in
adolescent girls, ages 13-18 matched for height and weight?

Hypothesis: As age increases, walking and running economy will improve

Statistical Aglvsis: A two-way AN OVA (age group and treadmill speed), with

repeated measures across one factor (treadmill speed). Stepwise linear regression

analysis to determine predictors of walking and running economy
Research Question 2: Does body mass index (BMI) affect walking and running
economy in adolescent girls?

Hypothesis: There is no difference in walking and running economy between

normal BMI and higher BMI girls

Statistical Analysis: ANOVA to determine differences between groups
Research Question 3: What is the reliability and validity of the two different methods
for measuring children’s RPE?

Hypothesis: The Borg and OMNI scales have similar reliability and validity

Statistical analysis: ANOVA to determine intraclass correlation for reliability and

 

Pearson correlations to determine validity, using percent of heart rate maximum
and percent of aerobic capacity as criterion measures
Research Question 4: Is there a relationship between exercise economy, RPE, and
physical activity?
Hypothesis: Both economy and RPE will be related to physical activity
Analysis: Pearson correlations to determine potential relationships between

economy, perceived exertion, and physical activity

CHAPTER 2

REVIEW OF LITERATURE

INTRODUCTION

Exercise economy is a potential determinant of physical activity in children.
Researchers have not examined the relationship between economy and physical activity,
and it is possible that there are some intervening variables that affect this potential
relationship. For example, economy could possibly affect one’s RPE, which also may be
related to physical activity. The focus of this literature review will be to outline what is
currently known about economy and RPE. At the present time, this author has not found
any studies examining the relationship of economy or RPE to physical activity.
EXERCISE ECONOMY

Exercise economy can be defined as the volume of oxygen (V02) required to
perform an activity at a given submaximal workload. During weight bearing activities
such as walking and running, the V02 is typically indexed by body weight (ml'kg'l'min'l).
It has been shown to be a factor affecting endurance performance, particularly when
individuals with similar fitness levels are compared 35. Several factors are thought to
contribute to economy, including age, sex, body size, maturation, pulmonary parameters,
cardiac function, resting energy expenditure, substrate utilization, running mechanics,
and training status 56’ 66. Also, allometric scaling, reliability of submaximal exercise
testing data (intra-individual variation), and treadmill habituation can affect economy
and/or how economy is expressed. A discussion of each of the potential determinants

follows.

Age and economy

Age has been shown to be a major factor affecting walking and running economy.
Pate et al. 58 showed that economy declines with age in 20-60 yr-old adult runners. They
measured economy in a stratified random sample of participants from the Carolina
Runners Study (the strata were based on age, gender, and weekly running miles) and used
multiple regression to determine predictors of economy (r2 for full model was 0.58).
Bailey and Pate 5 later reported that the relationship between age and economy is
parabolic, with younger and older individuals having poorer economy compared to those
of middle age. The researchers also found that body weight, in addition to age, was

I. 24 reported

significantly associated with economy in adults 58. Similarly, F reedson et a
evidence that seemed to indicate differences in the relationship between economy and age
during childhood and adulthood. Thus, it is important to consider that the nature of the
relationship between age and economy may change over time.

Many who have investigated the relationship between age and economy have
focused on child-adult differences. In all studies comparing children to adults,
researchers have reported that children are less economical than adults, with differences
ranging from 2-10 ml-kg'l-min'l 1’ '9’ 69’ 7" 78. Studies involving walking economy have
indicated that the economy difference between children and adults is greater in pre-

pubescent children 19 than pubescent/post—pubescent children 1 (3-5 ml-kg'l-min'l

difference vs. 2 ml-kg'l-min'1 difference, respectively). Adult-child differences in
economy appear to be greater during running. Unnithan found a difference ranging from
5-8 ml-kg'l-min’l in boys age 10 (pre-pubescent) versus men age 20 years 78. In another

study including pre-pubescent boys, the authors reported a child-adult economy

difference of 10 ml-kg'l -min’l 69. When examining child-adult differences between
premenarcheal girls and women, Rowland and Green 7‘ found a 5 ml-kg'l-min'l
difference. Differences of 4 ml-kg'lmin'l were found between postmenarcheal girls and
women in another study 1. Although it appears that the child-adult differences in boys
versus men may be higher than those in girls versus women, it is not clear how the
combined effects of pubertal status, gender, and running speed may affect economy

differences.

Other investigations of economy have focused on differences between children of
various ages or between children and adolescents. Astrand 4 was the first to examine
economy in children and adolescents. He found that younger participants used more 02
per kg body weight (were less economical) than older participants and that there was an
inverse, linear relationship between V02 and age. A

In addition to Astrand, other researchers have found a linear relationship between
age and V02 in children/adolescents. MacDougall et al. 42 found a linear relationship
between age and V02 in a cross-sectional sample of children ages 7-16. As the
participants’ ages increased, the V02 decreased (economy improved), and differences in
running economy were not more than 5 ml~kg"omin'l 42 between the youngest and oldest
children (ages 7-16). In a sample of children ages 6-12 and adolescents ages 13-19,
Waters et al. 83 found a linear relationship when they plotted age versus V02 and
determined a line of best fit using linear regression. Results showed a significant
difference (3 mlokg'lomin'l) in walking economy between the two groups, with the older

children being more economical than the younger ones.

McCormack et al. 53 divided participants into three age groups, 6-8 years, 9-11
years, and 12-14 years. All study participants ran at 5.0 mph on a motorized treadmill.
The authors found that running economy increased significantly with age across the three
groups (no more than 5 ml-kg'l-min'l), but they also examined the results with respect to
gender. On average, the oldest group of boys was more economical than the younger two
groups, while the older two groups of girls were more economical than the youngest girls.
A limitation of examining the differences between genders was small group sample size.
Specifically, subject number ranged from 6-14 among the groups. Another limitation in
these studies comparing both children and adolescents is lack of assessment of maturity
status. It is unknown how maturation may have affected any of these results.

Some researchers contend that the relationship between age and economy may not
be linear. Freedson et al. 24 reported a curvilinear relationship between speed and V02 in
children. Although Freedson et al. did not examine the relationship between age and
economy itself, their results suggested that age may affect economy differently during
childhood versus adulthood due to the relationship between body size at various ages and
speed.

All studies mentioned thus far included only cross-sectional data. The best
method to determine the effect of age on economy is to perform longitudinal studies that
would show change over time within individuals. Unfortunately, there are not adequate
longitudinal data available to determine how an individual’s economy changes
throughout his or her lifespan. It is difficult to maintain a sizable study cohort for an

extended period of time. The longitudinal studies that do exist will be discussed next.

In one longitudinal study, Ariens et al. examined adolescents at age 13 and
followed them until age 27 2. The researchers did not collect data on participants every
year, but they selected various time points (ages l3, 14, 15, 16, 21, 27). The authors
claimed that the largest change in economy occurred at puberty (ages 13-16) 2.
However, they noted that the participants may not have been at steady state during
exercise because the test stages were only two minutes. Four to five minutes is usually
considered the minimal time needed to achieve a metabolic steady state 67. Further, the
authors did not note the pubertal status of the girls. If their sample included early
maturers, then perhaps some other physiological aspect of maturation (e. g., maturation of
the pulmonary system) aside from puberty itself may have affected change in economy in
the 13-16 year-old age group.

Two research groups measured economy at the beginning and end of three and
five year time periods, and could not account for changes that may have occurred in the
interim 23’ 35. In both studies, researchers began testing economy when the participants
were pre-pubescent. The first study lasted three years. Based on the children's ages it is
likely that they were still pre-pubescent at the end of the investigation, although the
authors did not account for this 23 . The researchers reported an economy difference of
approximately 6 ml-kg'l-min'l from beginning to end. The other study included
participants who were most likely post-pubescent at the end of the study, according to
their ages (mean age = 17 years) 35. Economy differences ranged from 0.6-3.0 ml-kg'l-
min'1 from the first test to the second test, which was 7 years later. However, the
participants did not perform the same protocol at both time periods. Rather, they

completed a protocol that the researchers thought would evoke the same relative exercise

intensity at both ages. Thus, it is difficult to determine what economy changes truly
occurred.

Researchers have also examined exercise economy at yearly intervals during
longitudinal studies. Rowland et al. 64 studied participants every year over five years.
All participants were pre-pubescent at the beginning of the study and were pubescent or
post-pubescent (according to parents assessment of pubic hair development, voice
change, or facial hair) at the end of the study. The economy difference cited was 4-5
ml-kg'l-minl overall, with the greatest changes occurring between years three and four of
the study. In another study, Daniels et al. used a mixed-longitudinal design, and results
indicated a general linear trend toward an increase in economy with age 16. However,
their results showed a deviance from the linear trend at puberty, and training status may
have confounded the results.

Gender and economy

It is possible that gender plays a role in exercise economy. Many exercise
economy studies have focused specifically on boys or men. There are few studies in
which male versus female comparisons have been made in children, and the existing
literature regarding economy differences between boys and girls is equivocal. Some
studies suggest that girls display better economy than boys. In pre-pubescent children,
Freedson et al. 24 found a gender difference in economy of 3-4 ml-kg'l-min'l, with girls
being more economical than boys. This difference was more apparent at higher speeds.
In another study involving pre-pubescent boys and girls, the girls had lower V02 (by 1

ml-kg'l-min") during walking than boys 83. It is questionable whether this mean

10

difference has biological significance, as measurement error for V02 is approximately 2

ml-kg'l-min'l 29.

Astrand 4 was the first to show a gender difference in economy during childhood
and adolescence. The difference of 1-3 ml-kg'l-min‘l was not apparent until higher
treadmill speeds of 6.8-8.7 mph were achieved. He noted that although the girls
performed at a higher percentage of their V02max they were still more economical than
the boys.

Only one longitudinal study in which gender differences were examined was
found 2. Girls had lower V02 than boys by 1-2 ml-kg'l-min‘l, but the authors also pointed
out that the girls worked at a higher percentage of V02max than the boys. Thus, the girls
had better economy despite being closer to their aerobic capacities. It is questionable
whether the small amount of difference found had biological significance.

Studies that have shown no gender difference in economy vary in their designs.

23, 43

Some have included samples of pre-pubescent children , while others used pubescent

or post-pubescent children 42‘ 64. Some studies have been designed to examine walking

economy 23 64, while others examined running economy or a combination of walking and

running 42 43.
Body size and economy

Body size may play an important role in walking and running economy. It is
common knowledge in the field of biology that smaller animals use more oxygen relative
to body mass, both at rest and at maximal exercise, than larger animals 66. According to
66

the surface law, animals generate heat according to surface area instead of body mass

resulting in larger animals having a lower surface area2mass ratio than smaller animals.

11

1TH

Rowland 66 has speculated that since children have a greater surface area to body mass
ratio than adults, they are at an energy disadvantage because they must produce more
energy in order to make up for that which is lost to the environment.

Overweight and obesity may also affect economy. Bailey and Pate 5 have
suggested that segmental mass distribution may be an important factor in energy
expenditure and indicated that adding mass to the trunk may improve economy. Support
for this idea was provided by Pate et al. who found heavier runners were more
economical than lighter ones in a study involving 20-60 year-old adults 58.

Child-adult exercise economy differences in pre-pubescent males 69 and pre-
menarcheal females 7' have been accounted for when V02 is expressed relative to body
surface area rather than weight. In a study where adolescent girls and young women
were matched for height and weight (thus having no difference in body surface area),
economy differences persisted when participants were compared relative to body surface
area 1. Despite this evidence, one should consider that an individual displaces his or her
body mass, rather than surface area, during exercise.

Researchers have focused on obese children in two economy studies. In both
studies, there was no difference in V02 (in ml'kg'lmin'l) between obese and non-obese
children. One study included pre-pubescent participants, and there was a 16 kg
difference between the obese and non-obese children 43 . The other study included post-
pubescent children, and there was a 24 kg difference between the obese and non-obese
groups 68. There were no height differences between participants in either study. The
first study included participants who walked on the treadmill at an 8% grade 68, while the

other included participants who walked and ran at 11 different treadmill speeds 43 . Thus,

the existing literature shows that excess body weight may improve economy in adults, yet
have no effect on economy in children. Without more data it is difficult to make a
conclusion about the relationship between overweight/obesity and economy.

Maturation

The process of maturation most likely plays a major role in economy differences
between children and adults. According to Malina 50, “the influence of maturity status on
submaximal exercise capacity is mediated almost entirely by the effects of maturity status
on body mass, which in turn is closely related to the size of the heart and skeletal muscle
mass”. Unfortunately, it is difficult to assess maturation because it varies in both timing
(when it occurs) and tempo (the rate at which it occurs). Most researchers do not attempt
to account for maturation in their analyses because of the variation. It is likely that
maturation of various body systems could be involved in economy, but there is no real
evidence to support this notion.

Most studies in which researchers have examined exercise economy in children
have not included assessment of maturity indicators. In some studies with young
children, researchers have avoided the issue of maturity status by assuming their
participants were pre-pubertal 19' 24. In others, researchers have classified participants
based on menarcheal status 1' 71. It is not entirely correct to assume a given pubertal
status based on menarche because it is a late pubertal event 50. A pre-menarcheal girl
could easily have begun to reach pubescence, and a post-menarcheal girl is not
necessarily fully mature.

Rowland has considered maturity in his studies by using the criteria set by Tanner

to determine sexual maturation of participants. Tanner’s criteria 73 are based on both

13

pubic hair and genital development. Rowland used only the pubic hair criteria in his
studies, but he included facial hair assessment as well 64’ 69’ 70. Further, Rowland used

64' 69. In the other study, pubic hair

parental assessment of pubic hair in two studies
development was determined by the children themselves or by parents. Although
assessment of sexual maturation using Tanner’s criteria can be arbitrary 50, it is better to
consider maturation rather than ignore it.
Pulmonary parameters

Children ventilate more than adults for each liter of oxygen consumed during
submaximal and maximal exercise 67. This ratio of ventilation to oxygen consumption,
also known as ventilatory equivalent for oxygen (VE/V02), has been shown to decrease
with age 66. The greater ventilation during childhood is usually due to a higher
respiratory rate coupled with a lower tidal volume relative to body weight 66. There is
empirical evidence showing higher respiratory rate and higher VE/V02 in children versus

69' 7'. According to Dempsey 17, the oxygen cost of increased respiratory rate can

adults
average 3-5% of total body V02 during moderate exercise and 8-10% at V02max. Further,
there is evidence that obese children show an increase in ventilation without the
corresponding difference in ventilatory equivalent for oxygen 43. This suggests that size
is related to ventilation. However, a significant difference in ventilation has been found
in girls and women matched for height and weight 1, suggesting that in addition to size,
age may affect ventilation.

Armon et al 3 and Cooper et a1 15 have suggested that children’s higher sensitivity

to C02 than adults may be a mechanism for these respiratory differences. It is likely that

ventilation may account for a portion of the economy difference seen between children

and adults. However, no researchers have measured pulmonary function (e.g., forced
vital capacity, maximal voluntary ventilation) in children versus adults to see if
differences are related to economy. Since lung volume and function have not been
shown to limit aerobic exercise capacity in healthy individuals, it is unlikely that
pulmonary function would be related to economy.

Cardiac function

Turley and Wilmore 75 reported that cardiac output is lower in pre-pubertal
children than adults during submaximal exercise at the same workload on both treadmill
and cycle ergometer. They indicated that the primary reason for the lower cardiac output
was a lower stroke volume, despite a higher heart rate (HR) in children compared to the
adults. Rowland 66 has also found that children have lower stroke volumes than adults
when comparing individuals ages 6—25.

Heart rate is another factor involved in cardiac output. It appears that HR at a
given workload decreases as children age 67. Rowland 65 showed a higher HR and lower
stroke volume in pre-menarcheal girls versus women, but the difference in stroke volume
was not statistically significant. Other studies have shown higher HRS (compared to
adults) for a given submaximal workload in pre—pubertal 78 and pre-menarcheal 7'
children, as well as post-menarcheal girls 1.

One longitudinal study involving pre-pubescent children (who were still pre-
pubescent at the end of the study) indicated that HR at a given workload decreased with
age 23. Another longitudinal study involving pre-pubescent children indicated that girls
had higher HRs at a given workload than boys 63 . Further, obese, pre-pubescent boys and

girls were found to have higher heart rates than non-obese children at a submaximal

15

workload 43. Regardless of HR and stroke volume differences, if one calculates how
much energy is consumed by the myocardium during exercise, the differences found in
these studies are likely to be small (<1 ml‘kg’l'min'l) and not contribute significantly to
total body V02 1’ 33’ 65‘ 75.
Resting energy expenditure

Resting energy expenditure related to body size declines with age 67, and
researchers have hypothesized that it may be a factor in exercise economy differences.
Maffeis et al found that obese, pre-pubescent children appeared to have higher absolute
resting metabolic rate than non-obese children, but the difference disappeared when the
value was expressed with respect to kilograms of fat-free mass 43. MacDougall et al. 42

determined that resting energy expenditure could only account for 1-2 ml-kg'l-min'l of the

economy difference between children and adolescents. A difference of 0.5 ml-kg'l -min’1

was found between post-menarcheal, adolescent girls and women in another study ‘.
Thus, it appears that resting energy expenditure may account for only a small portion of
economy differences.
Substrate Utilization

Greater oxygen consumption is required to metabolize fat compared to
carbohydrate, and there is some evidence suggesting that children may rely more on fat
as an energy substrate during exercise than adults 67. This could be a factor in why
children use more oxygen at a given workload (are less economical) than adults. Pre-
pubertal male children, ages 9-13, were found to use more fat than adults ages 23-33 in a
cross-sectional study examining child-adult differences in economy 69. The amount of fat

metabolized during exercise is typically estimated by examining respiratory exchange

16

ratio (R) values measured during indirect calorimetry . Rowland et al. found that children
had lower R values at the same submaximal intensity (which was also the same percent
of V02max as the adults) and at maximal exercise. However, in another cross-sectional
study, post-menarcheal girls and women of equal size were found to have no difference
in substrate utilization (based on R values) I. Results from a longitudinal study involving
pre-pubescent children (ages 7-10 years) indicated that there was no difference in
substrate utilization at a given workload 23. Results from another longitudinal study of
prepubescent children indicated that girls use more carbohydrate than boys at a given
submaximal workload 62. However, the girls performed at a higher percent of their
V02max and did not differ significantly from the boys in economy. Based on these studies
it is unclear if substrate utilization contributes to the economy differences in children
versus adults.
Running mechanics

In two different studies comparing economy in pre-pubescent and pre-menarcheal
children versus adults, Rowland and colleagues found that greater stride frequency in

69’ 7'. Unnithan and Eston

children appeared to contribute to higher V02 during exercise
78 also determined that greater stride frequency contributed to a higher energy cost of
exercise in pre-pubertal boys, compared to men. In 1954, Astrand 4 was the first to
postulate that an increased stride frequency in children may be responsible for higher
V02. Several researchers after him have also come to this conclusion. In a cross-
sectional study involving children ages 7-16 years, MacDougall et al. 42 found that height

(and ostensibly, stride frequency and stride length) was partially related to economy

differences. However, MacDougall et al. did not actually count stride frequency. They

l7

assumed that stride frequency would be lower in taller individuals. However, in a study
comparing girls and women matched for height and weight, the authors found no
differences in stride frequency 1, despite the fact that the girls were less economical than
the women.

Other gait-related characteristics could be partially responsible for economy
differences between children and adults. Martin and Morgan 51 suggested that ground
reaction force may affect economy. Ground reaction force reflects the net effect of
muscular action and segment accelerations while the body is in contact with the ground5 1.
The authors suggested that, compared to rear foot strikers, individuals who strike the
ground more forward on the foot may rely more on musculature to assist with cushioning.
However, there are no data to support this theory as it relates to running economy in
children. A similar concept was examined in a study where researchers studied muscular
efficiency in children and adolescents 30. The authors calculated efficiency as the ratio of
work rate to V02. They found that muscular efficiency increased with age in a sample of
boys ages 7-15. Their protocol involved having the boys walk uphill on a treadmill. The
researchers also found no relationship between submaximal V02 and leg length, but
stride frequency was not measured.

Cocontraction is the simultaneous activation of agonist and antagonist muscles
during a task 26, and has been implicated as a possible factor for poorer economy in
children than adults. Frost et al. 26 found that younger children experienced more
cocontraction than older children. However, results from another study indicated that 11-

14 year-old children displayed gait patterns similar to those found in adults 28. Thus,

there may be issues related to walking and running mechanics that are responsible for
economy differences.
Training

Training status may play a role in submaximal exercise economy in children, but
research in this area is limited, and results are equivocal. Some researchers have found
no economy difference between fit individuals and less fit individuals 36‘ 79, although
Unnithan et al. 79 did find a difference in HR, R, and RPE between trained and untrained
participants. On the other hand, researchers have found that age and training status
appear to have a combined effect on economy 16’ 35. More research is needed in this area
in order to separate the effects of exercise and sports training from normal growth and
maturation.
Other factors

Other factors to consider when examining economy are allometric scaling,
reliability of V02 measurements, and treadmill habituation. Allometric scaling is defined
as a means of evaluating structural and functional consequences of changes in size of
animals 67. Resting metabolic rate relative to body weight increases as size of the animal
decreases 67. Differences in weight between an elephant and a mouse could vary as much
as a factor of 107, but the difference in weight between a human adult and child might
only be a factor of 10 65 67. Allometric scaling allows researchers to examine
relationships through manipulation of scaling exponents instead of the typical ratio
standard (expressing V02 relative to body mass). The classic use of allometric equations

involves the equation Y=a(Xb), with X representing body weight.

 

Different exponents (representing the variable b in the equation) have been
suggested as appropriate to use when scaling data. Two commonly reported values are
0.67 and 0.75 62. Rogers et al. 62 attempted to determine appropriate scaling factors for
pediatric data during submaximal and maximal exercise. They tested both the 0.67 and
0.75 exponents in addition to the typical ratio standard method and body surface area.
The authors determined use of a body surface ratio (ml~m'2-min'l) or the 0.67 exponent
(ml-kg'm-min'l) was a more appropriate way to express V02 independent of size
compared to the typical ratio standard method (ml-kg'l-min'l). Ebbeling et al. 19 also
implied that the body surface ratio may be superior to the typical ratio standard.
However, the assumption ofien made when using allometric scaling is that growth of
body segments is isometric, which is not the case in children. Further, there is not
consensus for a given scaling factor that always works with children or adolescents at a
particular age.

Another issue to consider is if a single V02 measure can describe the exercise
response reliably. Rogers et al. 63 examined reliability of V02 in children ages 7-9.
Participants ran at steady-state at two different treadmill speeds, and reliability
coefficients were higher for two trials (0.84-0.86) than they were for a single trial (0.72-
0.75). In another reliability study involving walking in girls ages 5-10, one trial was
found to be sufficient, despite the fact that the repeat measurements were taken six weeks
apart (coefficient of variation was 12.4%). Morgan 55 has suggested that if the
experimental setting is well-controlled, a single measure is adequate if group economy

scores are of interest and the subject pool is sufficiently large.

20

An issue closely tied to reliability is treadmill accommodation versus habituation.
Researchers have suggested that participants should undergo practice sessions on the
treadmill prior to measuring economy. Frost et al. 25 suggested the two adjustments made
when acclimating to treadmill exercise are accommodation and habituation.
Accommodation refers to the development of a stable gait pattern, while habituation
refers to lack of between-day or within-day‘differences. The authors tested children ages
7-11 during both walking and running on treadmill. Differences in V02 across trials were
not found to be significant, and the authors concluded that multiple trials across multiple
days did not necessarily improve stability of metabolic or kinematic variables, which
refers to accommodation. Morgan has suggested that 10-60 minutes of treadmill
habituation could be necessary prior to data collection 55.
Summary

Many factors can contribute to exercise economy in individuals. The evidence
shows a distinct difference between children and adults in exercise economy, with
children being less economical. Although researchers have examined reasons for child-
adult differences, it remains unclear at what point between childhood and adulthood these
differences begin to disappear. It is likely that maturation plays a role in economy
differences, but researchers have frequently avoided the topic. Arguments can be made
for body size, pulmonary function, cardiac function, substrate utilization, and stride
frequency all playing a role in child-adult economy differences. However, we recently
found that none of these factors contributed to the economy difference between

adolescent girls and women matched for size 1. Further studies are needed to determine

the cause of the economy differences.

2]

RATING OF PERCEIVED EXERTION

Perceived exertion has been defined as the ability to detect and respond to
sensations that arise as a result of physiologic responses to exercise 57. An individual’s
cognitive awareness of these sensations is considered to be a form of biofeedback in
which central, peripheral, and metabolic changes during exercise are assimilated. Borg
designed the first RPE scale and concluded that it was the best indicator of degree of

physical strain 13.

In addition, Borg asserts that perceived exertion can be used to study
“individuals at work or in leisure time, in diagnostic or clinical situations, and in
epidemiological evaluation of daily exercise intensities” ‘3 . It is possible that the decline
in participation in leisure time physical activities from childhood to adulthood is related
to an individual’s perceptions of exertion 7.

Borg’s RPE category scale was designed to increase linearly with exercise
intensity for work on a cycle ergometer 13. Scale values range from 6 to 20. There are
also verbal descriptions associated with the odd numbered values (see Appendix A). The
scale was devised originally for use with adult males, and each number value on the scale
is meant to correspond to HRS ranging from 60-200 b-min'l (e.g., an RPE score of 14
would correspond to an HR of 140 b-min'l). Many researchers and practitioners use the
RPE scale on a regular basis '3, but Borg himself admitted “there may not be one perfect
scale for all kinds of subjective intensities in all kinds of situations” ‘3.

The concept of RPE specific to children has not been well-studied 40. Some
researchers have questioned the use of the traditional Borg scale in children and

20, 41, 54, 85

adolescents , while others seem to think it is adequate for use with these

populations 22’ 61’ 86. In addition, Bar-Or found that obese children report higher RPE than

22

non-obese at the same workload 9. He hypothesized the difference in perception may be
related to the fact that obese children are less physically active than the non-obese. There
are several issues to consider when assessing use of RPE in children and adolescents.
Exercise mode and intensity, measurement issues, and development of new, child-
specific scales must be considered when evaluating RPE in children.
Exercise mode and intensity

Researchers have chosen cycle ergometry as the exercise mode in approximately
half the studies examining RPE in children. Bar-Or can be considered one of the
pioneers of RPE research in children 40, and he used the cycle ergometer in many of his
early investigations. He cited two main RPE differences between children and adults as a
result of his research. First, children assign lower ratings at a given workload than adults
and second, children’s RPE scores for a given workload appear to decrease faster than
those of adults following aerobic training and heat acclimatization 7. He based these
conclusions on a series of eight studies of 7-68 year old males. Research protocols
involved submaximal cycle ergometer exercise in which both HR and RPE were
measured. Bar-Or reported that younger participants gave lower RPE scores at a given
HR (4-4.5 point difference), even when values were expressed as percent of HR max.
This represents a large difference between younger and older participants. Thus, Bar-Or
concluded that Borg’s idea to multiply RPE level by l0 to estimate HR does not hold true
for children, adolescents, or young adults.

Bar-Or also concluded that during training, children decrease their RPE response
(at a given workload) faster than adults 7. Bar-Or's study participants were grouped by

10-year age increments. He plotted the correlation between RPE and HR versus the age

groups. Based on his analysis, Bar-0r speculated that either adults describe their exertion
levels less accurately than children, or children can better or more honestly perceive
changes in physiological strain than adults. Bar-0r noted that HR can decrease as a
function of training, but he stated that reduction in RPE following training was above and
beyond that of the HR.

There are two studies that have compared RPE assessments in pre-pubertal boys
versus men at ventilatory threshold 44’ 45. Participants in the first study 44 completed a
cycle ergometer V02max test, during which ventilatory threshold was determined. In
subsequent testing bouts, participants exercised at 80%, 100%, and 120% of ventilatory
threshold and reported RPE at each intensity. No difference in RPE was found between
boys and men. However, the researchers admitted that the negative results may have
been due to low power because of small sample size (9 boys and 9 men).

The second ventilatory threshold study also involved a cycle ergometer V02max
test 45 . In this study, children reported higher RPE at ventilatory threshold than adults.
However, the authors noted that ventilatory threshold occurred at the same percent of
V02max in both children and adults. HR was higher in children than adults at ventilatory
threshold, which may partially explain their higher RPE values.

A third study involved measuring RPE ventilatory threshold during treadmill
exercise, rather than cycle ergometry, in pre-pubertal boys only 46. The children reported
a wide range of RPE values at ventilatory threshold, but the numbers were lower than
those reported previously by adults. The authors noted this may be important because
ventilatory threshold usually occurs at a higher percent of V02mx in children compared to

adults.

24

Another study including pre-pubescent boys involved comparison of RPE during
three different treadmill protocols 47. Participants reported higher RPE at maximal
exercise performed by walking than for a running or combination run/walk protocol.
Although the RPE was higher, V02max was lower during the walking protocol. Results
from this study indicate that the higher RPE in the walking protocol may be due to leg
fatigue or protocol length since treadmill elevation was very steep by test end.

Other investigators studying RPE have compared children and adolescents during
cycle ergometer exercise. Williams et a1 studied children/adolescents, ages 11-14
years“. The children first completed a submaximal exercise test that was used to predict
V02max and familiarize them with the Borg RPE scale. Participants were instructed in a
subsequent exercise bout to work at RPE levels corresponding to 9, l3, and 17. The
authors concluded the participants understood the RPE scale since HR increased
significantly with each increasing workload.

22 assessed RPE at power outputs corresponding to 30%,

Eston and Williams
60%, and 90% of predicted V02max and found that the 15-17 year old participants Were
able to use the Borg scale correctly. RPE, HR and power output had linear relationships
with each other, with correlations ranging from r=0.74-0.78. Power outputs were set
based on percentages of predicted, rather than measured V02mu, so it is questionable
whether the participants were truly exercising at the intensities the authors cited.

Regardless of what the true exercise intensities may have been, the linear relationships

found among RPE and HR most likely were not affected.

25

Measurement issues

Reliability, validity, and study design/statistical issues must be considered when
studying RPE measures. Most investigators have found the Borg scale to be reliable for
use with children and adolescents. Mahon and Marsh measured test-retest reliability of
RPE at ventilatory threshold during treadmill exercise in pre-pubertal children 46. The
correlation between trials, which were 2-16 days apart, was r=0.78. The mean RPE at
ventilatory threshold was slightly but significantly higher during the first trial. The
authors noted that there was large inter-individual variation in RPE, which was not due to
differences in ventilatory threshold. They concluded researchers should use RPE with
caution in children ages 8-12, despite the moderately high correlation coefficient.

Bar-Or and Ward examined test-retest reliability in two studies that included
different groups of adolescent, obese children ”’77. In the first study, the investigators
found a correlation coefficient of r=0.92 (averaged across all exercise intensities) for 11
obese adolescents 8’. In the second study, they found a correlation of r=0.59 at the lowest
exercise intensity (20% of V02max) and F089 at the highest exercise intensity (80% of
V02max) 10. Thus, RPE might be a better predictor of exercise intensity at higher exercise
workloads.

Lamb et al questioned the use of traditional methods of correlation coefficients to
determine reliability (repeatability) of RPE measurements 4' . The authors used a
technique called limits of agreement suggested by Bland and Altman '2 to assess
reliability of their data, as well as the typical Pearson interclass and ANOVA intraclass
correlation techniques. Participants were college-age men who completed two identical

treadmill protocols, 2-5 days apart. Results of the correlation analysis revealed test-retest

26

reliabilities ranging from r=0.60-0.81 (Pearson correlation) and from r=0.75-0.82
(intraclass correlation). For both analyses, the correlation coefficients declined as
exercise intensity increased. Limits of agreement results indicated variability in test-
retest results of up to 3 RPE units and the same trend of higher variability in response as
exercise intensity increased. The authors concluded that the Borg 15-point RPE scale
may not have good test-retest reliability. However, results from correlation analysis
followed the same trends as the limits of agreement method, suggesting that the analytic
methods were of similar quality.

Validity is another important measurement issue, which is more important than
reliability. There are two methods researchers have used to validate RPE in children and
adolescents. One method is to compare it to HR, and the other is to test the participants’
ability to set exercise intensity to a given RPE level. RPE and HR are both considered to
be linearly related to workload7. Validating RPE with HR involves checking to see if this
linear relationship holds true. On the other hand, if one holds a work output to a given
RPE level, then changes in RPE should be directly related to the corresponding HR.
Only one study was found in which validity of RPE was assessed in a sample including
pre-pubescent children only. All other studies of youth included both children and

l. 54 included prepubescent children (classified as such by

adolescents. McManus eta
Tanner’s criteria for pubic hair and genital development) in a study consisting of 3-
minute treadmill exercise stages. Significant correlations between HR and RPE, ranging
from r=0.20-0.28, were found in the second, third, and fourth exercise stages, and it

appeared that the children underestimated their exercise intensities. Despite the fact that

the correlation coefficients were statistically significant, values were low, leaving it

27

questionable whether RPE assessment is valid in pre-pubescent children. In fact,
Robertson and Noble 6‘ speculated there may be a critical age threshold involved with
RPE validity. Due to lack of evidence, they are not sure what the age threshold should
be.

In studies that include older children and adolescents as participants, validity
coefficients have been higher than those reported in studies including pre-pubescent
children. Bar-Or 7 conducted a series of studies including participants ages 7-68 years.
All participants completed a submaximal cycle ergometer test, with different workloads
used for individuals of different age groups. Bar-Or reported correlations between RPE
and HR in individuals ages 7-20 ranging from 0.70 - 0.88 during cycle ergometer
exercise. Gillach et al. 27 examined the relationship between RPE and HR in participants
ages 11-14 years versus adults. Sessions were performed two times, one year apart, using
cycle ergometry as the exercise modality. When the researchers examined the data with
respect to the mean of individual correlations, the relationship between RPE and HR in
children was r=0.92 for the first testing session and F 0.94 for the second. When Gillach
et a1. examined the data based on group means, the correlations between RPE and HR in
children were r=0.64 for the first test and F065 for the second. The authors reported
that there was a wide range of HRs for a given RPE level. These results led the authors
to conclude that RPE is good for monitoring changes in exercise intensity, but they were
not sure whether it helped predict level of physical strain. This study had two major
limitations. The first is that the testing periods were a year apart, which introduces the
possibility that the children’s RPE would change as a function of age. The second

limitation is that the exercise stages in the cycle protocol were only two minutes. Thus, if

28

metabolic steady-state is crucial for RPE validity, it is questionable whether participants
achieved this during the various exercise stages.

In another study, Eston and Williams 7'2 explored the validity of RPE using HR as
a criterion measure. Their study participants were 15-17 years old. The protocol
involved two cycle ergometer tests that had three different submaximal exercise
intensities. The relationship between RPE and HR was significant (r=0.74, P<0.01), as
was the relationship between RPE and power output (r=0.78, P<0.01).

Two studies involved children/adolescents performing exercise at levels 7, 10, 13, .
and 16 8° 82 on the 6-20 Borg scale, all using the same protocol. The protocol was set in
4 two stages. The first stage required participants to perform a cycle ergometer test
including four submaximal exercise intensities and an aerobic capacity test. The second
stage required participants to exercise at RPE levels 7, 10, 13, and 16 on both cycle
ergometer and a walking track. One study included overweight children/adolescents 80,
ages 9-15, and the other compared children/adolescents ages 8-14 to adults 82.
Overweight children (30-3 8% fat via skinfolds) were able to discriminate the different
RPE levels accurately on the cycle ergometer 80, but they overestimated (they indicated a
higher intensity than the required level) RPE during the track exercise. The differences
between HRs on cycle and track at RPE levels 7, 10, 13, and 16 were 30, 36, 32, and 28
b-min", respectively, with the track HRs higher at every RPE level. The authors
concluded that overweight children need additional instruction prior to using the Borg
scale. However, the authors were comparing RPE in overweight youth during a non-
weight-bearing versus a weight-bearing exercise activity, with the non-weight-bearing

activity as their criterion measure. Overweight children may give higher RPE during

29

weight-bearing exercise due to the nature of the activity. Thus, the true meaning of study
results may be questionable.

In the other study using the same protocol 82, children/adolescents and adults were
found to be similar in their abilities to reproduce cycle intensity at a given RPE level.
Also, both groups overestimated (overexerted) during the track exercise compared to
their original cycle ergometer test. Adults were better able to estimate the prescribed
intensity level close to the value established in the submaximal cycle ergometer test
compared to the children. The authors expected that children would overexert
themselves at the various RPE levels because children typically report lower RPE for a
given workload 7. However, the children under-exerted themselves instead. Again, one
should question the use of a non-weight-bearing mode of exercise as the criterion when
comparing it to weight-bearing exercise.

In another study, children/adolescents were asked to work at exercise intensities
corresponding to RPE levels of 9, l3, and 17 86. Study participants, ages 11-14,
performed a graded exercise test that was used to predict aerobic capacity. In a
subsequent exercise bout (2-3 weeks later), children were required to reproduce RPE
levels of 9, l3, and 17. Results showed that the children were able to produce the
appropriate exercise intensity at given RPE levels and that the children had higher HRs
than would normally be expected based on RPE level (e. g., HRS at RPE level 9 were
higher than the expected 90 b-min'l). It is important to note that the previously
mentioned study designs did not include workload randomization. Randomizing may
have given a better indication of the children’s abilities, since performing the workloads

in the order of increasing intensity gives the children a point of reference.

30

In smnmary, it appears that, the RPE scale may be more valid in older children
and adolescents, compared to younger children. It is interesting to note that only one
study examining RPE in children and adolescents accounted for pubertal status.
Although emotional and physical maturation may not occur at the same time, it appears
that some type of maturation assessment would be helpful when determining the
reliability and validity of a particular RPE scale.

There are other factors to consider when determining the validity of RPE
assessments, as highlighted in a review by Dishmanls. He notes that there is consensus
regarding the fact low-intensity RPE assessments by adults seem to be governed by local
factors (e.g., muscle sensation). With increasing exercise intensity, central factors (e.g.,
increase in blood lactate, hyperventilation) play a more significant role. He also states
that RPE and relative V02 correspond at all exercise intensities independent of exercise
mode and indicates that V02 may be a better criterion measure than HR when
determining RPE validity. It is important to note that the association of %V02peak and
RPE can change based on fitness level and that RPE may more closely linked with
%V02p¢at< after training in adults”. It is unknown what effect fitness level and/or training
may have on RPE assessment in children. It appears to be clear that the use of both V02
and HR as potential criterion measures for RPE assessment is warranted.

Development of child-specific scales

Due to concerns about use of the traditional Borg scale in a pediatric population,
researchers have developed child-specific instruments that measure RPE. The Children’s
Effort Rating Table (CERT) 85, the OMNI Perceived Exertion Scale 6", and the Cart and

Load Effort Rating (CALER) Scale 21 are three such scales. None of these scales have

3l

had adequate testing for reliability and validity in both child and adolescent populations.
The literature that exists to date will be discussed in the following paragraphs.

Williams et al. 85 developed the CERT scale for use with children, ages 6-9.
Derivation of the CERT followed several steps. First, children ages 5-9 were initiated to
walking and running on a treadmill, stepping on a bench, and pedaling a cycle ergometer.
HR was measured during all activities, and the children were asked about their exercise
perceptions, including their understanding of the Borg scale. Next, a different group of
children performed running, walking, and jumping rope at different speeds and time
periods outside on the playground. After the activities, the children were asked to
describe their efforts, using pictures and words. Based on the children’s responses, the
CERT was devised using phrases appropriate to their descriptions.

The CERT is somewhat similar to the Borg scale in design 20. It has a numerical
range of 1-1 0, with a verbal anchor corresponding to each number value (see Appendix
B). The numeric RPE values are designed to correspond with HR, assuming a linear
relationship between the two measures. The HR value is supposed to equal 100 plus 10
times the RPE value, resulting in a range of 110-200 b-min-l. The verbal anchors range
from “very, very easy” to “so hard I’m going to stop”. The authors contend that the
wording for the CERT is more developmentally appropriate for children than what is
found on the Borg scale.

The first CERT field test involved a stepping at different loading conditions, and
the children described their RPE using the CERT. HRS were recorded during the testing
period. The next phase of field testing involved children working at level 5 and 7 of the

CERT while stepping, and HRS were again recorded. CERT ratings for a given intensity

32

level decreased with age, but the relationships were not analyzed statistically. Also,
children’s HRS were higher than expected at CERT levels 5 and 7, but the authors
attributed this result to unfamiliarity with the stepping task. The authors noted that this
was their first round of testing using the CERT, and they later examined its reliability and
validity.

When Eston et al. first tested reliability and validity of the CERT scale, they used
cycle ergometry as the mode of exercise, and the participants were slightly older, ages 8-
11 20. Children first performed an aerobic capacity test, with four-minute incremental
stages. Participants' CERT RPE scores were measured at the end of each testing stage.
The next testing phase involved the children cycling at intensities corresponding to levels
5, 7, and 9 (randomly ordered) on the CERT. The third phase was a repeat of the second,
and used to estimate test-retest reliability. Results showed significant correlations
between CERT RPE and HR (F076) and CERT RPE and work output (r=0.75). Results
also indicated that the work output values and HRS were lower than those predicted
during the maximal test. Reliability coefficients for the test-retest trial stages ranged
from r=0.9l-0.97 for work output at the three CERT levels and ranged from r=0.65-0.86
for HR at the three CERT levels. The lowest correlation coefficients occurred for
CERT level 9 for work output and CERT level 7 for HR. The authors concluded that the
CERT was reliable and valid, but possible gender differences may exist.

After initial testing, the CERT reliability and validity were evaluated further on
two occasions 38’ 39. The first test involved using both the CERT and Borg RPE scales 38.
Children 9-10 years old performed maximal exercise via cycle ergometer on two separate

occasions and reported RPE based on either CERT or Borg scales (participants were

33

randomly assigned to each group). Test-retest reliability over two trials for the CERT
was r=0.91, and r=0.90 for the Borg scale. Validity coefficients for the relationship
between perceived effort and both HR and power output ranged from r=0.90-0.98 for
both RPE scales. The author suggested that either scale would be appropriate for use
with 9-10 year old children.

Lamb performed a second reliability and validity study, which was an extension
of the first, involving the same participants 39. Participants were asked to reproduce
specific RPE levels from whichever scale they had used in the previous investigation.
This study was also conducted in two trials with cycle ergometer. Children in the CERT
group were asked to exercise at levels of 3, 5, 7, and 9, while children in the RPE group
were asked to exercise at levels of 8, 12, 15, and 18. The authors designed these CERT
and RPE levels to be at comparable intensities. Results indicated that the group using the
CERT had higher HRS and power outputs than the group using the Borg RPE scale.
Validity coefficients were higher when RPE, using either CERT or Borg scales, was
compared to power output, rather than HR. Test-retest reliability for HR and power
output at given CERT and Borg RPE levels were poor. The reliability coefficients were
also low for the lightest workload in the first study (r=0.43 for Borg and F023 for
CERT), but not as low as those in the second study (F008 for Borg and r=0.10 for
CERT). It is unclear why some of the reliability coefficients in the second study were
lower than the first. However, the low coefficients corresponded to the lowest
workloads, perhaps indicating that younger children can report RPE more reliably at

higher workloads.

34

The OMNI scale was developed for use with male and female African American
and Caucasian children ages 8-12 60. Robertson et a1 had a graphic artist design a series
of four pictorial descriptors showing a child riding a bicycle uphill. Children were asked
to describe the four pictures, which gave the scale’s creators words to illustrate the
different intensity levels. Six descriptions provided by the children were chosen to
represent different exertion levels that matched up with the pictures. The descriptions
corresponded to numbers placed along the scale, ranging from zero to 10 (see Appendix
C).

When investigators tested the scale, they used a mixed race and sex sample of
children. Participants completed a cycle ergometer test consisting of four steady-state
exercise stages. HR was measured each minute of the test, and V02 was measured in the
last minute of each exercise stage. RPE was measured via the OMNI scale in the last
minute of each exercise stage. The investigators did not report any reliability testing
associated with the scale, but they did report validity 60. They used regression analysis to
determine if there was a linear relationship between the OMNI RPE scores and both HR
and V02, The r2 values for the models for HR and V02 were 0.86 and 0.88, respectively.
The authors concluded that participants were able to use the scale successfully, but noted
that the response linearity may not apply all the way to V02max because the testing
protocol did not involve the participants exercising to their aerobic capacities. The
standard deviations ranged from 0.22-0.51 at the highest RPE levels reported, and the
authors concluded that relationships between OMN I RPE and HR, and OMNI RPE and
V02 were linear. 60. Further, it is questionable whether the scale is usable because

instrument reliability is unknown.

The CALER scale was developed by the same researchers who devised the
CERTZ'. The authors thought an RPE scale with pictures, as well as numbers and verbal
anchors would be more appropriate for younger children. The authors did have
knowledge of the OMNI scale, but chose to create their own. The CALER scale is
Similar to the OMNI scale, in that there is a picture of a child riding a bicycle at various
stages of exertion (see Appendix D). The difference between the pictures in the two
scales is that the CALER scale depicts the child pulling a cart that is progressively loaded
with bricks. The number of bricks in the cart corresponds to the numbers on the scale,
and the verbal anchors are five statements from the CERT. There is also a difference
between the CALER and OMNI scales in the numeric choices. The OMNI scale has RPE
levels from zero to ten, while the CALER has choices from one to ten.

Eston et al. 2‘ determined the reliability of the CALER scale but not the validity.
Study participants were 7-10 year-old children who were asked to work at CALER levels
of 2, 5, and 8 on the cycle ergometer. Each exercise stage lasted three minutes. The
children participated in four trials. Correlation coefficients were higher between trials 3
and 4 than between the first 2 trials, suggesting that the instrument was more reliable
after the children were able to practice. The authors concluded that the scale is reliable
and valid for use with 7-10 year-old children. However, validity was never tested using
physiological variables as criterion measures.

Summary

The use of RPE in children and adolescents has gained popularity with

investigators over the last few years, most likely because it is potentially related to

physical activity behavior. Reliability and validity studies of the traditional Borg 15-

36

point scale have not confirmed its usefulness with children or adolescents. Researchers
have developed new scales in response to this uncertainty, but it is not clear if these
instruments are more reliable or valid than the original Borg scale. Based on the data
available so far, the OMNI scale has the most promise for use with child and adolescent
populations, but more research is needed. In particular, it is important to determine at
what age certain scales may be most appropriate since RPE use appears to be more valid
in older children than in younger children. It is also important to ascertain which

variables are best to use as criterion measures when determining validity of these scales.

37

CHAPTER 3

RUNNING ECONOMY IN ADOLESCENT GIRLS

ABSTRACT

The relationship between age and walking and running economy [submaximal
oxygen consumption (V02submax)] in adolescent girls (n=5 8), ages 13-18 years, was
examined in this investigation. Most participants were matched for height (158.8 i 3.8
cm) and weight (54.8 d: 5.0 kg). A group of participants with higher body mass index
(BMI) who fit the same height criteria were also recruited. Anthropometric measures
(height, weight, sitting height, breadths, skinfolds) and pre-exercise V02, as well as a
habituation session, were obtained prior to a submaximal (consisting of 3 different
stages) and maximal treadmill test. The primary purpose of this study was to determine if
exercise economy improves with age in a cross-section of girls ages 13-18 years who
were matched for size. The secondary purpose of this study was to determine the effect
of higher BMI on the girls' economy. Results of repeated measures analysis of variance
(AN OVA) showed that economy did not significantly improve with age in this sample.
Stepwise linear regression showed that HR was the best predictor of economy in all three
treadmill stages. Ventilation and body surface area were predictors of economy in the
first and third treadmill stages. V02max was an additional predictor of economy in the
first treadmill stage, and stride frequency was an additional predictor of economy in the
third treadmill stage. There were not enough higher BMI participants to perform
inferential statistics (n=4), but the higher BMI girls did not appear to differ from the

normal BMI participants in economy. Results indicated that the running speed may have

38

resulted in the girls having inefficient running mechanics due to the above average fitness
levels of the participants.
INTRODUCTION

Economy can be defined as the oxygen consumption (V02) required to perform
submaximal, steady-state exercise at a given workload. The lower the V02 is, the better
the economy. During submaximal exercise in children, a steady-state can usually be
achieved within three to four minutes 67. Economy has been documented as a factor
influencing endurance performance, particularly among individuals of a similar aerobic
capacity 35, and may also influence capacity to perform recreational and occupational
physical activities 67.

Cross-sectional and longitudinal studies have shown that children are less
economical (i.e., greater submaximal oxygen consumption) than adults, both at a given

7 - '
2' 69' 70’ 8 and at a given relative percentage of

submaximal treadmill speed and grade
aerobic capacity 35. Also, there is some evidence that the difference between children and
adults becomes larger with increased exercise intensity 7" 75. The reason(s) for
submaximal economy differences between children and adults remain(s) unclear, but
researchers have suggested many factors that may be responsible. Body surface-to-mass
ratio, ventilatory parameters, gait kinematics, cardiac function, and substrate utilization
have all been implicated as factors that may affect economy. Several factors may be a
result of size differences between study participants. In most economy studies comparing
65, 69-71.

children/adolescents to adults, the adults are larger than their younger counterparts

Thus, researchers have not been able to discount size as a factor affecting economy.

39

To remove the effects of body size from consideration, we recently conducted a
study I that included girls (ages 12-14 years) and women (ages 18-24 years) matched for
stature and weight. Despite being matched for size, the girls used more oxygen per
kilogram body weight (were less economical) than the women in both treadmill walking
and jogging. The economy differences between girls and women in our study were

6 71 - .
5’ . Also, the difference 1n economy

somewhat less than those found in other studies
between groups was greater for jogging than for walking. These results indicated that
size explains only a portion of the difference in exercise economy between children and
adults. We concluded that factors other than size contribute to the economy difference
between girls and women.

We postulated that the reasons for the remaining difference in economy may be
due to resting energy expenditure, myocardial oxygen consumption, or ventilation. We
estimated that resting energy expenditure accounted for ~25% of the V02 difference
during walking and ~1 1% of the difference during running. Based on previous research
33’ 65’ 75, we estimated that myocardial V02 due to higher heart rate (HR) values in the
girls would not have added significantly to the total body V02. Also, higher ventilation
during both walking and running (both due to higher respiratory rates) in the girls was not
likely to have contributed much to the economy difference 17. Further, there were no data
available regarding the maturity status of the girls in our study, other than the fact that the
girls were post-menarcheal. Having information about menstrual cycle regularity would
give an indication of overall maturity status, since achievement of menarche does not

represent complete maturity. Thus, the reasons for differences in economy between girls

and women matched for size remain unclear.

40

An issue directly related to size differences is whether being overweight or obese
affects submaximal energy expenditure. Bailey and Pate 5 suggested that segmental mass
distribution may be an important factor in economy. They hypothesized that extra mass
in the trunk region may improve economy, although it may not be beneficial from other
standpoints. Indeed, Pate et al. 58 showed that heavier adult runners were more
economical than lighter runners. However, obese youth have been shown to have no
difference in walking economy expressed as V02 per kg body mass 68 or V02 per kg fat
free mass 43 when compared to their non-obese counterparts.

Obese, pre-pubescent children were also found to have similar economy
compared to their non-obese peers during running 43. While there was no attempt to
match for height in either study, there was no significant difference in height between the
obese and non-obese participants. Neither study included stride frequency (or other gait
parameters), or pulmonary function measures as possible contributors to energy
expenditure. One might not expect a difference in gait or pulmonary function because
the children were the same height, but it is unknown how being overweight may affect
these variables.

Our recent study 1 showed that matching girls and women for size removed some
of the difference in economy normally seen between children and adults. However,
matching for size only removed part of the difference, which indicates there are other
factors (e.g., maturity) involved. Based on our previous study, those factors must be
related to changes that occur between the ages of 13-18 years. The primary purpose of
this study was to determine if exercise economy improves with age in a cross-section of

girls ages 13-18 years who were matched for size. The secondary purpose of this study

41

was to determine the effect of higher BMI (same height range, but heavier in weight than
the normal BMI participants) on the girls' economy.

METHODS

Participants. Study participants were post-menarcheal, adolescent girls, ages 13-18
years, who fit a particular range of height and weight criteria (N=5 8). ' Height and weight
criteria were 156-164 cm and 48-57 kg, which are the same ranges used in our previous
study I. These values represent slightly taller than average 13-year-olds, and average to
slightly shorter than average height l7-year-olds, who fit within the 15-85th percentile
range of body mass index (BMI) percentile rank according to CDC growth chart
normative values 77. In addition, an attempt was made to recruit a group of higher BMI
girls. All participants in this group (N =4) had a BMI greater than the 85th percentile for
girls of the same age 77. The majority of participants were not trained distance runners,
defined as running more than 10 miles per week. Girls who played on sports teams were
not excluded. All participants except three played on at least one Sports team (basketball,
gymnastics, soccer, softball, swimming, tennis, and, track), despite an attempt to include
non-athletes as well as athletes in the study.

Participants were recruited through contacts with a church youth group, daughters
of Michigan State University employees, school teachers, and sport team coaches.
Written assent to participate and parental consent were obtained for each participant (see
Appendix E). This research was approved by the University Committee on Research
Involving Human Subjects (UCRIHS) at Michigan State University (see Appendix F).
Study Protocol for Day One. Each participant was familiarized with the laboratory

environment. Height and weight were measured, followed by pulmonary function testing

42

that included an assessment of forced vital capacity (F VC) and maximal voluntary
ventilation (MVV). Background information (including menstrual history, asked by a
female investigator) was also obtained. Each participant participated in a treadmill
habituation session, consisting of at least ten minutes at various speeds and grades.
Pulmonary Function Testing. Each participant was provided with a clean, disposable
mouthpiece to place on the end of a hose attached to a spirometer (SensorMedics; Yorba
Linda, CA). She was given instructions for how to perform the F VC and MVV
protocols. The F VC test involved the participant breathing normally for a few seconds,
followed by a maximal inspiration. Immediately after the inspiration, each participant
was required to perform a forced, maximal exhalation, as quickly as possible. The MVV
test involved the participant breathing as fast and forcefully as She could, for 12 seconds.
Each test was performed at least two times, with the participant in a seated position. The
average of the two trials was used for analytical purposes.

Background information. Each participant was asked some basic background questions
including date of birth, age at menarche, and menstrual history for the last four months
prior to study entry. Girls who could not recall their past four months of menstrual
history were asked to track their cycles for the following few months. The menstrual
history data provided information regarding cycle length, which was to be used as an
indicator for maturity '1’ 74. Sexual maturity assessment using Tanner’s criteria was not
employed due to concerns that there would be little variation in maturity status between
participants.

Treadmill Habituation. Each participant engaged in a practice bout of treadmill walking

and jogging. Previous research 55 has shown that 10 minutes is usually adequate time for

43

habituation to occur. More time was spent if the investigator felt it was necessary, or if
the subject requested it. Expired gases were collected, and participants wore a heart rate
telemetry belt (Polar; Seattle, WA) during this time period in order to familiarize them
with wearing the chest strap and breathing through the one-way valve during actual
testing.

Study Protocol for Day Two. Anthropometric measures were obtained, followed by the
treadmill exercise test. In most cases, there was approximately a one-week time period
between testing days, but in no more than 15 instances there were 2-4 weeks between
laboratory visits.

Anthropometry. Height (without shoes) and Sitting height were measured to the nearest
0.1 cm using a stadiometer that was calibrated with a steel tape measure of known length.
Weight was measured to the nearest 0.1 kg on a kilo-pound beam balance that was
calibrated with known weights certified by the Bureau of Standards. Biacromial and
biiliac breadths were measured to the nearest 0.1 cm using a spreading caliper. Skinfolds
were obtained at five sites: triceps, subscapular, suprailiac, thigh, and medial calf.
Triplicate measures were taken and values recorded to the nearest 0.5 m using a Lange
caliper. All anthropometric measurements were taken by the same female investigator.
Treadmill test. During all phases of treadmill testing, participants wore heart rate
monitors and breathed into a mouthpiece connected to a (SensorMedics 2900; Yorba
Linda, CA) metabolic measurement cart for analysis of expired respiratory gases. Prior
to beginning exercise, each participant sat in a chair on the treadmill for 10 minutes.
Expired gases were collected during this time in order to provide an index of pre-exercise

V02. Although this value represented a value higher than basal metabolic rate, it

44

provided a value typical of each participant’s resting energy expenditure. The girls were
instructed to refrain from exercising and eating and drinking (except for water) three
hours prior to the test. Participants typically wore t-shirts, shorts, and athletic shoes
appropriate for treadmill exercise.

The first stage of the treadmill test involved walking at 2.0 mph, 0.0% grade for
six minutes. In the second and third minutes, stride frequency for the minute was
determined by counting the number of right foot heel strikes in 30 seconds and
multiplying by two. Walking economy was calculated from the mean V02 during the
final three minutes of the stage.

The second stage of the treadmill test involved walking at 3.0 mph, 5.0% grade
for six minutes. In the second and third minutes, stride frequency was counted in the
same manner as the first stage. Economy was calculated from the mean V02 during the
final three minutes of the stage.

The third stage of the treadmill test involved jogging at 4.5 mph, 0.0% grade for
six minutes. Again, stride frequency was counted in the second and third minutes of the
stage in the same manner as outlined for the first stage. Running economy was calculated
from the mean V02 during the final three minutes of the stage. Previous testing from our
laboratory has shown that girls with the higher BMI values would be able to maintain a
speed of 4.5 mph. Previous testing from our laboratory had also shown that a speed of
4.5 mph would most likely not be an awkward pace for the normal BMI group. Any
participant who attempted to walk at this pace was asked to jog instead.

At the end of the 18 minutes of walking/running, speed remained at 4.5 mph and

grade was increased 2.5% each minute until the participant reached volitional fatigue.

45

Criteria for reaching V02max included two of the following: 1) Respiratory exchange ratio
(RER) > 1.0, 2) maximum heart rate > 95% age-predicted maximum, 3) plateau or
decrease of V02 (< 2 ml-kg‘l-min’l) during the final minute of the stage. Although the
investigator verbally encouraged the participants to continue to run as long as poSsible,
each subject decided when to terminate the test and indicated her desire to stop by
grabbing the front bar of the treadmill. Three participants did not reach V02max according
to our criteria, but this did not affect analysis of the economy data. Immediately, the
treadmill speed and grade were decreased to 2.0 mph and 0.0% grade. The participant
walked slowly until her HR fell below 120 b-min‘l, at which point She came off the
treadmill.

Statistical analysis. The outcome variable in this study was V02 during each stage. A
power analysis revealed that 16 participants per group was necessary for assessment by
analysis of variance (AN OVA), assuming an expected difference of 4.0 ml-kg'l-min'l,
power of 0.8, alpha level of 0.5, and effect size of 1.0. Repeated measures ANOVA was
used to determine if age affected economy during the three exercise stages. Further,
stepwise linear regression analysis was used to determine which factors predicted
walking and running economy. Three correlation matrices were run to determine which
factors had the best relationship to V02 in the three economy stages of the treadmill
protocol and to determine which factors might be related to each other (see Appendix G).
Variables with the six highest correlations to V02 in each stage were entered into the

regression analysis. Statistical software packages used were Statview SE (Abacus

Concepts Inc., 1988) and JMP (SAS Institute Inc., 1996).

46

RESULTS

Anthropometric participant characteristics are reported in Table 3.1, and
physiological data are reported in Table 3.2. The participants in this study, particularly
the normal BMI girls, were fitter than average girls their age 67. They also did not show
the typical decline in V02max with age normally seen across this age group 37 (see
Figure 3.1). To determine the effect of age on economy, repeated measures ANOVA was
run. The analysis revealed that age had no effect on economy in any of the exercise
stages (F = 0.45, P = 0.8). Closer inspection of the data revealed that eight girls, across
all ages, displayed poorer economy than the rest of the sample during the third treadmill
stage (see Table 3.3 for characteristics). Eight girls did not represent enough to perform
statistical tests, but inspection of the data showed negligible differences between them
and the rest of the sample for the first two treadmill stages. There was a difference in HR
between the eight girls and the rest of the sample during the third stage of the treadmill
test.

Since age had no effect on economy, stepwise linear regression analysis was used
to determine the factors affecting economy in 13-18 year old girls. Prior to data analysis,
a correlation matrix was run for potential variables affecting economy during the first,
second, and third stages of the treadmill test. The variables included in the correlation
matrix were any which were potentially related to economy, based on the literature (see
Appendix G). The variables with the six highest correlations to V02 during the three
stages of the treadmill test were selected and placed into stepwise linear regression. The
six highest correlations were used so that we did not include too many variables in the

analysis for the number of participants.

47

Age was not highly correlated with V02 during any of the exercise stages (r=-0. 10
for all three stages). The six variables placed into regression analysis for stage 1 were
HR, VE, V02max, body surface area (BSA), age at menarche, and BMI. Results of linear
stepwise regression showed that four factors influenced economy during the first stage of
the treadmill test: HR, V02max, V5, and BSA (see Table 3.4 and Figure 3.2). However,
the final model could only account for 55% of the variance in V02 during the first stage.

The Six variables placed into regression analysis for stage 2 V02 were HR, age at
menarche, body surface area, sitting height, MVV, and VE (see Table 3.4 and Figure 3.2).
The only variable that entered the model was HR, which represented 31% of the variance.

Table 3.1 Age and anthropometric characteristics

 

 

Sample (N=58) High BMI (N=4)
Mean St. Dev. Mean St. Dev.
Age (years) 15.3 1.5 16.7 1.9
Age at menarche (years) 12.5 1.0 12.2 1.3
Height (cm) 158.8 3.8 160.3 3.1
Weight (kg) 54.8 5.0 72.3 5.6
BMI (kg-m'z) 21.7 1.9 28.2 2.4
Sum of Skinfolds (mm) 77.6 15.2 112.9 13.1
Sitting height (cm) 84.6 2.6 85.6 2.5
Leg length (cm) 74.2 2.8 74.7 2.5
Sitting heightzstature ratio 0.53 0.01 0.53 0.01
Biacromial breadth (cm) 35.2 1.5 37.0 1.3
Biiliac breadth (cm) 25.6 1.2 28.2 1.7
Body surface area (m2) 1.55 0.08 1.76 0.07

 

NOTE: Median values for BMI and sum of Skinfolds in the normal BMI group were 21.8
and 77.0, respectively. Values were 27.9 and 112.1 for the higher BMI group.

The six variables placed into regression analysis for stage 3 V02 were HR, VE,
stride frequency, sitting height, BSA, and biiliac breadth. Results of stepwise linear
regression showed that four factors influenced economy during the third stage of the

treadmill test. Heart rate, VE, stride frequency, and BSA all entered into the final model

48

(see Table 3.4 and Figure 3.2). HR during the third stage of the treadmill test had the
most influence on V02, accounting for 55% of the variance. Ventilation, BSA, and stride
frequency all had a much smaller effect on economy, accounting for 7%, 8%, and 4% of
the variance, respectively. Thus, the final model accounted for 74% of the variance
during the third treadmill stage.

Despite a vigorous recruiting effort, it was difficult to find higher BMI girls who
were willing to participate in the study. Only four higher BMI girls were included in the

Table 3.2 Physiological data at rest and during exercise

 

 

Sample (N=58) High BMI (N=4)
Mean St. Dev. Mean St. Dev.

Rest
Rest HR (b-min'l) 83 12.2 80 4.0
Rest V02 (ml-kg'lmin'l) 4.4 0.7 3.7 0.8
Maximal exercise
v02max (ml-kg'l-min'l) 46.9 5.3 40.1 7.9
HR...ax (b-min") 202 9.3 197 9.1
Stage 1
HR (b-min‘l) 104 12.6 102 7.0
vo2 (ml-kg'l-min'l) 11.5 1.0 10.8 0.9
vE (L-min") 17.6 2.2 19.7 0.9
%v02m,,. 24.7 3.0 27.4 4.0
Stride freq. (steps-min") 50 2.5 51 1.3
Stage 2
HR (b-min") 136 15.0 137 11.2
vo2 (ml-k "-min") 20.9 1.2 19.7 0.6
v5 (Lmin' ) 30.5 4.0 35.1 2.4
%v02m,,. 45.2 5.9 51 10.8
Stride freq. (steps-min") 59 _ 2.5 59 3.0
Stage 3
HR (b'min'l) 168 17.4 172 16.4
W; (ml-k "-min") 31.6 2.8 30.2 0.6
vE (L-min' ) 49.8 9.0 59.8 6.8
%v02max 68.3 10.5 77.7 16.9
Stride freq. (steps'min'l) 80 4.3 80 2.1

 

49

Table 3.3 Physiological differences in eight girls with poor economy

 

Sample (N=50) Poorer Economy (N=8)

 

Mean St. Dev. Mean St. Dev.
Age (years) 15.1 1.5 16.1 1.5
Age at menarche (years) 12.5 1.0 12.7 0.6
BMI (kgm'z) 21.8 1.9 21.3 1.6
v02max (ml-kg'l-min'l) 47.6 5.1 45.9 2.6
HR“...x (b-min“) 201 9.3 208 7.1
Rest
Rest HR (b-min'l) 81 10.4 96 17.8
Rest v02 (ml-kg'l-min'l) 4.4 0.6 4.7 1.0
Stage 1
HR (b°min'l) 102 10.6 120 15.3
V02 (ml'kg’l-min'l) 11.3 0.8 12.8 0.8
vE (L-min“) 17.2 2.2 19.2 1.7
Stride freq. (stepsmin'l) 50 2.5 49 1.8
Stage 2
HR (b'min'l) 133 13.1 155 15.3
V02 (ml-kg’l-min'l) 20.7 0.9 22.7 1.4
vE (L-min'l) 29.6 3.8 33.5 3.1
Stride freq. (steps-min'l) 59 2.6 58 1.6
Stage 3
HR (b-min’l) 165 16.1 188 12.1
v02 (ml-k "-min") 30.9 2.0 36.8 1.7
VE (L°min' ) 47.1 7.5 61.2 6.0
Stride freq. (steps-min'l) 81 4.3 80 4.6

 

final sample, which did not allow for use of inferential statistics in analysis of their data.
Tables 3.1 and 3.2 Show the differences in anthropometric and economy data between the
lower and higher BMI girls. The average age of the higher BMI girls was slightly older
than that of the lower BMI girls, since three of them were 17 years old. The higher BMI
girls were heavier and slightly taller than the lower BMI girls. They also had a greater
sum of five Skinfolds. The frame Size of the higher BMI girls was slightly larger than
that of the lower BMI girls, with both biacromial and biiliac breadths being greater. The
higher BMI girls had higher VE than the lower BMI girls during all three treadmill

exercise stages. Finally, the higher BMI girls had slightly lower V02 values at rest and

50

during the three exercise stages than the lower BMI girls, despite being at a higher
percentage of their aerobic capacities.

Table 3.4 Results of linear stepwise regression analysis

 

 

 

Factor (N=58) R R Square R Smmre A
Stage 1

Stage 1 HR .492 .242 .242
V02max .628 .395 .153
Stage 1 V5 .684 .468 .073
BSA ' .739 .547 .079
Stage 2

Stage 2 HR .555 .308 .308
Stage 3

Stage 3 HR .742 .55 .55
Stage 3 V5 .787 .619 .069
BSA .838 .703 .084
Stride frequency .861 .742 .039
DISCUSSION

The primary purpose of this study was to determine if exercise economy improves
as a function of age in girls 13-18 years old. It was expected that a significant negative
relationship would exist between age and V02. That is, as the girls' age increased,
exercise economy would improve 2’ 65’ 69’ 7'. Repeated measures AN OVA showed that
there was no relationship between age and economy in our sample. The lack of
relationship between age and economy was surprising. Further analysis Showed that age
was poorly correlated with economy during all stages of the treadmill test (1: -O.10 for all
three stages, see Figure 3.2).

Once it was determined that there was a lack of relationship between age and
economy, we searched for the factors that predicted economy using stepwise linear
regression analysis. Age did not enter into any of the regression models for V02 during

any submaximal treadmill stage. The best regression model came from stage three and

51

accounted for 74% of the variance in V02. The four variables entering the model were
HR, VE, BSA, and stride frequency. It was expected that stage 3 would show the best
model since researchers have shown that economy differences between children and
adults increase with increasing exercise intensity" 70’ 75. The larger difference might
allow for better comparisons among potential factors affecting economy.

Although stage three provided the best regression model, HR was found to be a
predictor of V02 in the regression models from all three treadmill stages. HR accounted
for 24% of the variance in stage one, 31% of the variance in stage two, and 55% of the
variance in stage three and was the best overall predictor of economy. Thus, it appears
that HR has more influence on economy as exercise intensity increases. The relationship
between HR and V02 is well known as being linearsz, but the linearity of the relationship
and the existence of the relationship do not explain the lack of relationship between age
and economy.

V5 and BSA entered the regression models predicting V02 in both stage one and
stage three. VE accounted for 7% of the variance in the first and third treadmill stages.
BSA accounted for 8% of the variance in stages one and three. Previous research has
Shown that adolescent girls ventilate more than women who are matched for body size ‘
during submaximal treadmill exercise. However, according to calculations we performed
based on the research of Dempsey 17, VE most likely did not account for more than 1
ml-kg'l-min'l of the economy difference in our previous study. BSA area may have had
more of an effect on economy if the participants had not fit into the height and weight
criteria used for this investigation, but it is important to note that it had an effect despite

the matching.

52

The remaining variable that predicted economy in the third treadmill stage was
stride frequency, which accounted for 4% of the variance. Previous research has shown
that stride frequency in pre-pubertal children was greater than in adults 69’ 71' 78. We
found only one investigation examining stride frequency differences between adolescents
and adults, in which there were no differences in stride frequency between post-
menarcheal girls and women l. The girls in the study were matched for height and
weight, which may have removed a potential difference in stride frequency normally seen
between children and adults. However, the girls in this study were also matched for
height and weight.

The finding that age and economy were not related was unexpected. Upon closer
inspection of the data, we found eight girls, across all ages, who displayed poorer
economy than the rest of the sample during the third treadmill stage. The only
remarkable characteristic displayed by the eight girls that was different from the rest of
the sample was higher HR at rest, during the three economy stages, and at maximal
exercise. HR was the best predictor of economy in all three treadmill stages for the
overall sample. Kitamura et al.33 showed that there is a good correlation between HR and
myocardial V02. Based on rough calculations from Kitamura et al.'s paper, it appears
that even the greatest HR difference of 23 b-min'l would only account for approximately
0.2-0.3 ml-kg" min". Thus, it is doubtful that the increased HR in the eight girls could
account for the difference in energy expenditure between the eight girls with poorer
economy and the rest of the sample.

We believe the likely reason for the lack of relationship between age and

economy in the third treadmill stage is running Speed. Although there are no established

53

“normal values” for V02nm, our study participants showed greater than average fitness
compared to other girls their age (2.6 L-min'l vs. 2.0 L-min'l)67. Anecdotal evidence
indicated that several girls wanted to walk at the 4.5 mph speed because it was a slow
running pace for them. The slower pace was chosen so that girls of any fitness level
would be able to participate throughout the entire protocol.

The slow jogging speed may have caused participants to use an inefficient gait
pattern. Stride frequency was found to be a predictor of economy in the third treadmill
stage, but there were no apparent differences in stride frequency between the eight girls
with poorer economy and the rest of the sample for any of the treadmill stages. Rowland
believes that the increased stride frequency in children versus adults is a result of shorter
leg length in the children“. However, there was a poor correlation between leg length
and stride frequency in all three treadmill stages (r=<-0.23, r=-0.33, and r=-0.02 for
stages 1, 2, and 3, respectively). This could suggest that the speed, particularly during
running, may have produced an inefficient gait pattern and masked any economy
differences that might be present across the sample.

Our results may not surprise researchers who believe that child-adult differences
in economy are related to body size 71. Rowland 66 has speculated that individuals who
have a greater surface area to body mass ratio may be at an energy disadvantage because
they need to produce more energy in order to make up for that which is lost to the
environment. BSA did fit into the first and third stage regression models, but it only
explained 8% of the variance in each. This is most likely because our participants fit into
a particular height and weight range, which did not allow much difference in BSA

between participants. However, based on our previous research I we expected to remove

54

all effects of body size and BSA by matching for height and weight, so that it would be
easier to tease out other factors affecting economy.

The secondary purpose of this study was to determine the effect of BMI on
economy. Unfortunately, only four girls with BMI above the 85th percentile 77
volunteered to participate, and inferential comparisons could not be made with the
normal BMI girls due to lack of statistical power. From a descriptive standpoint, the
higher BMI girls differed little from the normal BMI girls. The higher BMI girls were
slightly taller, with larger biacromial and biiliac breadths than the normal BMI girls.
Since they were taller and had larger lungs, the higher BMI girls had larger VE than the
normal BMI girls. The higher BMI girls also had slightly lower V02 values at rest and
during exercise than the normal BMI girls. It is doubtful whether these differences were
biologically significant, let alone statistically significant.

There were some limitations to this study. The sample population may have been
biased due to the girls being recruited from a convenience sample. The participants were
fitter than average girls their age, which may have affected economy. Although the
literature appears divided regarding whether or not training status affects
economy [6’ 35’ 36’ 79, the running pace may have been awkward for the girls from a
biomechanical perspective. No biomechanical data other than stride frequency were
collected, however. Another limitation to this study was the small number of higher BMI
girls who participated. If more of the higher BMI girls had participated in the study,
better comparisons could have been made between the high and normal BMI groups.

In summary, results of this study indicated that there was no relationship between

age and economy in adolescent girls matched for height and weight. HR was a predictor

55

of economy during all three treadmill phases. The only other factors that appeared to
relate to economy were VE, body surface area, and stride frequency in the third stage of
the treadmill test. However, these other factors did not play a large role in predicting
economy. Results of our previous economy study indicated that size played a role in
economy differences between adolescent girls and adults, but there might be other factors
involved. Results of this study did not clarify what other physiological or biomechanical
factors may be related to economy. More research is needed on what role walking and/or

running speed may play in treadmill economy.

56

V02max (ml/kg/min)

Figure 3.1. Relationship between age and V02max

 

60

1:1 1:: y = 57.335 - 0.65367x R02 = 0.041

'3 V02max
9 V02max (K)

 

y = 63.479 - 1.4457x R"2 = 0.958

 

 

 

30 r r ' I ' I
12 14 16 18 20
Age (years)

NOTE: V02max (K) represents data from Krahenbuhl et al., 1985

57

V02 (ml/kg/min)

4O

3O

20

10

Figure 3.2. Relationship between age and

economy in stages 1, 2, and 3

 

 

 

 

y = 22.316 - 8.7051e-2x R"2 = 0.012

o O
o o o ° °
o o :01: o (0:63: o 0°
0 U 0 u 0
o O 8 o O o
o oo o 0° 8°
‘3 o o
- y = 35.097 - 0.22405x R02 = 0.014
. V
o 9 ’ .
O O N . 9 . O
- ’ 9 fl 2 9 . ., "o '

 

y = 12.537 - 6.5715e-2x R"2 = 0.011

_Wf a

 

 

12

I I I

14 16

Age (years)

58

20

D
O

O

StgiVOZ
SthVOZ
SthVOZ

CHAPTER FOUR

RATING OF PERCEIVED EXERTION IN ADOLESCENT GIRLS

ABSTRACT

The test-retest reliability and validity of the Borg and OMNI rating of perceived
exertion (RPE) scales were examined in a population of adolescent girls (N=58,
age=15.3i1.5 years). Participants were randomly assigned to which scale they used, as
well as which treadmill exercise condition (walking, walking uphill, or jogging). Results
of intraclass correlation (AN OVA) indicated that the OMNI scale (R=0.95) was more
reliable than the Borg scale (R=0.78). Single-day reliability estimates were slightly lower
with OMNI (r=0.91) more reliable than the Borg scale (r=0.64). Standard error of
measurement for the Borg scale was 1.24 scale units, while standard error of
measurement for the OMNI scale was 0.49 scale units. Criterion validity was determined
using Pearson Product Moment correlation analysis, with % HRmax and %VO2max as the
criterion measures. Validity estimates were higher for the OMNI scale compared to the
BORG. Specifically, validity coefficients (r) were 0.86 and 0.89 for the OMNI,
compared to 0.66 and 0.70 for the Borg, for %HRmax and %VO2max comparisons,
respectively. Although the OMNI scale has not been tested in adolescent girls, it appears
to be more reliable and valid than the Borg scale for use in this population during
treadmill exercise.

INTRODUCTION
Perceived exertion has been defined as the ability to detect and respond to

sensations that arise as a result of physiological adaptations to exercise 57. The cognitive

59

awareness of these sensations is considered to be a form of biofeedback in which central,
peripheral, and metabolic changes during exercise are assimilated. Borg designed the

first rating of perceived exertion (RPE) scale and believes RPE to be the best indicator of
degree of physical strain 13 .

Borg’s 6-20 RPE instrument is a category scale that was designed to increase
linearly with exercise intensity for work on a cycle ergometer 13. There are also verbal
descriptions associated with odd numbered responses (see Appendix A). For example, a
Borg RPE value of 7 corresponds to “very, very light”, while number 19 corresponds to
“very, very hard”. The scale was originally devised for use with adult males, and each
number value on the scale is meant to correspond to heart rates (HR) ranging from 60-
200 b-min’l (e.g., an RPE score of 14 would correspond to an HR of 140 b-min'l). Many
researchers and practitioners use the RPE scale on a regular basis '3, but Borg himself
admitted that “there may not be one perfect scale for all kinds of subjective intensities in
all kinds of situations” 13.

The use of RPE in the pediatric population has not been well-studied 40, and it is
not clear whether or not the original Borg scale is appropriate for use with children.
Children ages 8-12 years reported a wide RPE range for a given treadmill intensity in one
study 46. In fact, Robertson and Noble 6‘ indicated there might be an age threshold
associated with use of the Borg scale. Lamb et al. 4' questioned the use of Borg’s RPE
scale in. young adult males, but other researchers have found it to be reliable in children
and adolescents 46’ 47‘ 61’ 86. Further, Bar-0r and Ward found that obese adolescents
reported higher RPE for a given workload than non-obese. However, the obese

participants were working at a higher percentage of their aerobic capacities 10.

60

The Children’s OMNI Scale of Perceived Exertion was developed specifically for
use with children of mixed race and sex 60. This scale uses numerical ratings that range
from 0-10 and a series of pictures Showing a child on a bicycle (see Appendix C). The
zero end of the scale is represented by flat ground, and the rest of the exercise levels are
depicted with the child riding the bicycle uphill. Validation work has been done on the
OMNI scale, and results have shown it is valid over a range of exercise intensities 60.
However, it is not clear that the OMNI has been validated against higher-end exercise
intensity, and all validation performed so far has used cycle ergometer exercise in 8-12
year old children. The authors noted they do not know if the scale is generalizable to
other exercise modes or children or adolescents of different ages. Thus, the OMNI scale
appears to have promise for use with children and adolescents but has not been well-
studied.

Since RPE is an indicator of physical strain, it may be a factor in physical activity
behavior. However, this issue has not been addressed in the literature. One reason for
not addressing a potential relationship between RPE and physical activity is contradictory
evidence presented thus far regarding the appropriateness of the various scales that are
available. It is not clear which RPE scales are most reliable and valid for use with this
population. The purpose of this study was to compare the reliability and validity of the
Borg and OMN I RPE scales during submaximal treadmill exercise in a population of
adolescent girls ages 13-18.

METHODS

Participants. Study participants were adolescent females, ages 13-18. Participants were

excluded if they were trained runners (i.e., ran more than 10 miles per week), but were

61

not excluded if they participated in other physical activities or team sports. Written assent
to participate and parental consent was obtained for each participant. This research was
approved by the University Committee on Research Involving Human Subjects
(UCRIHS) at Michigan State University.

Instruments. Two different RPE scales were used in the testing protocol. Prior to
performing the exercise tests, participants were familiarized with each scale.

Borg 6-20 RPE (category) scale. The Borg category scale 13 is designed to describe
subjective perceptions of exercise (physical stress). The scale consists of numbered
choices, 6-20, and verbal cues, from “very, very light” to “very, very hard”. During the
last minute of each phase of the testing protocol, participants were asked how hard they
felt they were exercising. All instructions given to participants followed the guidelines
outlined by Bar-Or 8.

Children ’s OMNI scale of perceived exertion. The OMNI scale was developed
specifically for use with children of mixed race and sex status. The OMN I is an 11-point
scale, consisting of numbered choices, 0-10, and verbal cues, from “not tired at all” to
“very, very tired”. The scale also has a set of four pictures of a boy riding a bicycle. The
first picture is placed at the level zero and depicts the boy riding on flat ground. The
other three pictures Show the boy riding uphill, with the boy appearing more and more
tired as the scale progresses. These other three pictures are positioned between the 2-4,
5-7, and 8-10 on the scale. Participants were asked how they feel according to this scale
during each minute of the last phase of the exercise test.

Testing protocols. Data were collected on two occasions, approximately one week apart.

The testing protocols involved treadmill exercise, with the first day including habituation

62

and the second day including a test with three different exercise stages. During both
treadmill bouts, expired gases were collected continuously via indirect calorimetry
(SensorMedics 2900; Yorba Linda, CA). Heart rate was measured via telemetry (Polar;
Seattle, WA) and recorded for each minute of exercise. Prior to data collection,

I participants were randomly assigned to use one of the RPE scales for one of the three
exercise stages. Random assignment involved creating a list of all possible combinations
of RPE scale and treadmill stage. Each combination was assigned a number, and a
random number table was used to choose the number of the combination each participant
would use.

The protocol for the habituation session on the first day lasted 10-15 minutes.
Participants always began the habituation day by walking 2.0 mph, followed by 3.0 mph
at 5.0% grade, followed by jogging at 4.5 mph. Each girl spent six minutes in the
randomly assigned stage and two to five minutes in each of the other two stages. After
the sixth minute HR value for the assigned stage was recorded, participants were asked to
rate their perceived exertion level according to the appropriate RPE scale (Borg or
OMNI, whichever they were assigned).

The protocol for the second treadmill testing session involved sitting quietly for
10 minutes followed by three, six-minute, steady-state exercise stages. The three
exercise stages corresponded to the same intensities previously mentioned for the
habituation day. Participants were asked their RPE during the last 30 seconds of the
same stage at which they were asked RPE during the habituation test.

Statistical analysis. Reliability was determined using intraclass correlations

(AN OVA). Single test reliability was determined using the Spearman-Brown prophecy

63

formula: r = K * (r1)/ [1 + (K-l) * r1] where K is the number of trials performed and r, is
the reliability coefficient established for K trials. Standard errors of measurement (SEM)
’ were calculated using the formula: SEM = SD * SQRT(l-r) where SD is the standard
deviation of the RPE scores, and r is the intraclass correlation coefficient. Validity was
determined using Pearson correlation, with both percent of HR maximum (%HRmax) and
percentage of aerobic capacity (%V02ma.() as criterion measures.
RESULTS

Borg RPE scores differed by approximately one point between the two days.
OMNI RPE scores did not differ by more than half a point between the two days. RPE
scales are presented in Table 4.1. lntraclass correlation (via repeated measures ANOVA)
was used to determine reliability for each scale. The OMNI scale showed better
reliability (R=0.95) than the Borg scale (R=0.78). Single-day reliability estimates (which
represent an estimate of the reliability if the scale was given at only one point in time)
were slightly lower with OMNI (r=0.91) being more reliable than the Borg scale
(r=0.64). Standard error of measurement for the Borg scale was 1.24, while standard
error of measurement for the OMN I scale was 0.49.

Table 4.1 RPE values for days one and two

 

 

 

Day One Day Two

Mean St. Dev. Median Mean St. Dev. Median
Borg Stage 1 (n=9) 8.2 2.2 7.0 8.0 1.7 8.0
Borg Stage 2 (n=9) 11.1 1.8 11.0 12.0 1.7 12.0
Borg Stage 3 (n=10) 11.5 2.3 12.0 12.7 2.1 13.0
OMNI Stage 1 (n=10) 0.3 0.5 0.0 0.7 0.8 0.5
OMNI Stage 2 (n=10) 2.3 1.3 2.0 2.4 1.0 2.0
OMNI Stage 3 (n=9) 4.8 1.6 5.0 5.2 1.6 5.0

 

64

Criterion validity was determined using Pearson Product Moment correlation
analysis, with % HRmax and %VO2max as the criterion measures (see Table 4.2 for
o/oHRmax and %VO2max data). Validity estimates (r) were higher for the OMNI scale
compared to the BORG. Specifically, validity coefficients were 0.86 and 0.89 for the
OMNI, compared to 0.66 and 0.70 for the Borg, for %HRmm and %VO2mam comparisons,
respectively.

Table 4.2 RPE level and %HRm;,x and %VO2max by exercise stage

 

 

 

 

Stage %HR,,m % V02max
Mean St. Dev. Mean St. Dev.

Borg Stage 1 53.3 2.8 24.2 2.5
Borg Stage 2 70.8 7.1 47.5 5.8
Borg Stage 3 82.6 5.0 65.8 4.7
OMNI Stage 1 51.2 4.9 23.1 2.3
OMNI Stage 2 64.0 6.7 44.0 4.9
OMNI Stage 3 87.9 6.8 73.4 12.8
DISCUSSION

The reliability estimate for the Borg scale in this study is similar to those found in
other studies. Mahon and Marsh46 reported the same reliability estimate for the Borg
scale in children ages 8-12 at ventilatory threshold as was found in the present
investigation. In adolescent, obese children Bar-Or and Ward found an average RPE
reliability of R=0.92 across different exercise intensities”). They also found coefficients
of R=0.59 at 20% of aerobic capacity and R=0.89 at 80% of aerobic capacity in another
group of obese adolescents.

The reliability estimate for the OMNI scale in this study was excellent. We were
unable to compare our estimate to previous estimates because no other data are presently

available regarding the reliability of the OMNI scale. Thus far, the creators of the scale

65

have only reported validity but not test-retest reliability. There was less than 0.5 point
difference between both days of OMNI RPE estimates for all three treadmill stages.
Also, the SEM was low for the OMNI scale, indicating that a measurement error of
approximately half a point affects the true RPE value.

Traditionally, a common criterion used to validate RPE scales has been HR at a
given workload7’ 20' 27. However, in this study validity was determined by using both
%HRmax and %VO2max as criterion measures. We considered these to be the best
criterion measures because a) the participants’ actual HRmax and V02max values were
known and b) these max values differed among the subjects.

In studies where HR was used as a criterion measure for RPE, Borg scale validity
coefficients have been good to excellent. Gillach et al. 27 found validity coefficients
between Borg scale RPE and HR of r=0.92 and r=0.94 in 11-14 year-old participants on
two different occasions. In contrast, Eston et al. 22 found a somewhat lower validity
coefficient between Borg scale RPE and HR of r=0.74 in a study with adolescent boys.
Correlations between Borg scale RPE and HR fell anywhere between 0.70 and 0.88 for
individuals ages 7-20 years in Bar-Or’s studies7. All previously mentioned studies were
performed using cycle ergometer.

Correlations between Borg scale RPE and %HRm.x and %VO2max in the present
investigation were F066 and r=0.70. The values for %HRmax are slightly lower than
those reported for previous studies, but two of the previously mentioned investigations
correlated RPE score directly with HR rather than %HRmaxzz’ 27. Bar-Or appears to be the
only investigator who has examined RPE related to %Han. He plotted the data rather

than calculating the correlation coefficient, however. Based on the plots, the RPE-

66

%HRmax relationship displayed by his participants appears similar to the participants in
the current investigation for the first two treadmill stages. During the third treadmill
stage, our participants reported lower RPE values at a higher %HRmax than Bar-Or’s
participants. It is unknown why this phenomenon occurred. However, our participants
were fitter than average girls their age 67, and it is possible that they reported lower RPE
scores than less fit individuals might report. The fitness level of Bar-Or’s participants is
unknown. However, his participants were all males.
When the original validation of the OMNI scale was performed, Robertson et al.60
used both HR and V02 as criterion measures. The authors employed regression analysis
to determine if there was a linear relationship between OMNI RPE and both HR and
V02. The model for HR had F093, and the model for V02 had an F094. The scale
was developed for, and validated on children ages 8-12 during cycle ergometer exercise.
To our knowledge, the present investigation is the first to examine the validity of the
OMNI scale for use with adolescents during treadmill exercise. Results showed that
validity estimates were higher than those found with the Borg scale, regardless of
whether %HRmax (F086) or %VO2max (F089) were used as criterion measures. These
values are similar to those found in the original validation study of the OMNI scale.
Each participant used one RPE scale during one stage of the treadmill testing. It
is unknown how the reliability and validity coefficients may have been affected by this
protocol. Most other studies examining reliability and validity of RPE scales have
included assessment of RPE at several stages of exercise. This could be viewed as either

a limitation or a strength of this study. It could be a limitation because individual

participants did not assess RPE at different intensities. It could be viewed as a strength if

67

one assumes that RPE assessment in one stage did not affect subsequent ratings. We
considered both points while designing the study. Since our purpose was to determine
reliability and validity of the scales, we decided it was best to use RPE at one given stage,
rather than allow earlier assessment to affect later ones.

Another potential limitation of this study was the method by which reliability was
determined. The conditions for assessing RPE were not exactly the same on the first and
second days. All first-day assessments of RPE were measured during a Six-minute
treadmill stage that was one of the three treadmill stages completed on the second day.
Despite the fact that participants did perform the other two stages on the first day, the
stages were not performed for the full six-minute time period that they were on the
second day. Our idea was to assess RPE for one given stage so that any RPE
measurement would not be affected by previous measurements, and we believe we
achieved this.

Overall, our results indicate that the OMN I scale is a reliable and valid instrument
to use with adolescent girls performing treadmill exercise. The results are particularly
interesting to note because our study is the first to include OMNI reliability data on an
adolescent population during treadmill exercise. Although the OMNI scale was
originally developed for use with 8-12 year-olds performing cycle ergometer exercise, we
found it to be more reliable and valid than the Borg scale for treadmill exercise in
adolescent girls. Recommendations for further research include testing the OMNI scale
with younger age participants and examining its reliability and validity using other

exercise modalities.

68

CHAPTER FIVE

RELATIONSHIP BETWEEN ECONOMY, RATING OF PERCEIVED
EXERTION, AND PHYSICAL ACTIVITY

Physical inactivity is a major cardiovascular disease risk factor for adults. It has
been found to have an inverse association with coronary heart disease incidence and
morbidity and mortality of several chronic diseases 59. Although cardiovascular disease
risk factors often do not manifest until adulthood, investigators have begun to focus
attention on physical activity during childhood. Researchers are currently searching for
determinants of physical activity during childhood and adolescence with the hope that
active children will become active adults. Unfortunately, current evidence Shows that
physical activity declines during adolescence, especially in females 48. Although there is
only moderate evidence that physical activity tracks (i.e., remains stable) from childhood
to adulthood 49, researchers generally agree that it is important for children to become and
stay physically active 6. The Surgeon General’s Report on Physical Activity and Health
indicates that children should participate in regular physical activity, and continue to do
so throughout their lives 76.

A search of existing literature uncovered no evidence that any researchers have
examined either exercise economy or rating of perceived exertion (RPE) as potential
determinants of physical activity 72. It is conceivable that an adolescent who has poor
walking and running economy feels he/she cannot adequately perform and is
subsequently discouraged from participating in physical activities involving this skill. It

is also possible that an individual who perceives a physical task to be intense may be

deterred from participating in the activity. Without a theory to test or any data available

69

regarding potential relationships, it is difficult to determine what those relationships
might be. The purpose of the fourth study hypothesis was to examine potential
relationships between economy, RPE, and physical activity (see Figure 5.1 for a model).

METHODS

Each study participant visited the Human Energy Research Laboratory on two
occasions. Participants were asked to recall physical activities from the previous day
using the Previous Day Physical Activity Recall (PDPAR; see Appendix H) which has
been validated in children and adolescents 84. Completing the PDPAR involves
identifying the duration and intensity of physical activity behaviors for the entire day and
recording the values in the corresponding boxes on the activity recall form. The first
page of the recall includes pictorial and verbal clues regarding how to identify physical
activity intensity. The four physical activity intensity choices range from “very light:
slow breathing, little or no movement” to “hard: hard breathing, moving quickly”. The
original plan was to schedule testing days to provide PDPAR assessments that would
represent two different types of day (i.e., weekday and weekend), but due to scheduling
difficulties this rarely occurred. Participants completed the PDPAR during the beginning
of each laboratory visit.

Study participants also answered questions from the Youth Risk Behavior
Surveillance Survey 34 (see Appendix I). The Centers for Disease Control (CDC)
developed the survey to focus attention on important health problems in youth. After
reviewing leading causes of morbidity and mortality, representatives from the CDC
decided the survey should cover six areas: behaviors that resulted in unintentional and
intentional injuries, tobacco use, alcohol and other drug use, sexual behaviors, dietary

behaviors, and physical activity 34. A steering committee was appointed by the CDC to

70

choose questions that would address prevalence of behaviors in the six main areas. The
total number of questions needed to be limited so that students would be able to complete
the questionnaire in a 45-minute class period. Eight questions pertain to physical
activity, which the participants in this study answered during their first visit to the
laboratory.

Data analysis. The PDPAR data were collected as minutes spent in activity. These data
were averaged across both days and placed into five different categories: Total PDPAR
minutes, very light activity minutes, light activity minutes, medium activity minutes, and
hard activity minutes. The YRBSS questions were all structured using number of days,
number of minutes, or number of sports teams. A lower numbered response to a YRBSS
question means less days, minutes, or sports teams, depending on the question.

YRBSS questions 1, 4, 6, 7, and 8 were considered the most relevant to physical
activity behaviors, as well as the most likely to have relationships to both economy and
RPE. Question 1 asks the number of days in the last week the girls participated in
activities that made them sweat and breathe hard for 20 minutes or more (an attempt to
measure vigorous physical activity). Question 4 asks the number of days in the last week
the girls walked or bicycled for at least 30 minutes (an attempt to measure moderate
physical activity). Question 6 asks how many minutes the girls are active in an average
physical education class. Questions 7 and 8 ask about sport team participation. To
determine potential relationships, a correlation matrix was run between economy, RPE,

and physical activity (see Appendix G).

71

RESULTS AND DISCUSSION

The correlation matrix between economy, RPE, and physical activity appears in
Appendix G. A negative correlation would be expected for the relationship between
economy and physical activity, assuming that a more economical person would be more
physically active (have more PDPAR minutes and more days, minutes, or sports teams on
the YRBSS). A negative correlation would also be expected for the relationship between
RPE and physical activity, assuming that someone who perceives an activity as less
intense would be more physically active. A positive correlation would be expected for
the relationship between economy and RPE, assuming that a more economical person
(i.e., lower oxygen consumption) would perceive an activity as less intense.
Relationship between economy and physical activity. It is possible that economy
could affect physical activity behavior, or that physical activity behavior could affect
economy. Although the expected direction of correlations between economy and
physical activity is negative, most of our relationships between PDPAR scores and
economy were positive. None were significant (Appendix G).

Questions from the YRBSS that are most likely to show a good picture of
physical activity are numbers 1, 4, 6, and 7/8. Participants in this sample were more
vigorously (question 1) and moderately (question 4) physically active and played on more
sports teams (questions 7 and 8) than both the national and state averages (see Table 5.1).
Fewer participants in this sample participated in 20 minutes or more of physical activity
during physical education class (question 6) than the national or state averages.
However, only 39% of this sample did pg: take physical education classes, compared to

48% of the national and 72% of the Michigan samples.

72

The highest (negative) correlations between economy and YRBSS questions were
significant and occurred for treadmill stage 3, although questions number 1, 2, 3, and 7
correlated well with economy in all three treadmill stages. Question 1, which asks on
how many of the last 7 days the girls performed vigorous activity, correlated best for all
treadmill economy stages (F -0.4, F -0.5, F -0.5 for stages 1, 2, and 3, respectively,
P<0.05). Question 7, which asks on how many school sports teams the girls participated,
displayed the next best correlation coefficients (F ~04 for both stage 2 and stage 3,
P<0.05). These results indicate that the more economical girls spent the most time being
physically active in the week prior to data collection and participated on the most school
sports teams. Whether good economy makes one more likely to be active (including
Sports), or whether being active and playing sports improves one’s economy cannot be
discerned from these data.

Table 5.1. Prevalence of behaviors in YRBSSquestions
Current sample (%) Nationwide average (%) Statewide average (%)

 

Question 1 77 57 57
Question 4 71 16 21
Question 6 52 70 80
Question 7 90 48 53

 

NOTE: Data for questions 1, 6, and 7 is from Kann et al., 2000 32

Data for question 4 is from Kann et al., 1998 31

When examining the relationship between exercise economy and physical
activity, one must consider that the PDPAR scores did not correlate well with the YRBSS
scores. It seems reasonable to expect fairly high, positive correlations between the
PDPAR minutes and YRBSS answers, assuming that the girls who reported the most
activity minutes were active on most days (or minutes) and participated on the highest

number of school-sponsored sports teams. However, the correlations between the

73

PDPAR and YRBSS were mostly negative and ranged from ~F -0.2 to 0.2. A possible
reason for this apparent lack of correlation is the fitness level of the study participants.
Participants would often classify events such as soccer games as “medium” or even
“light”. It is likely that some activities the girls performed might be classified as more
intense by girls who were less fit. In fact, it is possible that some physical activities were
not considered as such and thus, were not reported on the PDPAR, despite prompting
from the investigator. In contrast, the YRBSS questions asked specifically about
activities in which participants “breathed hard” and “sweat”, and perhaps were a more
valid measure of physical activity in our sample.

It is possible that the PDPAR was ndt a good instrument to use with this
population. It is also possible that there is no relationship between economy and physical
activity. However, the increase in the correlation coefficient between treadmill economy
and YRBS Question 1 with increasing exercise intensity suggests that perhaps these two
variables are related. If a more objective monitoring device had been used to assess
physical activity, different results may have been found. It is also possible that the large
(and typical) standard deviations found in PDPAR scores masked a true relationship.
Further, if the treadmill speed (particularly in stage 3) truly was inefficient for several of
the girls in the sample (as the data indicate), then a better relationship between economy
and physical activity might be found at a slightly faster pace.

Relationship between RPE and physical activity. It is difficult to infer much from the
correlation coefficients between RPE and physical activity. There were only 9-10
participants for each RPE-treadmill stage combination. Nevertheless, there were a few

significant correlation coefficients between RPE and the PDPAR minutes. Borg RPE for

74

stages 1 and 2 correlated moderately (F -0.5 and F -0.4, P<0.05) with number of very
light activity minutes and number of light activity minutes, respectively. OMNI RPE for
stage 1 correlated moderately with total PDPAR minutes (F -0.5, P<0.05). OMNI RPE
for stages 2 and 3 correlated moderately (F -0.4, P<0.05) with number of minutes spent
in light activity.

There were many significant (P<0.05) correlation coefficients in the expected
direction for RPE versus the YRBSS questions. The first YRBSS question correlated
moderately (F -0.5) with OMNI RPE for the second stage. The fourth YRBSS question
correlated moderately with Borg RPE for stage 2 (F -0.6). The sixth YRBSS question
correlated moderately (F -0.4) with OMNI RPE for stage 1 and well (F -0.7) with
OMNI RPE in stage 3. The seventh YRBSS question correlated moderately (F -0.4)
with Borg RPE in stage 2 and OMNI RPE in stage 3.

Given the exploratory nature of this study, it is difficult to state exactly what these
correlations mean in terms of a true impact of a girl’s RPE on her on physical activity
behavior. Since the first two treadmill stages were of low and moderate intensity, one
might expect a good relationship between RPE during these stages and very light, light,
and medium minutes on the PDPAR. However, there was only a moderate relationship
between Borg RPE for stage 1 and very light activity minutes, and Borg RPE for stage 2
and light activity minutes. It is interesting to note that OMNI RPE for stage 1 had a
moderate relationship with overall PDPAR activity minutes. Thus, there is some
evidence that a relationship exists between RPE and physical activity, but the topic

warrants further investigation with a more focused design and more study participants.

75

Relationship between economy and RPE. Out of six possible correlations, only three
were in the expected @ositive) direction. Only one of these three was significant
(P<0.05), which was OMNI RPE during stage 3 (F07). This represents an interesting
finding, as the girls who were less economical duringjogging reported higher RPEs. The
small number of participants (n=9) included in this correlation is a limitation, but this
topic also warrants further investigation with more participants.

There is good potential for further investigation into these relationships.
Economy should be examined over a wider range of treadmill speeds. Physical activity
Should be measured objectively with motion-detecting devices such as a pedometer or the
Computer Science and Applications (CSA) accelerometer. Use of a portable metabolic
analyzer to assess economy in a field setting may also provide a better correlation
between economy and physical activity, Since most life activities do not occur while an
individual walks and jogs on a treadmill or rides a cycle ergometer. Also, sampling
designs must better insure that participants represent the intended population. Although
it is difficult to find girls who are less fit and/or active to participate in exercise studies,
their data are important to consider. This is important from a public health perspective
Since it is the less fit/active girls who are most in need of physical activity.

There are also methods to consider that may help better examine those
relationships between RPE and economy. A large sample would allow collection of
several data points for an RPE assessment at a given stage. This way, one RPE
measurement would not affect subsequent measurements, but sufficient data would be
available to make valid comparisons. RPE could also be assessed in combination with

measures obtained from portable metabolic analyzers and objective physical activity

76

measures. RPE should be assessed during activities in field settings to determine if
traditional laboratory testing has provided accurate and generalizable data.
Developmental psychologists may need to be consulted in order to determine how
children learn to perceive exertion. Despite creation of child-specific scales, the idea that
children can detect and report perceived exertion in the same way as adults is still applied
in all research to date.

The best method for examining all these relationships is to conduct longitudinal
studies. Longitudinal studies would provide the best assessment of how variables change
over time. They would also help researchers understand if improvements in one’s
economy of movement or physical activity behaviors can be made.

In summary, it appears that the concepts of exercise economy and RPE as
determinants of physical activity may be warranted as topics for future investigations.
Physical activity is a multi-faceted behavior that is affected by many factors. Factors
affecting one individual may not affect another. If physical education instructors can
improve children’s motor skill performance at young ages, they may help to improve
children’s economy. Although the correlations found in this study do not imply
causality, it is possible that improvement in exercise economy may affect RPE and/or
physical activity behavior. Without longitudinal research spanning from childhood
through adolescence, it is difficult to project which factors are the most important in

convincing children to lead healthy lifestyles as adults.

77

Figure 5.1 Proposed model for relationships among economy, RPE, and physical
activity

 

 

Determinants
-Sex
-Ethnicity
-Age
-Achievement

orientation
-Intention
-Perceived physical

competence
-Depression

-Sports participation
-Support from
parents and peers

  
 

 

Physical
Activity

 
 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Economy

Determinants /
-Age Determinants
-Sex -Central factors
-Body size -Peripheral factors
-Maturation -Exercise mode
-Pulmonary function -Exercise intensity
-Cardiac function -Exercise frequency
-Resting energy -Exercise duration
expenditure -Metabolism
-Substrate utilization -Cognition
-Running mechanics -Measurement scale
-Training

 

 

 

78

Appendix A

79

 

Appendix A

Borg’s RPE Scale

6-
7- Very, very light
8-

9- Very light

10-

11- Fairly light

12-

13- Somewhat hard
14- l

15- Hard

16-

17- Very hard

18-

19- Very, very hard
20-

80

Appendix B

81

Appendix B

Children’s Effort Rating Table (CERT)

1 ...... Very, very easy

2 ...... Very easy

3 ...... Easy

4 ...... Just feeling a strain
5 ...... Starting to get hard
6 ...... Getting quite hard
7 ...... Hard

8 ...... Very hard

9 ...... Very, very hard

10.....So hard I am going to stop

82

Appendix C

Appendix C

Children’s OMNI Scale of Perceived Exertion

3 getting more
tired

 

2
a little
fired

0
not tired

 

84

at all

Appendix D

85

 

Appendix D

Cart and Load Effort Rating (CALER) Scale

d9... 9 meow
8; Ema om

BE .95

EB now
2 waESm

 

 

r86

Appendix E

87

 

Appendix E
Consent Form
Title: Running economy in adolescent females

1, , the parent / guardian of grant
permission to enroll my child in a study to examine the variables which affect running
economy. The entire amount of time required per participant is about two hours total.

As children grow, the amount of energy required to do certain types of exercise (e.g.,
walking, jogging) changes. The reasons for this remain unclear. The purpose of this
study is to examine these changes between adolescents of different ages.

On both test days, the participant’s height and weight will be measured prior to any
testing that takes place. On the first day, the participant will undergo pulmonary fimction
testing. This testing involves breathing into a device called a spirometer, which will
measure lung volume. A new, separate mouthpiece is used for each child. In addition,
each participant will answer questions pertaining to physical activity behavior, physical
activity self-efficacy, and background, including date of onset of menstruation and
menstrual history. These questions will be asked by a female investigator. Finally, each
child will have a practice session on the treadmill to become familiar with treadmill
walking and running. The practice session will last 10-15 minutes, or until she is
comfortable on the treadmill. She will also wear a heart rate transmitter belt around her
chest in order to familiarize her with use of the instrument.

On the second test day, each participant’s sitting height, shoulder and hip breadth, and
skinfold thicknesses of her calf, arm, hip, thigh, and back will be measured (by a female
investigator). She will again be asked to answer questions about her physical activity
behavior. In addition, she will perform an exercise test. The exercise test will involve
walking and jogging on a motorized treadmill. Prior to the test, the participant will sit for
10 minutes while seated in a chair on top of the treadmill. The treadmill test itself will
begin with a six minute warrn-up at 2 mph, 0.0% grade. Next, she will walk at 3.0 mph,
5.0% grade for six minutes. Then she will jog at 4.5 mph, 0.0% grade for six minutes.
After 18 minutes of walking and jogging, the grade will increase 2.5% each minute until
she can no longer keep up. Then the test will be stopped when the participant indicates
that she wishes to stop.

During the treadmill test, breath will by analyzed to determine the amount of oxygen
each participant uses during the test. This will involve her inhaling room air through a
mouthpiece and exhaling expired air into the same mouthpiece, which will be connected
toga hose that will channel the air into a machine for analysis. Also, she will be wearing
an elastic chest strap that will measure her heart rate during the test.

I understand that there are possible risks in the study. However, they are very minimal.
No one can absolutely guarantee that any person, no matter how fit, or how young, will

88

not suffer a heart attack or related problem during the test. The risk of this in a
population with KNOWN heart disease is less than 1 in 100,000 hours of testing. It is
likely that the risk is much less in healthy adolescents. All researchers involved with the
testing have been trained in cardiopulmonary resuscitation (CPR).

If the participant has any medical condition (e.g., asthma, orthopedic problems, heart
condition, etc.) which might be a reason not to exercise, the investigators should be
informed prior to any testing.

A record of each child’s participation in the study will be kept in a confidential file at
Michigan State University. Each participant’s privacy will be protected to the maximum
extent allowable by law. She is free to withdraw from the study at any time and for any
reason. Such withdrawal will notjeopardize her future treatment. There will be no cost
for participating in this study and each participant will receive a movie pass and $10.00 at
the end of the test to defer the cost of transportation.

If one is injured as a result of participation in this research project, Michigan State
University will provide emergency care if necessary. If the injury is not caused by the
negligence of MSU, you are personally responsible for the expense of this emergency
care and any other medical expenses incurred as a result of this injury. My signature
below, and my child’s signature, acknowledges her voluntary participation in this
research project. Such participation does not release the investigators, institutions,
sponsor, or granting agency from their professional and ethical responsibility to my child.

If there are any questions regarding this research project, including those regarding
participants’ rights as human subject, feel free to contact any of the following individuals:
Karin Allor, MS, Project Coordinator 355-4734; James Pivarnik, Ph.D., Project
Supervisor 353-3 520; David Wright, Ph.D., Chair of University Committee on Research
Involving Human Subjects 355-2180.

I have read this consent form and received answers in those areas which were not clear. I
am willing to give consent for my child to participate in this study.

 

 

Study Participant Signature Date

 

 

Phone Number

 

 

 

 

 

Address
Parent/ Guardian Signature Date
Investigator Signature Date

89

 

Appendix F

90

Appendix F

Approval form from University Committee on Research Involving Human Subjects

 

517355-21”

FAX: 517353-2976

m maumm
E-Mail: mm.“

bwwem
”UMM'

mum
Noam

MICHIGAN STATE

UNIVERSITY

October 26. 2000

'TO: James PIVARNIK
3 LM. Sports Circle

RE: IRB# 00-607 CATEGORY:2-D
APPROVAL DATE: October 19, 2000

TITLE: RUNNlNG ECONOMY lN ADOLESCENT FEMALES

The University Committee on Research Involving Human Subjects' (UCRIHS) review oi this
project is complete and I am pleased to advise that the rights and welfare of the human
subjects appear to be adequately protected and methods to obtain intormed consent are
appropriate. Therefore. the UCRIHS approved this project.

RENEWALS: UCRIHS approval is valid tor one calendar year, beginning with the approval
date shown above. Projects continuing beyond one year must be renewed with the green
renewal lorm. A maximum of four such expedited renewals possible. Investigators wishing to
continue a project beyond that time need to submit it again for a complete review.

REVISIONS: UCRIHS must review any changes in procedures involving human subjects. prior
to initiation ol the change. If this is done at the time of renewal, please use the green renewal
.torm. To revise an approved protocol at any other time during the year, send your written
request to the UCRIHS Chair, requesting revised approval and referencing the project's IRBtt
and title. Include in your request a description of the change and any revised instruments,
consent forms or advertisements that are applicable.

PROBLEMS/CHANGES: Should either at the following arise during the course at the work,
notify UCRIHS promptly: 1) problems (unexpected side effects, complaints, etc.) involving
human subjects or 2) changes in the research environment or new information indicating
greater risk to the human subjects than existed when the protocol was previously reviewed and

approved.

If we can be of further assistance. please contact us at 517 355-2180 or via email:
UCRIHS@msu.edu. Please note that all UCRll-lS forms are located on the web:
http://wwwmsu.eo'uluseriucri‘ns

Sincerel ,
I .u/ I LW

Ashir Kumar, MD
Interim Chair, UCRIHS

AK: br

CC: Karin Allor
3 1M. Sports Circle

91

Appendix G

92

 

Appendix G

Correlation Matrices

 

out; Sq 5:95.38 5908 "outdo-80> do; cow-ta moon :5. uon coca—nuance combs n<mm<NO> .mo-a comm-Em anon n<mm
.332— 955 "£325 528:?” 3c mo E3 u Eo-Ezm .5985 02:3 "035m £6.35 328855 "8885 6:222: an own Human—z ”ME-OZ

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2.- m8. #8.- an E... E.- w..- EQ- m:. E. S.- moo. 2:.- E. >>2
SN.- 27 :9. Mam. a8. 02.. oz. 2N. Em. SN. 89. 2. mg. m8. 0?.
8m. 8m. o8. 8m. 8,”. so- 8m. 2:. 2:. 2:.- oos. 8m. 2o- oS. E1
_ 2:. OS. 95.- :9- m3- 8m. 8”. 5.- 8? 8m. 2:.- SN. 5.. a:
a Na. m3- 3? 8m.- SN. SN. 02.. oe. cow. 8? Ba. 2:.- 2.132;
a 8m. 2:. 8N.- SW. 08. o8. 8c- :fi. 8m. 2:. 2:. $82:
_ «3.. SN; M5. 21 an. E: 8N.- MS. ”2.- E. <li
_ m8.- Eo. E. E. ”2. m8.- 8? z: .- as. 332.: .
_ 8m. 8m.- 8? 2:. 8o. 08.. 08. 2:. . BEE-.2
_ :m. as. $1- :5. Ba. 08.- 37 22:... "I
_ 3m. mm:- 8? SN. m:. e5.- 8:51
a m3- 8? an 2. E. .52851
_ 8N. Sn- 8. SN- 52.231
a 8m.- SN. 2: .- ~o>cm
_ 3.? SN. Em
_ :o. .552
_ ow<
out: «an 2 a; flaw .Eeua 5:: ~o>
m: a?» \No> <2 gm 2.2% 25 :55 -E 23 3 ES 5:52 m.3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

omfim 5:68: Em 2: E >Eocooo wage-mm 322$ 3388 how anE 5:20:00 Ad 053-

’93

0223 23 25222222222228 combs. nov5m\~o> .288 com-2.22222 2223 222222 222 28222222225228 combs n<mm<No> .2298 cum-2.222.». .2225 n<mm
2223022 madam n222w22222m $2822.53 922.2 L2o E25 u 2.282222% 2222:3222 22222222 "22:25 222228222 22222228022222 ":8me 6228522222 222 own ".2292 HmE-OZ

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2.- mg. 2.8.- 222. 22:. :2.- 2- E2.- 22. En 8.. Be. 2:.- 3. >22
2:.- 2:. 222. 2%. 25. 2:.- 22. RN. E. 222. 2:.- mm. 223. m8. u>:
2:.. 2:. 2m. 2;. 2:. 2:.. 22m. SN. 82. 22.- 2:. 25. 2:.- 2:. m>
2 2:. 2:. 2:.- 2:”.- 2:.- 8m. 2:. .22.- 22? ace. 2:. 2K. 222. MEI
2 2x. 222.. 2:.- 22m.- 222. 2:2. 2:. 2:.- 222. 2:. 2:. 2:.- 2.2-.228?
2 22m. 2:.- SN.- 22». 22”. 2:. 2:.. 2:. 22m. 222. 2:. 522282
2 82. SN- 22:. :1 2m. 2.: .- 8.- m3. 222 .- $2. <21
2 2:.- 2222 3. %2. ”N2. 222.- 25.. 2222- 08. 2.222222
2 228.. 221- 22.. 228. 28. 2: .- 2:. 2:. .2222 2.2: 1
2 :m. and 2:.-.- 222. So. 022.- 22.- 2.2825
2 Em. 2:.- -.- EN. 22. as.- 2222222
2 228.- m2.- mmm. N2. N2. 22522222
2 223.. So- 8. 2:2.- 2:22:23
2 2:.- 25 2:.- 3>ch
2 SN- 2-N. 22221
2 :22. 5:522
2 mmfli
«22222 222mm 22.22. 222-2 4222 .528 I322. ~o>
:22 23> 28> <22 .5 9225 22:5 2222222 -222 ~o> am 22222 525:2 a?

 

 

owe; 222226322 2:288 222 222 28202280 mason-um $023.2 22222228022 20.2 22-222: 220222250 .Nd 0322.2-

94

2222226 22222 2822222222228 220» 228 novt222~o> 8222 8222226 22:22.2 222222 22222 25222222222228 :0» 228 u<mm<~o> 8222 822-2225 22222 u<mm
.222m222 9222222 "222w2u2222m mus-22222222 32.2 20 Eat n 2220222226 2222222222 22222222 ”0222225 222228222 222-2222822222 "2222220225 22922222222 222 um“: i 2222202 .mE-OZ

 

omfim 22222222822 22.22222 0222 22 2222202208 $2228.23 22028.2 22222228022 20.2 2222222222 2202222202200 .00 22222-2-

95

 

“WEE

”2.222an2 0222 2222 2222222222 2282922222 22222 2222a 2222222222 .222»: .222w22 .229» .20 8222222222 09229222 u WEI-93> #222922 222 220 22322022
225292222 .222 8222222222 $22296 u @253.QO 2822222., owns-mam 222 m-2 2220 2:222 m2 m2om 2m 222222 .N .2 8mg 222 2222222280 822 NO>mw2m-NO>2w2m “-22.02

 

N.-

NO>nw2m

 

N.-

NO>Nw2m

 

2.-

23: 2m

 

2.-

h.

2.-V

m.

.w.

m.

«.22:

 

2.v

N.

m.

N.-

2.v

as:

 

2..-

22.-

2...-

2.-

2.v

“>23

 

22..

N.-

2.V

N.-

m.-

2.v

$22-22»

 

2.-

c.

2.-

m.

2.

N.

méggm

 

2.v

2.v

2.-

m.-

2.-

N.-

N.-

2.-

2.

ama>

 

m.-

2.-

2.-

T

m.-

22.-

2.-

2..-

N—m'fN—‘N
I.‘ .0.

5mg?

 

2V

2.-v

2..-

2.-

m.-

m.

2.-

ema>

 

2.V

2.v

m.-

m.

n.

N.

N.-

mmm¢>

 

N.-

N.-

2.-

2.-

2.

2.-

o.-

N.-

2.

O .
I 0 '

9wm¢>

 

N.-

2.-

2.-

2.-

m.-

2.-

m.-

m.-

2

m.-

m.-

nmm¢>

 

m.-

N.-

N.-

T

22.

w.-

2.-

m.-

m.-

N.-

Nmm¢>

 

N.-

N.-

.-v

—1

2.-

m.-

2.V

m.-

#020062

V

N.-

m.-

m.-

m.-

T

2mm->

 

”>22:

 

 

”.32

 

”>224

 

”>221;

 

w;
”:59.—

 

n
mzzc

 

N
27220

 

 

22
M25

 

N
”.222:

 

2
M22222

 

3>
$5

 

8>
222m

 

8>
222m

 

 

 

222232022 282222222 222222 {”72 22282208 .20 22222302222202 20.2 222222222 282228.50 v.0 22222-2-

96

Appendix H

97

 

Appendix H

Previous Day Physical Activity Recall (PDPAR)
Activities Scale

On the next page is a scale which records the main activities you
did yesterday. Please be sure to write on the scale what day
yesterday was.

1. For each time period write in the number(s) of the main
activities you did in the boxes on the time scale.

2. Then try to remember how HARD you felt like you were
working and how LONG you did the activity. Write how LONG
you did the activity in the box that corresponds to how HARD
you felt like you were working in the boxes on the time scale.

. Very Light- Slow breathing little or no movement

 

Please be as accurate as possible.

98

 

 

 

Activity Numbers

Eating

1. Meal
2. Snack
3. Cooking

Sleep/Bathing
4. Sleeping
5. Resting
6. Shower/bath

Transportation

7. Ride in car/bus

8. Travel by
walking

9. Travel by bike

Work/School

10. Job (list)

11. Homework

12. Household
chores (list)

Spane Time

13. Watch TV/Play
video games

14. Go to
movies/concert

15. Liam to music

16. Talk on phone

17. Hang around

18. Shopping

19. Hobby (list)

20. Other (list)

Physical Activities ’

21. Bicycle (easy

pace)

22. Walk (slowly)

23. Dance (for fun)

24. Swim (for fun)

25. Skateboard

26. Play organized
span (slow
pace)

27. Play individual
sport (slow
pace)

28. Bicycle (fast)

29. Walk (fast)

30. Jog/run

31. Swim laps

32. Lift weights

33. Aerobic dance

34. Play organized
span (fast pace)

35. Individual
exercise (fast
pace

 

 

Previous Day Physical Activity Recall

2. Write how long

you did the activity

in the box corresponding
to how hard the activities

U were
U

1. Put activity numbers
in this column

Time Activity Very

7:00
7:30
8:00
8:30
9:00
9:30
0:00
10:30
11:00
11:30
12:00
12:30
1:00
1:30
2:00
2:30
3:00
3:30
:00
:30
:00
:30
:00
:30
:00
:30
:00
:30
.30
0:00
0:30

 

 

Write the day of the week you did these activities

99

 

Appendix I

100

Appendix I
Youth Risk Behavior Surveillance System Questions
. On how many of the past 7 days did you exercise or participate in sports activities for

at least 20 minutes that made you sweat and breathe hard, such as basketball, jogging,
fast dancing, swimming laps, tennis, fast bicycling, or similar aerobic activities?

a. 0 days e. 4 days
b. 1 day f. 5 days
c. 2 days g. 6 days
d. 3 days h. 7 days

. On how many of the past 7 days did you do stretching exercise, such as toe touching,
knee bending, or leg stretching?

a. 0 days e. 4 days
b. 1 day f. 5 days
c. 2 days g. 6 days
d. 3 days h. 7 days

. On how many of the past 7 days did you do exercises to strengthen or tone your
muscles, such as push-ups, sit-ups, or weight lifting?

a. 0 days e. 4 days
b. 1 day f. 5 days
0. 2 days g. 6 days
(1. 3 days h. 7 days

. On how many of the past 7 days did you walk or bicycle for at least 30 minutes at a
time (including walking or bicycling to or from school)?

a. 0 days e. 4 days
b. 1 day f. 5 days
c. 2 days g. 6 days
d. 3 days h. 7 days

. In an average week when you are in school, on how many days do you go to physical
education (PE) classes?

a. 0 days 0. 2 days e. 4 days

b. 1 day (1. 3 days f. 5 days

. During an average physical education (PE) class, how many minutes do you spend
actually exercising or playing sports?

I do not take PE

Less than 10 minutes

10 to 20 minutes

21 to 30 minutes

More than 30 minutes

rip-99‘s»

101

 

During the past 12 months, on how many sports teams run by your school did you
play? (Do not include PE classes)

a. 0 teams c. 2 teams

b. 1 team d. 3 or more teams

During the past 12 months, on how many sports teams run by organizations outside
your school did you play?

a. 0 teams c. 2 teams

b. 1 team (1. 3 or more teams

 

102

REFERENCES

103

10.

11.

REFERENCES

Allor, K. M., J. M. Pivarnik, L. J. Sam, and C. D. Perkins. Treadmill economy in
girls and women matched for height and weight. J Appl Physiol. 89:512-516,
2000.

Ariens, G. A., W. van Mechelen, H. C. Kemper, and J. W. Twisk. The
longitudinal development of running economy in males and females aged between
13 and 27 years: the Amsterdam Growth and Health Study. Eur J Appl Physiol.
76:214-220, 1997.

Armon, Y., D. M. Cooper, and S. Zanconato. Maturation of ventilatory responses
to 1-minute exercise. Pediatr Research. 29:362-368, 1991.

Astrand, P. 0. Experimental Studies of Physical Working Capacity in Relation to
Sex and Age. Copenhagen: Ejnar Munksgaard, 1952.

Bailey, S. P. and R. R. Pate. Feasibility of improving running economy. Sports
Med. 12:228-236, 1991.

Baranowski, T., C. Bouchard, O. Bar-Or, T. Bricker, G. Heath, S. Kim, R. M.
Malina, E. Obarzanek, R. R. Pate, W. B. Strong, B. Truman, and R. Washington.
Assessment, prevalence, and cardiovascular benefits of physical activity and
fitness in youth. Med Sci Sports Exerc. 26:8237-247, 1992.

Bar-Or, O. Age-related changes in exercise perception. In: Physical Work and
Eflort. G. Borg (Ed.) Oxford: Pergamon Press, 1977, pp. 255-266.

Bar-Or, 0. Pediatric Sports Medicine for the Practitioner. New York: Springer-
Verlag, 1983

Bar-Or, 0. Rating of perceived exertion in children and adolescents: clinical
aspects. In: Psychophysics in action. G. Ljunggren and S. Domic (Eds) New
York: Springer-Verlag, 1989, pp. 105-1 13.

Bar-Or, O. and D. S. Ward. Rating of perceived exertion in children. In: Advances
in Pediatric Sport Sciences. 0. Bar-Or (Ed.) Champaign, IL: Human Kinetics,
1989.

Billewicz, W. 2., H. M. Fellowes, and A. M. Thomson. Post-menarcheal

menstrual cycles in British (Newcastle upon Tyne) girls. Ann Human Biol. 7: 177-
180, 1980.

104

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

Bland, J. M. and D. G. Altman. Statistical methods for assessing agreement
between two methods of clinical measurement. Lancetz307-310, 1986.

Borg, G. Psychophysical bases of perceived exertion. Med. Sci. Sports Exerc.
14:377-381, 1982.

Casperson, C. J., K. E. Powell, and G. M. Christenson. Physical activity, exercise
and physical fitness: definitions and distinctions for health-related research.
Public Health Rep. 100: 126-130, 1985.

Cooper, D. M., M. R. Kaplan, L. Baumgarten, D. Weiler-Ravell, B. J. Whipp, and
K. Wasserman. Coupling of ventilation and CO2 production during exercise in
children. Pediatr Research. 21:568-572, 1987.

Daniels, J ., N. Oldridge, F. Nagle, and B. White. Differences and changes in V02
among young runners 10 to 18 years of age. Med Sci Sports Exerc. 10:200-203,
1978.

Dempsey, J. A., C. A. Harms, and D. M. Ainsworth. Respiratory muscle perfusion
and energetics during exercise. Med Sci Sports Exerc. 28:1123-1128, 1996.

Dishman, R. K. Prescribing exercise intensity for healthy adults using perceived
exertion. Med Sci Sports Exerc. 26: 1087-1094, 1994.

Ebbeling, C. J ., J. Hamill, P. S. Freedson, and T. W. Rowland. An examination of
efficiency during walking in children and adults. Pediatr Exerc Sci. 4:36-49,
1992.

Eston, R. G., K. L. Lamb, A. Bain, A. M. Williams, and J. G. Williams. Validity
of a perceived exertion scale for children: a pilot study. Perceptual and Motor
Skills. 78:691-697, 1994.

Eston, R. G., G. Parfitt, L. Campbell, and K. L. Lamb. Reliability of effort
perception for regulating exercise intensity in children using the cart and load
effort rating (CALER) scale. Pediatr Exerc Sci. 12:388-397, 2000.

Eston, R. G. and J. G. Williams. Exercise intensity and perceived exertion in
adolescent boys. Brit. J. Sports Med 20:27-30, 1986.

Forster, M. A., G. R. Hunter, D. J. Hester, D. Dunaway, and K. Shuleva. Aerobic
capacity and grade-walking economy of children 5-9 years old: A longitudinal
study. Pediatr Exerc Sci. 6:31-38, 1994.

Freedson, P. S., V. L. Katch, T. B. Gilliam, and S. MacConnie. Energy

expenditure in prepubescent children: influence of sex and age. Am. J. Clin. Nutr.
34:1827-1830, 1981.

105

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

Frost, G., O. Bar-Or, J. Dowling, and C. White. Habituation of children to
treadmill walking and running: Metabolic and kinematic criteria. Pediatr Exerc
Sci. 72162-175, 1995.

Frost, G., J. Dowling, K. Dyson, and O. Bar-Or. Cocontraction in three age
groups of children during treadmill locomotion. J. Electromyogr. Kinesiol. 7: 179-
186, 1997.

Gillach, M. C., J. F. Sallis, M. J. Buono, P. Patterson, and P. R. Nader. The
relationship between perceived exertion and heart rate in children and adults.
Pediatr. Exerc. Sci. 1:360-368, 1989.

Hausdorff, J. M., L. Zemany, C. K. Peng, and A. L. Goldberger. Maturation of
gait dynamics: stride-to-stride variability and its temporal organization in
children. J Appl Physio]. 86:1040-1047, 1999.

Howley, E. T., D. R. Bassett, and H. G. Welch. Criteria for maximal oxygen
uptake: review and commentary. Med Sci Sports Exerc. 27:1292-1301, 1995.

Kanaley, J. A., R. A. Boileau, B. H. Massey, and J. E. Misner. Muscular
efficiency during treadmill walking: The effects of age and workload. Pediatr
Exerc Sci. 1:155-162, 1989.

Kann, L., S. A. Kinchen, B. 1. Williams, J. G. Ross, R. Lowry, J. Grunbaum, P. S.
Blumson, J. L. Collins, and L. J. Kolbe. Youth risk behavior surveillance: United
States 1997. MMWR. 4721-89, 1998.

Karin, L., S. A. Kinchen, B. I. Williams, J. G. Ross, R. Lowry, J. Grunbaum, and
L. J. Kolbe. Youth risk behavior surveillance: United States 1999. MMWR. 49:1-
96, 2000.

Kitamura, K., C. R. Jorgensen, F. L. Gobel, H. L. Taylor, and Y. Wang.
Hemodynamic correlates of myocardial oxygen consumption during upright

exercise. J App] Physio]. 32:516-522, 1972.

Kolbe, L. J ., L. Karin, and J. L. Collins. Overview of the youth risk behavior
surveillance system. Public Health Rep. 108:2-10, 1993.

Krahenbuhl, G. S., D. W. Morgan, and R. P. Pangrazi. Longitudinal changes in
distance-running performance of young males. Int J Sports Med. 10:92-96, 1989.

Krahenbuhl, G. S. and R. P. Pangrazi. Characteristics associated with running
performance in young boys. Med Sci Sports Exerc. 152486-490, 1983.

Krahenbuhl, G. S., J. S. Skinner, and W. M. Kohrt. Developmental aspects of
maximal aerobic power in children. Exerc. Sport Sci. Rev. 132503-538, 1985.

106

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

Lamb, K. L. Children's ratings of effort during cycle ergometry: An examination
of the validity of two effort rating scales. Pediatr. Exerc. Sci. 7:407-421, 1995.

Lamb, K. L. Exercise regulation during cycle ergometry using the children's effort
rating table (CERT) and rating of perceived exertion (RPE) scales. Pediatr. Exerc.
Sci. 8:337-350, 1996.

Lamb, K. L. and R. G. Eston. Effort perception in children. Sports Med. 23:139-
148, 1997.

Lamb, K. L., R. G. Eston, and D. Corns. Reliability of ratings of perceived
exertion during progressive treadmill exercise. Br. J. Sports Med. 332336-339,
1999.

MacDougall, J. D., P. D. Roche, O. Bar-Or, and J. R. Moroz. Maximal aerobic
capacity of Canadian schoolchildren: Prediction based on age-related oxygen cost
of running. Int J Sports Med. 4:194-198, 1983.

Maffeis, C., Y. Schutz, F. Schena, M. Zaffanello, and L. Pinelli. Energy
expenditure during walking and running in obese and nonobese prepubertal
children. J. Pediatr. 123:193-199, 1993.

Mahon, A. D., G. E. Duncan, C. A. Howe, and P. Del Corral. Blood lactate and
perceived exertion relative to ventilatory threshold: boys versus men. Med. Sci.
Sports Exerc. 29:1332-1337, 1997.

Mahon, A. D., J. A. Gay, and K. Q. Stolen. Differentiated ratings of perceived
exertion at ventilatory threshold in children and adults. Eur. J. Appl. Physiol.
18:115-120,1998.

Mahon, A. D. and M. L. Marsh. Reliability of the rating of perceived exertion at
ventilatory threshold in children. Int. J. Sports Med. 13:567-571, 1992.

Mahon, A. D. and M. L. Ray. Ratings of perceived exertion at maximal exercise
in children performing different graded exercise test. J. Sports Med. Phys. Fitness.
35:38-42, 1995.

Malina, R. M. Physical activity and fitness of children and youth: questions and
implications. Med. Exerc. Nutr. Health. 42123-135, 1995.

Malina, R. M. Tracking of physical activity and physical fitness across the
lifespan. Res Q Exerc Sport. 67:48-57, 1996.

Malina, R. M. and C. Bouchard. Growth, Maturation, and Physical Activity.
Champaign, IL: Human Kinetics, 1991

107

 

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

Martin, P. E. and D. W. Morgan. Biomechanical considerations for economical
walking and running. Med Sci Sports Exerc. 24:467-474, 1992.

McArdle, W. D., F. I. Katch, and V. L. Katch. Exercise physiology: energy,
nutrition, and human performance. 4 ed. Baltimore, MD: Williams & Wilkens,
1996, 850.

McCormack, W. P., K. J. Cureton, T. A. Bullock, and P. G. Weyand. Metabolic
determinants of l-mile run/walk performance in children. Med Sci Sports Exerc.
23:611-617, 1991.

McManus, A. M., N. Armstrong, B. Kirby, and J. Welsman. Ratings of perceived
exertion in prepubescent girls and boys. In XIXth International Symposium of
Pediatric Work Physiology. Exeter: E & FN Spon, pp. 253-257, 1997.

Morgan, D. W. and M. Craib. Physiological aspects of running economy. Med Sci
Sports Exerc. 242456-461, 1992.

Morgan, D. W., P. E. Martin, and G. S. Krahenbuhl. Factors affecting running
economy. Sports Med. 72310-330, 1989.

Noble, B. J. and R. J. Robertson. Perceived Exertion. Champaign, IL: Human
Kinetics, 1996, 320.

Pate, R. R., C. A. Macera, S. P. Bailey, W. P. Bartoli, and K. E. Powell.
Physiological, anthropometric, and training correlates of running economy. Med
Sci Sports Exerc. 24:1128-1 133, 1992.

Powell, K. E., P. D. Thompson, C. J. Casperson, and J. S. Kendrick. Physical
activity and the incidence of coronary heart disease. Ann. Rev. Public Health.
8:253-287, 1987.

Robertson, R. J ., F. L. Goss, N. F. Boer, J. A. Peoples, A. J. Foreman, I. M.
Dabayebeh, N. B. Millich, G. Balasekaran, S. E. Riechman, J. D. Gallagher, and
T. Thompkins. Children's OMN I scale of perceived exertion: mixed gender and
race validation. Med Sci. Sports Exerc. 32:452-458, 2000.

Robertson, R. J. and B. J. Noble. Perception of physical exertion: Methods,
mediators, and applications. Exerc. Sport Sci. Rev. 25:407-452, 1997.

Rogers, D. M., K. R. Turley, K. I. Kujawa, K. M. Harper, and J. H. Wilmore.
Allometric scaling factors for oxygen uptake during exercise in children. Pediatr
Exerc Sci. 7:12-25, 1995.

108

 

 

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

Rogers, D. M., K. R. Turley, K. I. Kujawa, K. M. Harper, and J. H. Wilmore. The
reliability and validity of running economy in 7-, 8-, and 9-year old children.
Pediatr Exerc Sci. 6:287-296, 1994.

Rowland, T., L. Cunningham, L. Martel, P. Vanderburgh, T. Manos, and N.
Charkoudian. Gender effects on submaximal energy expenditure in children. Int J
Sports Med. 182420-425, 1997.

Rowland, T., K. Miller, P. Vanderburgh, D. Goff, L. Martel, and L. F errone.
Cardiovascular fitness in premenarcheal girls and young women. Int J Sports
Med. 21:117-121, 2000.

Rowland, T. W. Developmental aspects of physiological function relating to
aerobic exercise in children. Sports Med. 102255-266, 1990.

Rowland, T. W. Developmental Exercise Physiology. Champaign, IL: Human
Kinetics, 1996, 269.

Rowland, T. W. Effects of obesity on aerobic fitness in adolescent females. Am J
Dis Child. 145:764—768, 1991.

Rowland, T. W., J. A. Auchinachie, T. J. Keenan, and G. M. Green. Physiologic
responses to treadmill running in adult and prepubertal males. Int J Sports Med.
8:292-297, 1987.

Rowland, T. W., J. A. Auchinachie, T. J. Keenan, and G. M. Green. Submaximal
aerobic running economy and treadmill performance in prepubertal boys. Int J
Sports Med. 9:201-204, 1988.

Rowland, T. W. and G. M. Green. Physiological responses to treadmill exercise in
females: adult-child differences. Med Sci Sports Exerc. 202474-478, 1988.

Sallis, J. F ., J. J. Prochaska, and W. C. Taylor. A review of correlates of physical
activity of children and adolescents. Med Sci Sports Exerc. 322963-975, 2000.

Tanner, J. M. Growth at Adolescence. 2 ed. Oxford: Blackwell, 1962.

Treloar, A. E., R. E. Boynton, B. G. Behn, and B. W. Brown. Variation of the
human menstrual cycle through reproductive life. Int J Fertility. 12277-126, 1967.

Turley, K. R. and J. H. Wilmore. Cardiovascular responses to treadmill and cycle
ergometer exercise in children and adults. J App] Physio]. 83:948-957, 1997.

U. S. Department of Health and Human Services. Physical activity and health: a

report of the surgeon genera]. Atlanta, GA: National Center for Chronic Disease
Prevention and Health Promotion, 1996.

109

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

US. Department of Health and Human Services. Advance data from vital and
health statistics of the centers for disease control and national center for health
statistics. Atlanta: Centers for Disease Control and Prevention National Center for
Health Statistics, 2000.

Unnithan, V. B. and R. G. Eston. Stride frequency and submaximal treadmill
running economy in adults and children. Pediatr Exerc Sci. 2: 149-1 55, 1990.

Unnithan, V. B., J. A. Timmons, R. T. Brogan, J. Y. Paton, and T. W. Rowland.
Submaximal running economy in run-trained pre-pubertal boys. J Sports Med
Phys Fitness. 36:16-23, 1996.

Ward, D. S. and O. Bar-Or. Use of the Borg scale in exercise prescription for
overweight youth. Can. J. Sport Sci. 15:120-125, 1990.

Ward, D. S. and O. Bar-Or. Usefulness of RPE scale for exercise prescription
with obese youth. Med Sci Sports Exerc. 19, Suppl.:15, 1987.

Ward, D. S., J. D. J ackman, and F. J. Galiano. Exercise intensity reproduction:
Children versus adults. Pediatr. Exerc. Sci. 3:209-218, 1991.

Waters, R. L., H. J. Hislop, L. Thomas, and J. Campbell. Energy cost of walking
in normal children and teenagers. Develop Med Child Neurol. 25: 1 84-1 88, 1983.

Weston, A. T., R. Petosa, and R. R. Pate. Validation of an instrument for
measurement of physical activity in youth. Med Sci Sports Exerc. 29: 1 38-143,
1997.

Williams, J. G., R. G. Eston, and B. Furlong. CERT: A perceived exertion scale
for young children. Perceptual and Motor Skills. 79: 1451-1458, 1994.

Williams, J. G., R. G. Eston, and C. Stretch. Use of the rating of perceived

exertion to control exercise intensity in children. Pediatr. Exerc. Sci. 3:21-27,
1991.

110

   
   

M G

 

\lllllllllllfl

  

ll

3

   

A

iiiiiiiiii