M

l

    

\
\

NWllHlHNlWWHINWHHI(NW

\

 

 

é 1|

 

203

 

LIBRARY
Michigan State
University

Roﬂ'

This is to certify that the
thesis entitled

NUTRITION RELATED HEALTH BEHAVIORS:
PREVALENCE AND ASSOCIATION WITH OVERWEIGHT AMONG
ADOLESCENTS, GRADES 9-12, IN MICHIGAN, 2001 YRBS

presented by

RACHEL A. ROSENBAUM, R.D.

has been accepted towards fulﬁllment
of the requirements for the

M. S. degree in EPIDEMIOLOGY

 

 

majSFProfessor’s Signature
7/7/0?

Date

MSU is an Aﬁ‘innative Action/Equal Opportunity Institution

.-.-u-~-I-I-a-c-n-O-I-I-l-I-I-I-l-l-O-o-o-o-n-a---n.

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 pilClRC/DaIeDue.indd-p.1

NUTRITION RELATED HEALTH BEHAVIORS:
PREVALENCE AND ASSOCIATION WITH OVERWEIGHT AMONG
ADOLESCENTS, GRADES 9-12, IN MICHIGAN, 2001 YRBS
By

Rachel A. Rosenbaum, R.D.

A THESIS
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Epidemiology

2007

ABSTRACT
NUTRITION RELATED HEALTH BEHAVIORS:
PREVALENCE AND ASSOCIATION WITH OVERWEIGHT AMONG
ADOLESCENTS, GRADES 9-12, IN MICHIGAN, 2001 YRBS
By
Rachel A. Rosenbaum, R.D.
Pediatric overweight is currently one of the leading health issues among youth in the
United States. The National Health and Nutrition Examination Survey identiﬁed 34% of
adolescents aged 12-19 years to be either overweight or at-risk of overweight in 2003-
2004. Few studies have examined the association between overweight and nutrition-
related health behaviors (NRHB) simultaneously among adolescents. The objectives of
this study were to examine adherence to national health recommendations regarding
NRHB (ﬁ'uit and vegetable intake, physical activity, television viewing, and no risky
dieting behaviors), and the association between these behaviors and overweight among
Michigan youth. Data is ﬁom the Michigan Youth Risk Behavior Survey (YRBS) 2001, a
representative sample of publicly educated high school youth, grades 9-12 (n=2,891).
Results were weighted using stratum and individual weights. In the state of Michigan,
9.4% of adolescents participated in all four deﬁned NRHB, and only 7.7% participated in
all four NRHB and maintained a healthy body weight (BMI-for-age 5"'-85‘h percentile).
Girls who participated in 3-4 NRHB were signiﬁcantly more likely to maintain a healthy
body weight than girls with only 0-1 NRHB (84.8%, 95% CI: 81.9-87.8 compared to
72.8%, 95% CI: 67.2-78.5, respectively). A similar trend was noted among boys. In

conclusion, < 10% of adolescents adhered to national recommendations regarding NRHB

and youth that did not adhere to recommendations were more likely to be overweight.

ACKNOWLEDGEMENTS

I would like to take this opportunity to thank everyone who has both supported me
through this process and assisted me along the way. To my family, thank you for all of
your love, kind words, and encouragement always. To the team at the Center for
Statistical Training and Consulting, especially Connie Page, Weixing Song, and Irina
Murtazashvili, thank you for all of your technical assistance. Special thanks to Jianling
Wang, for all of your statistical knowledge, input, advice, and hard work. I would also
like to thank Earl Watt, Laurie Bechhofer, and Kim Kovalchik from the State of
Michigan for providing access to the YRBS data set, and special thanks to Steve Kinchen
from the Centers for Disease Control and Prevention for patiently addressing all of my
questions. Thank you to all of the supportive staff in the Department of Epidemiology,
also.
Special thanks to my committee members Dr. Dorothy Pathak and Dr. Mathew Reeves,
all of your time and input is appreciated. 9
Finally, I need to thank my primary thesis advisor, Dr. Ellen Velie, without whom this
work would not be complete. Ellen, I appreciate all of the time that you have invested in
helping me achieve this goal. I know that it has been a long road, and all of your time,

input, advice, editing, and support was instrumental in helping me ﬁnish this project.

iii

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LIST OF TABLES vi
CHAPTER 1: INTRODUCTION
1.1 Overview 1
1.2 Rationale 3
1.3 Study Aims 6
1.4 Organization of paper 7
CHAPTER 2: BACKGROUND
2.1 Deﬁnition of Overweight among Children and Adolescents_8
2.2 Prevalence of Overweight among Children and Adolescents_8
2.3 Health Consequences of Childhood and Adolescent Obesity_9
2.4 Tracking of Childhood and Adolescent Obesity 11
2.5 Risk Factors for Childhood and Adolescent Overweight 12
2.5.1 Individual Lifestyle Factors 13
2.5.1.1 Physical Activity 14
2.5.1.2 Television Viewing 15
2.5.1.3 Fruit and Vegetable Intake 17
2.5.1.4 No Risky Dieting Behaviors 18
CHAPTER 3: METHODS
3.1 Youth Risk Behavior System 20
3.1.1 Background of the YRBSS 20
3.1.2 Michigan 2001 YRBS 20
3.2 Deﬁnition of Outcome, Nutrition-Related Health Behaviors,
And Covariates 20
3.2.1 Outcome 20
3.2.1.1 Body Mass Index (BMI) 20
3.2.1.2 Validity and Reliability of BMI-for-age,
and self-reported height and weight 21
3.2.2 Nutrition-Related Health Behaviors 22
3.2.2.1 Physical Activity 23
3.2.2.2 Television Viewing 23
3.2.2.3 Fruit and Vegetable Intake 24
3.2.2.4 Dieting Behaviors 25
3.2.3 Covariates 26
3.2.3.1 Sports Participation 26
3.2.3.2 Race 26

 

3.2.3.3 Grade in School 27

 

iv

CHAPTER 4:

CHAPTER 5:

APPENDIX

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3.3 Sample Characteristics 27
3.3.1 Description of Sample 27
3.4 Statistical Analysis 29
3.4.1 Software Used in Analysis 29
3.4.2 Weighting Factors 29
3.4.3 Data Analysis by Aim 30
3.4.3.1 Aim 1 30
3.4.3.2 Aim 2 30
RESULTS
4.1 Paper for publication 32
4.1.1 Abstract 32
4.1.2 Introduction 33
4.1.3 Methods 35
4.1.4 Results 40
4.1.5 Discussion 44
DISCUSSION/CONCLUSIONS
5.1 Summary of Findings 58
5.2 Comparison of Findings to Prior Literature 58
5.3 Strengths and Limitations 62
5.4 Suggestions for Future Research 63

 

vii

LIST OF TABLES

Chapter 3:
Table l: Nutrition-Related Health Behaviors; Michigan YRBS, 2001 .................. 22
Table 2: Sample Selection and Exclusion Criteria; Michigan YRBS, 2001 ............ 28
Table 3: Population Demographics of Sample Analyzed, compared to adolescents with
missing variables, Michigan YRBS, 2001, weighted data ................................. 29
Chapter 4:

Table 1: Population Demographics of Sample Analyzed. Michigan YRBS, 2001,
weighted data ..................................................................................... 50

Table 2: Prevalence of Nutrition — Related Health Behaviors by Demographic
Characteristics among Adolescents grades 9-12. Michigan YRBS,
2001, weighted data .............................................................................. 51

Table 3: Prevalence of at least one, at least two, at least three, or all four Nutrition —
Related Health Behaviors by Demographic Characteristics among Adolescents
grades 9-12. Michigan YRBS, 2001, weighted data ....................................... 52

Table 4: BMI-for-age percentiles by Demographic Characteristics among Adolescents
grades 9-12. Michigan YRBS, 2001, weighted data ....................................... 53

Table 5a: Individual Nutrition-Related Health Behaviors (Fruit and Vegetable Intake &
Physical Activity) and BMI-for-age by Demographic Characteristics among Adolescents
grades 9-12. Michigan YRBS, 2001, weighted data ....................................... 54

Table 5b: Individual Nutrition-Related Health Behaviors (N o Risky Dieting Practices &
Television Viewing) and BMI—for-age by Demographic Characteristics among
Adolescents grades 9—12. Michigan YRBS, 2001, weighted data ........................ 55

Table 6: Nutrition-Related Health Behaviors and BMI-for-age by Demographic
Characteristics among Adolescents grades 9—12. Michigan YRBS,
2001, weighted data .............................................................................. 56

Table 6b: Addendum to Table 6: showing only those students with two Nutrition-Related

Health Behaviors and BMI-for-age by Demographic Characteristics among Adolescents
grades 9-12. Michigan YRBS, 2001, weighted data ....................................... 57

vi

APPENDIX

Table 1: Deﬁning Risky Dieting Behaviors: Fasting, diet pills, and/or vomiting to lose
weight; compared to students who are either not dieting, or are eating less and/or
exercising to lose weight: Michigan YRBS, 2001, weighted data ........................ 66

Table 2a: Students with zero or one Nutrition Related Health Behaviors and BMI-for—age
by Demographic Characteristics among Adolescents grades 9-12. Michigan YRBS,
2001, weighted data .............................................................................. 67

Table 2b: Students with two, three, or all four Nutrition- Related Health Behaviors and
BMI-for-age by Demographic Characteristics among Adolescents grades 9-12. Michigan
YRBS, 2001, weighted data ..................................................................... 68

Table 3: Nutrition-Related Health Behaviors and BMI-for-age by Demographic
Characteristics among Adolescent Males grades 9-12. Michigan YRBS, 2001,
weighted data ...................................................................................... 69

Table 4: Nutrition-Related Health Behaviors and BMI-for—age by Demographic

Characteristics among Adolescent Females grades 9-12. Michigan YRBS, 2001,
weighted data ...................................................................................... 7O

vii

CHAPTER 1: INTRODUCTION

1.1 Overview:

Pediatric overweight is currently one of the leading health issues among youth in the
United Statesl. In 2003-2004, 18% of children aged 6-17 were considered overweightz,
and among adolescents aged 12-19 years, 34.3% were either overweight or at-risk of
overweight3. There has been a continued upward trend, where in 1988-1994 11% of
children aged 6-17 years were considered overweight“, and in 1999-2000 15.5 % of
children aged 12-19 were overweight5 . Health behaviors such as dietary intake, physical
activity, and time spent in leisure activities have been associated with overweight in
children and adolescents6.

Childhood overweight has many possible consequences, some with a short-term
impact, others Which stretch across the lifespan“. Childhood overweight can have both
physiological and psychological effects, from diseases like sleep apnea and cholelithiasis,
to emotional outcomes like a negative self-esteem9’lo. Overweight or obesity as a child or
adolescent is a predictor of adult overweight and obesity, and childhood overweight can
also affect adult morbidity and mortality1 1. Long-term childhood overweight results in an
increased risk for high blood pressure (HTN), hyperlipidemia, and Type 2 Diabetes
Mellitus (”IQDM)”.

Risk factors for overweight and obesity include both genetic and environmental
inﬂuencesé, including socio-economic status, education level, and the built / food
environments in which people live'3’l6. Individual-level risk factors include dietary

intake, physical activity, and leisure time spent in sedentary activitiesl.

National recommendations therefore address these modiﬁable risk factorsn'lg. The
2005 Dietary Guidelines for Americans, jointly issued by the United States Department
of Agriculture and the United States Department of Health and Human Services,
recommends maintaining a healthy weight, daily physical activity, and increased fruit and
vegetable intake"). For children, they also recommend limiting the amount of time spent
being sedentary, and increasing physical activity”. Physical activity should be
undertaken by children and adolescents on a daily or near-daily basis19‘21. Based on the
guidelines set by the International Consensus Conference on Physical Activity Guidelines
for adolescents, it is recommended by the CDC that adolescents should participate in
moderate or vigorous physical activity for at least twenty minutes at least three times per

R2031. The American Academy of Pediatrics recommends television viewing for

wee
children to be no more than one or two hours per day”. Fruit and vegetable intake is
recommended to be at least ﬁve servings per day, and is marketed nationally as the “5-
A-Day for Better Health” program, administered by the US. Department of Health and
Human Services and the CDC”.

Data will be analyzed from the 2001 Michigan Youth Risk Behavior Survey
(YRBS), a survey that examines the behaviors of publicly educated adolescents in the
state of Michigan, grades 9-12. This study seeks to determine the prevalence of nutrition-
related health behaviors in youth, and to describe the demographics of students who are
compliant with these recommendations in relationship to normal weight and overweight /
at-risk of overweight. This study also seeks to examine the association between these

nutrition-related health behaviors with overweight / at-risk of overweight among

adolescents. Nutrition-related health behaviors to be examined are ﬁ'uit and vegetable

intake, level of physical activity, television viewing, and having no risky dieting
behaviors, which include no laxative or diet pill use, no starvation, and no vomiting to
lose weight.

The Youth Risk Behavior Surveillance System (YRBSS) was developed by the
Centers for Disease Control and Prevention (CDC) to monitor behaviors that relate to
premature morbidity and mortality among adolescents“. The YRBSS has ﬁve
components, including state and national school-based surveys, called the Youth Risk
Behavior Survey (YRBS), of publicly educated adolescents, grades 9 4224. The other
three components include a household survey of youth 12-21 years, a Special population
survey, and a survey mailed to college students”. Behaviors that the YRBS are focused
on assessing include behaviors related to violence, tobacco and drug use, risky sexual

behaviors, unhealthy dietary behaviors and inadequate physical activity“.

1.2 Rationale:

The goals of this study are to determine the adherence of 9-12th grade publicly
educated Michigan youth, in 2001, to national recommendations regarding modiﬁable
lifestyle factors. Individual behaviors to be evaluated are fruit and vegetable intake,
physical activity, television watching, and no risky dieting practices. I will examine the
prevalence of each individual behavior, and the prevalence of students participating in at
least one, at least two, at least three and all four nutrition-related health behaviors. The
association of these behaviors to overweight / at-risk of overweight will also be assessed.

Several studies have examined the prevalence of individual nutrition-related health

behaviors among adolescents. Using data from the 1999 national YRBS, Lowry et a1

reported that 76.1% of students ate less than ﬁve servings of ﬁ'uits and vegetables per
dayzs. They also found that 30.5% of students did not meet the recommendations for
physical activity, deﬁned as 30 minutes of moderate physical activity at least 5 times per
week or 20 minutes of vigorous physical activity at least 3 times per week, and that
42.8% watched television for more than two hours per day”. In a study examining
dieting behaviors of adolescents ﬁom the 1993 YRBS, Story et al reported that 5.5% of
girls and 1.5% of boys had either used vomiting or diet pills to control their weight“.
Research involving the prevalence of television viewing, fruit and vegetable intake,
physical activity, risky dieting behaviors, and their individual association to BMI-for-age
among children and adolescents are abundant. No studies to my knowledge, however,
combine all four nutrition-related health behaviors in an attempt to describe compliance
to multiple national recommendations regarding health in adolescents. Television
viewing, diet, and physical activity were examined together in studies examining their
association with BMI-for-age27'29. In the F rarningham Children’s study, researchers
found that increased television viewing, deﬁned as more than 3 hours per day, when
combined with either a high fat diet (234% total calories from fat) or a low level of
physical activity (determined by using an electronic motion sensor), was associated with
an increased risk of having excess body fat in early adolescence”. Anderson et al found
that children who watched more television (_>_4 hours per day) and had lower levels of
participation in vigorous physical activity (53 times per week vigorous) had higher
BMI’S than children who exercised more and watched less television”. Using data from
the YRBS, Eisenmann found that increased physical activity was associated with less

129

television viewing and a lower BM . In research examining the association of physical

activity with other health behaviors, low activity levels were found to be associated with
lower fruit and vegetable intake and greater television viewing, among others”. Risky
dieting behaviors have been found to be associated both with an increased risk of
developing a serious eating disorder3 1'33, and with a lower diet quality and decreased
intake of fruits and vegetables34

Several studies have examined the prevalence of combinations of modiﬁable lifestyle
factors among adults35'37. In a study published in 2001, Ford et a1 examined four health
behaviors in adults: huh and vegetable intake (25 per day), adequate physical activity,
no smoking, and body mass index (<25)35. Study participants were from NHANES 111
(1988-1994). They found that 6.8% of adults in the US. had all four healthy behaviors.
Women had a higher prevalence than men (8.0% compared to 5.5%, and whites had a
higher prevalence than African-Americans and Mexican-Americans (7.2% compared to
3.4% and 4.0%, respectively)”. Kim et al also assessed the prevalence of health
behaviors of adults in 2004, using diet quality (Diet Quality Index — International, used to
measure and quantify diet quality), physical activity (Physical Activity Index), no
tobacco use, and alcohol use (Alcohol Consumption Index) as lifestyle indicators”. The
authors compared behavior patterns of people living in China to those of people living in
the US. Overall, they reported that people living in the US. had both a lower diet
quality and lower physical activity levels than those of the comparison group in China”.
Reeves and Rafferty also completed a study on lifestyle factors of US. adults, using the
2000 BRFSS. The health behaviors used in that study were also used by Ford et a1, and
included fruit and vegetable intake (25 per day), regular physical activity (230 minutes

25 times per week), no tobacco use, and maintaining a healthy body weight (BMI :25)”.

They found that the prevalence of US. adults with all four deﬁned healthy behaviors was
only 3%”.

This study seeks to contribute to the current knowledge regarding the associations of
multiple health-related behaviors among adolescents and their relationship with
overweight / at-risk overweight among that population. Results can hopefully assist
policy-makers and educators in their continuing efforts to address the epidemic of

overweight / at-risk overweight among Michigan youth.

1.3 Study Aims:

The speciﬁc aims of this study are:

1. To determine both the prevalence of individual and combined nutrition-related
behaviors, and the adherence to national recommendations regarding modiﬁable
. lifestyle factors, among a population-based sample of publicly? educated Michigan
youth, grades 9-12, as reported in the 2001 Michigan YRBS.
Nutrition-related behaviors include: ﬁlth and vegetable intake, physical activity,
television viewing, and not participating in risky dieting behaviors. The four
variables will be combined to determine the prevalence of students participating

in all four deﬁned nutrition-related health behaviors.

2. To examine the association between nutrition-related health behaviors and
overweight / at-risk for overweight, based on BMI-for-age calculated ﬁ'om self-

reported height and weight, among Michigan youth.

1.4 Organization of paper

This thesis includes a separate paper for publication in Chapter 4, thus there will be
some redundancy between chapters. Chapters 1, 2, 3 and 5 represent typical thesis
chapters of an introduction, background, methods and discussion. Chapter 4 is a stand-

alone paper which includes all the results.

CHAPTER 2: BACKGROUND

2.1 Deﬁnition of Overweight among Children and Adolescents

Childhood overweight is usually deﬁned using the CDC’S BMI-for-age growth
charts”. BMI is calculated using a child’s height and weight, and used as an indicator of
body fatness”. The growth charts are both gender- and age-speciﬁc for children aged 2-
19 years”. Using the BMI-for-age charts, overweight is deﬁned as 295“1 percentile, and
at-risk of overweight is deﬁned as Z85th — 95th percentile”. Normal weight is deﬁned as
25th -85til percentile, and underweight is deﬁned as <5‘h percentile”. There is no
deﬁnition of childhood obesity when using the CDC’s growth charts. However, current
literature uses the word “overwei t” for children in the 285th — 95th percentile, and
“obese” for children Z95th percentileé’m’l 1.
2.2 Prevalence of Overweight among Children and Adolescents

There has been a signiﬁcant increase in the prevalence of childhood and adolescent
overweight in the United States39’40. Among children aged 6-17 years of age, the
prevalence of overweight has increased from 6% in 1976-1980 to 18% in 2003-20042.
Using data from NHANES from 1971-2000, Jolliffe found that overweight among
children aged 2-19 years had increased from 10% in 1988-1994 to 14.4% in 1999-2000”.
The study also reported that children who are overweight are getting heavier. The
number of overweight youth increased by 182% between 1971 to 1999-2000, and they
are getting heavier, indicated by the extent of overweight, which increased by 247%
during the same time frame”. Ogden et al found similar increases in overweight

prevalence among children and adolescents using NHANES 1999-20005, and current

rates reported from NHANES 2003 -2004 data Show that 17.1% of children and
adolescents aged 2-19 years are overweightB. It has also been estimated, using current
overweight prevalence rates, that there is now a 30-40% lifetime risk of children in the
United States to develop Type II Diabetes Mellitus39. 1

Strong differences exist with respect to overweight among adolescents of different
racial and ethnic groups“. African-American and Hispanic children and adolescents
have higher prevalence rates of overweight and obesity when compared to white children
and adolescents“. By gender, in 1999-2000, Black non-Hispanic girls, aged 6-18 years,
had the highest prevalence of overweight at 24%, and Mexican-American boys, aged 6-
18 years, had the highest prevalence at 29%“. The rates for white, non-Hispanic girls
and boys aged 6-18 were 11% and 12%, respectively“. In 2003-2004, among
adolescents aged 12-19 years, non-Hispanic white girls had a prevalence rate of 15.4%,
non-Hispanic black girls were at 25.4%, and Mexican-American girls were at 14.1%“).
Among adolescent boys of the same age range, non-Hispanic white bOys had a
prevalence rate of 19.1%, non-Hispanic black boys were at 18.5%, and Mexican-
American boys were at 18.3%“).
2.3 Health Consequences of Childhood and Adolescent Overweight

There are a number of health risks associated with childhood and adolescent
overweight and overweight“. Physical health consequences include glucose intolerance,
insulin resistance and Type 2 Diabetes Mellitus, hypertension, hyperlipidemia [deﬁned
by increased low-density lipoproteins (LDL) and triglycerides with decreased high-
density lipoproteins (HDL)], cholelithiasis and hepatic steatosis, pseudotumor cerebri,

sleep apnea, polycystic ovary disease (PCOD), and orthopedic complicaﬁonsl’6’42.

Obesity, elevated plasma insulin, systolic blood pressure, and total / LDL cholesterol tend
to cluster in studies of children, and track into adulthood”. In a study investigating the
prevalence of metabolic syndrome in children, researchers found a 4.2% prevalence of
the indicators (deﬁned as having at least 3 of the 5: increaSed triglycerides, blood
pressure, fasting glucose, waist circrunference and decreased HDL) among adolescents
12-19 years of age“. The study also found a 29% prevalence of the indicators among
overweight adolescents, and 7% prevalence among adolescents at-risk of overweight“.
Criteria were based on modifying the National Cholesterol Education Program’s (NCEP)
deﬁnition of the metabolic syndrome for adults for use with an adolescent population.
Several studies have looked at the relationship between obesity and health-related
quality of life (QOL)45’46. The QOL includes physical, emotional, social, and school
functioning. In a study of 106 children and adolescents, aged 5-18 years, researchers
found that severely obese children (average BM1=34.7 kg/m2) reported signiﬁcant
impairment in the physical, psychosocial, emotional, social, school functioning and total
score when compared to healthy children and adolescents45 . Using the same QOL
assessment tool, Williams et al found that overweight and obese children, ages 9-12
years, differed most strongly in physical and social functioning from their normal weight
peers, with obese children having the lowest scores“. They also reported a trend of
decreasing QOL scores as weight increased“. Among females who were overweight or
obese during adolescence, rates of poverty are higher than those of normal-weight
females; years of advanced education, family income, and maniage rates are lower
among females who were obese in late adolescence than those of a normal weight‘s'43 .

Studies have also found that when presented with drawings of obese ﬁgures, children

10

associate the heavier ﬁgures with lower academic and social success”. Children
described the heavier ﬁgures in the drawings as “lazy”, “stupid”, “sloppy”, and “ugly“.
Emotionally, there are associated risks of developing low self-esteem, a negative body
image, and depression9’w. Children are also at risk ﬁom the stigmatization that occurs
ﬁom being obese, which can also carry into adulthood9’l0.
2.4 Tracking of Childhood and Adolescent Overweight into Adulthood
Overweight in childhood and adolescence is a predictor of adult obesityl 1’48’49. The
risk of being an overweight adult doubles for children who were overweight when
compared to normal weight kids”. Various stages of growth have been associated with
an increased risk of overweight and obesity in adulthood. During the toddler years,
obesity is related to body fat and parental weight”. Bray reports a four-fold risk of
obesity in adulthood for children ages 1-3 years who are greater than the eighty-ﬁfth
percentile weight-for-height and have at least one parent who is also overweight”. The
period, of adiposity rebound, deﬁned as weight gain between the ages of 5 to 7, has been
suggested to play a role in the development of obesity”. An early adiposity rebound is
associated with both parental obesity and the persistence of obesity into adulthood6’49’50.
Obesity as an adolescent is associated with adverse health outcomes later in life”. The
likelihood of being overweight in adulthood is greatly increased for overweight or obese
adolescents”. Teens with a BMI-for-age at or greater than the ninety-ﬁfth percentile
have a 5-20 fold chance of being overweight as an adult”. In the Fels Longitudinal
Study, Guo et al found that of the 347 participants, those overweight or obese at aged 35
years had signiﬁcantly higher BMI in childhood or adolescence than those who were not

overweight or obese5 1. Freedman et al, using data from the Bogalusa heart study, also

11

reported that childhood BMI was associated with adult adipositysz. They also speculated
that the magnitude of this relationship was dependent on the level of overweight during
childhood”.

The burden of obesity and its related co-morbidities do not end in childhood or
adolescence. Risk factors for adult diseases like increased blood pressure, dyslipidemia,
and associated factors of insulin resistance continue ﬁorn childhood to adulthood, and
can increase with increased weight gain over the lifespan11’48. Research using data hour
the Harvard Growth Study of 1922-1935 found that overweight in adolescence increased
the risk of several diseases, in both male and female participants”. The follow-up time
was 55 years, and the 508 participants were divided into two study groups: overweight,
deﬁned as >75‘h percentile BMI, and lean, deﬁned as 25-50'ch percentile BM153. The
study formd that in both genders, among the group deﬁned as overweight during
adolescence, the risk of atherosclerosis and coronary heart disease was increased by 7.3
and 1.8 fold, respectively”. Arthritis was higher among women in the overweight group,
and colorectal cancer and gout were higher among men53 .
2.5 Risk Factors for Childhood and Adolescent Overweight

Risk factors for childhood and adolescent overweight include a complex balance of
genetics and environmental factors6’15’49’54’5 5 . The underlying theory of weight regulation
is one of energy balance: energy intake must equal energy expenditure". One of the
basic concepts of weight regulation is that of “set point” maintenance“. This set point
refers to the body’s ability to maintain a weight given the variations of energy intake and
expenditure that ﬂuctuate on a day-to—day basis. Total energy expenditure is calculated

using three factors: resting metabolic rate, thennogenesis, and physical activity". Total

12

energy expenditure tends to be lower in obese individuals, though the role of resting
metabolic rate in the development of obesity is unresolved6’47. The other half of the
energy balance equation involves energy intake: if the body takes in more energy in the
form of calories that it can expend, energy will be stored as fat.

The family environment plays a large role in the development of childhood obesity6.
Studies documenting the activity levels, eating patterns, and relative adiposity of parents
have all shown a relationship with childhood obesity”. Eating habits and food
preferences are developed in children at a very young age“. Patterns of eating higher fat
or sweetened foods seen in children comes from the exposure to these foods in their
home environment5 9. These eating patterns also track from childhood to adulthood“.
Overweight or obese parents are also more likely to have overweight children57’59. This
effect is more likely to be seen in children under the age of ten. After this age, adiposity
of a parent is less likely to be a factor in predicting adult obesity”. Physical activity
patterns of adults affect their offspring. Studies have shown that activity levels of parents
can inﬂuence both childhood activity patterns and the development of obesity in
adolescenceé.

Socio-economic status has also been associated with obesity among both adolescents
and adults'3"5. Miech et al reported that among adolescents aged 15-17, poverty status
was associated with overweight; among poor students the prevalence of overweight was
50% higher than non-poor students”. Built and social environments may play a role in
this relationship15 ’16. Studies have shown that areas with a higher poverty rate have less
access to healthy food than do areas that are more prosperous‘s’m.

2.5.1 Individual Lifestyle Factors

13

Individual lifestyle factors that are associated with overweight and obesity include
physical activity and leisure time spent in sedentary activities, television viewing, dietary
intake including fruit and vegetable intake, and dieting behaviorsss'wﬁ.
2.5.1.1 Physical Activity 1

Physical activity plays a large role in the relationship of energy balance. Physical
activity levels have been shown to track across the lifespan, and are associated at an early
age with levels of parental physical activity6’64’65. Positive beneﬁts of physical activity
include increased aerobic and cardiovascular ﬁtness, increased bone mass, increased
levels of HDL cholesterol, and decreased risk of diseases including hypertension and
Type 2 Diabetes Mellitus65 . Children who have lower levels of physical activity have the
potential for increased blood pressure, insulin levels and abnormal lipid proﬁles, and are
more likely to be overweight“. Childhood activity levels have also been shown to track
into adultllood64’65.

Activity levels vary for children of different gender, grade, and race28'29’65 . The
Bogalusa Heart Study, found that boys, aged 9-15 years, participated in more vigorous
physical activity, while girls of the same age participated in more moderate and light
activities65 . Lower levels of physical activity were reported among students without a
school-based physical education class“. Anderson et al found that 65% of girls aged 14-
16 years participated in vigorous physical activity three or more times per week,
compared to 80% of boys in the same age bracket”. By grade, activity levels decreased
as grade increased from ﬁfth to eighth grade65 . Differences in trends are also seen by
race and gender, where 87.9% of non-Hispanic white boys, 77.6% of non-Hispanic black

boys, and 80.2% of Hispanic boys reported participating in vigorous exercise _>__3 times

14

per week. Among girls, 77.1% of non-Hispanic white girls, 69.4% of non-Hispanic black
girls, and 72.6% of Hispanic girls reported participating in vigorous exercise 23 times per
week65 . These race and gender trends were also noted by Eisenmann et al, who found
that, among adolescents aged 14-18 years, participation in. vigorous physical activity 23
times per week was higher among boys than girls (72.3% compared to 57.1%,
respectively); and higher among whites than African American and Hispanic subjects
(67.4% compared to 55.6% and 51.0%, respectively)”.

Physical activity is associated with BMI among children28’29’6o. In a study comparing
children, aged 11-15 years, of normal weight (BMI-for-age < eighty-ﬁfth percentile), to
overweight children (BMI-for-age 2 eighty-ﬁfth percentile), vigorous physical activity
was the only risk factor related to an increased BMI“). The study also foimd that
overweight students had decreased ﬁber intake and decreased time spent in vigorous
physical activity when compared to normal weight children“).
2.5.1.2 Television Viewing

Television viewing among children and adolescents is a common leisure-time
activity22’27’28’62. The American Academy of Pediatrics recommends limiting total
television viewing (and other entertainment media) time as 52 hours per dayzz.
Prevalence rates of television viewing differ by gender and race. Using data ﬁ'om
NHANES 1988-1994, it was reported that 24.3% of non-Hispanic white boys, aged 8-16
years, watched _>_4 hours of television per day, while rates among non-Hispanic black
boys and Mexican-American boys were much higher (42.8% and 33.3%, respectively)”.
Among girls of the same age, 15.6% of non-Hispanic white girls reported watching _>_4

hours of television per day, compared to 43.1% of non-Hispanic black girls and 28.3% of

15

Mexican-American girls28 Results from the 1999 YRBS showed similar trends: among
boys aged 14-18 years, 18.1% of white boys reported watching television 24 hours per
day, compared to 54.7% of African-American boys and 30.6% of Hispanic boys”. The
prevalence rates by race among girls aged 14-18 years were 12.9% for white girls, 53.7%
for Aﬁican—American girls, and 28.3% for Hispanic girls”.

In a policy statement issued in 2003, the American Academy of Pediatrics reported
that children who watched television less than two hours per day had signiﬁcantly lower
BMI’S than those children who watched television for four or more hours per day“.
Television viewing has been associated with an increased BMI among children, and has a
positive relationship with intake of less nutrient-dense foods, including an inverse
association with fruit and vegetable intake6'27'6l'63. Children who watched _>_3 hours of
television per day, when combined with either a lower level of physical activity or a high
fat diet, had the greatest risk of increased adiposity as adolescents”. In the Framingham
Longitudinal Study, Proctor et al reported that adolescents who watched television for
1.75 hours per day had a lower BMI, triceps, and sum of skin-folds when compared to
adolescents who watched for more than 3.75 hours per day”. Analyzing data from
NHANES 1988-1994, Andersen et al reported that 26% of the 4,063 children
participating in the study watched more than four hours of television per day, and
children with two hours per day or less television time had less body fat and lower BMI’S
than their peers who watched more television”. Among children aged 14-18, ﬁ‘om the
1999 YRBS, researchers found that 25% of participants watched television for four or
more horns per day, and that obesity prevalence among children aged 12-18 years

increased 2% with every additional hour of television viewed per day”. A positive linear

l6

relationship between BMI and hours of television viewed was also reported in the
California Teen Longitudinal Study62. Results ﬁ'om the F rarningham Children’s Study
found that children with the most television viewing had higher BMl’s at follow-up, and
Showed the greatest increases in mean BMI ﬁ'om ages 4-11 years”.
2.5.1.3 Fruit and Vegetable Intake

Fruit and vegetable intake is encouraged for all age groups. The protective beneﬁts
of increased intake of ﬁuits and vegetables for adults include a decreased risk of
cardiovascular disease, Type 2 Diabetes Mellitus, and some cancers“. The National 5-A-
Day for Better Health Initiative was created to help promote the recommendation of
consumption of 25 servings per day of fruits and vegetables“. Studies have shown that a
large percentage of children and adults do not meet this recommendation23’6l. In the 5-A-
Day Baseline Survey, researchers found that the median fruit and vegetable intake for
adults was only 3.4 servings per day”. Krebs-Smith et al, using the Continuing Surveys
of Food Intake by Individuals (CSFII) ﬁom 1989-1991, found that only one in ﬁve
children aged 2-18 years had consumed ﬁve or more servings of fruits and vegetables per
day, and 25% of vegetables consumed were French ﬂies“. Fruit and vegetable intake is
increasing among children in the United States“. Using data ﬁom the (CSFII) ﬁom
1989-1991 and 1994-1996, researchers found an increase in fruit and vegetable intake in
both boys and girls, where boys aged 12-19 years increased from 3.4 to 3.7 servings per
day of vegetables, compared to 2.7 to 2.8 servings per day for girls of the same age“.
Fruit intake for all children aged 6-19 years increased from 1.1 to 1.5 servings per day
over the same time period61 . Even accounting for the increases, children are not meeting

the recommended intake of 25 servings of ﬁ'uits and vegetables per day.

17

Fruit and vegetable intake has been Shown to be associated with other health
behaviors, like television viewing and risky dieting behaviors25’26’6'. Among public
school children in the 6th and 7th grade in a study in Massachusetts, researchers found an
inverse relationship between fruit and vegetable consumption and hours of television
viewed per day“. Using data ﬁom the 1999 YRBS, Lowry et al reported that among
adolescents, grades 9-12, television viewing was associated with insuﬂicient fruit and
vegetable intake in both male and female subj ectszs . In a study examining dieting
behaviors, researchers reported that extreme dieting was associated with decreased intake
of fruits and vegetables among girls, grades 9-1226. Neumark-Sztainer et al reported a
similar trend among middle- and high-school girls in the Project EAT study, conducted in
Minneapolis, Minnesota“.
2.5.1.4 No Risky Dieting Behaviors

Dieting is a concern among adolescents for a variety of reasons. Dieting has been
associated with a decreased intake of nutrient-dense foods, which may affect growth and
development“. High-risk dieting behaviors, like meal skipping, laxative use, and
vomiting may be associated with the development of eating disorders, like anorexia
nervosa, bulimia, or binge eating”. Long term health consequences of dieting include
increased weight gain and possible nutrient deﬁciencies, and may contribute to the
development of bulimia or other eating disorders32’68. Using a cohort of children from the
Nurses Health Study, Field et al found that 4.5% of girls and 2.2% of boys aged 9-14
years were frequent dieters“. Binge eating was more prevalent among girls, and was
associated with dieting for weight control among both sexes“. Dieters were also reported

to have gained more weight than non-dieters during the survey“. In a cohort study of

18

adolescents aged 14-15 years, completed in Victoria, Australia, it was reported that
females who were dieting at a severe level (assessed using the Adolescent Dieting Scale)
were 18 times more likely to develop an eating disorder than non-dieters3 1. Authors
concluded that moderate dieting and weight control behaviOrs like exercise carry a lower
risk of developing an eating disorder than behaviors that severely restrict dietary intake3 1.
Unhealthy dieting practices are prevalent among US. youth. Studies of adolescents
have shown that 57.2% of females and 31.6% of boys have used at least one unhealthy
dieting behavior (deﬁned as skipping meals, fasting, diet pill use, laxative use, tobacco
use, or vomiting) in attempts to lose weight“. Using data ﬁom the1999 YRBSS, Lowry
et al found that one in four students (32.5% for females and 17.1% of males, combined
27% overall) were using unhealthy dieting practices while trying to either maintain or
lose weight“. Girls in the 6-8th grades have been found to use methods like vomiting,
laxatives, or diet pills to lose weight”. In the Project EAT survey, a cross-sectional
survey of 5,144 middle school students in the St. Paul/Minneapolis area, 52.7% of girls
and 31.6% of boys reported using at least one unhealthy dieting behavior to lose weight”.
In a study of 10th graders in Los Angeles, of students who reported they were attempting
to diet, 44% were skipping meals3‘3 . Lowry et al reported that of the students attempting
to lose weight in the 1999 national YRBS survey, 27% of students were using unhealthy
weight control behaviors, while 54% of students trying to maintain or lose weight were

using exercise and a modiﬁed fat and calorie diet”.

19

CHAPTER 3: METHODS

3.1 Youth Risk Behavior System
3.1.1 Background of the YRBSS

The Youth Risk Behavior Surveillance System (YRBSS) was developed by the
Centers for Disease Control and Prevention (CDC) to monitor health-risk behaviors
among adolescents in the United Sates. The survey design is a complex cluster design.
Students in grades 9-12 are surveyed at the local level and are aggregated to the state and
national levels. The survey is representative of students both publicly and privately
educated in the United States. The samples are weighted to adjust for both non-response
and demographic characteristics. The survey is collected every other year, in odd-
numbered years").
3.1.2 Michigan 2001 YRBS

The state-level 2001 Michigan Youth Risk Behavior Survey (YRBS) is part of the
national effort to assess health risks among adolescents. In 2001, 43 public schools
(88%) participated in the survey, and 3,630 students (83%) completed the survey. The
overall response rate for the state of Michigan in 2001 was 73%, calculated by

multiplying the school participation rate (88%) by the student participation rate (83%)“.
3.2 Deﬁnition of Outcome, Nutrition-Related Health Behaviors, and Covariates

3.2.1 Outcome

3.2.1.1 Body Mass Index (BMI)

20

To assess the outcome of overweight or obesity among our sample, self-reported
height and weight were used to obtain each individual’s body mass index [BMI: weight
in kg / (height in m)2 ]. The CDC’S Growth Charts were used to assess BMI-for-age
based on percentile height and weight cut-offs”. The 2000 CDC Growth Charts are
gender-speciﬁc and age-speciﬁc. They were developed ﬁom national data sets
(NHANES I-III, but weight was not used ﬁom NHANES III for children over the age of
673), and are recommended to be used as a screening tool to identify overweight children

and adolescents”. The questions on the survey were:

“Q. 5 How tall are you without your shoes on?
Q. 6 How much do you weigh without yoru' shoes on?”

Self-reported height and weight were converted to kilograms and centimeters for
analysis. In order to use the CDC’s SAS program for Growth Chart percentiles, age was
speciﬁed in months.

The CDC deﬁnes overweight based on BMI-for-age as: BMI 295th percentile
overweight, and at-risk for overweight as BMI 285th - 95'h percentile. We also used the
deﬁnition of underweight, BMI-for-age 35‘“ percentile, and examined this group
separately, as this may be indicative of disordered eating or other health problems75 .
3.2.1.2 Validity and Reliability of BMI-for-age, and self-reported height and weight

BMI-for-age has been found to be an acceptable index to use for detecting
overweight among children in several studies, and has been found to be more accurate
than both the Rohrer Index (weight/height3) and the Benn Index (weight/height”, where p
is minimizing the association with height)for assessing body fatness, using averaged skin-
folds as a standard(correlations of r =0.47, r =0.39, and r =0.38, respectively)76’77. BMI-

for-age has also been associated with total and percent body fat, and is an acceptable

21

measure of fatness in children and adolescents/8. Self-reported height and weights have
been shown to be reliable among adolescents when compared to actual height and weight
(correlation range 0.87-0.94)”, with the caveat that girls are more likely to underreport
weight than boys, and overweight subjects are more likely to underreport weight than
those of a normal weight72’79’80.
3.2.2 Nutrition-Related Health Behaviors

Nutrition-related health behaviors were deﬁned based on current national guidelines
or were developed by this author based on other scientiﬁc work using similar nutrition-
related health behaviors or data sets. Categorical cut-off points for variables are deﬁned
as ‘healthy’ based on public health recommendations from Healthy People 2010, the
American College of Sports Medicine (ACSM), and a priori designation based on prior

work examining health behaviors among adolescents.

Table 1: Nutrition-Related Health Behaviors; Michigan YRBS, 2001

 

Nutrition-Related Health Behaviors: Deﬁnitions:

 

 

 

 

 

Fruit and Vegetable intake 25 servings per day a

Dieting behaviors No laxative/diet pills/emesis/fastingf
Television viewing 52 hours per day °

Physical Activity 23 times per week “vigorous” d

 

_>_5 times per week “moderate”
25 times per week
“vigorous” + “moderate”

 

8 Dietary intake of fruits and vegetables will be based on ﬁve questions regarding the intake of 100% fruit
juice, fruits, carrots, green salad, and other vegetables.‘1 The question regarding potato intake was not used
in the variable because it did not address ﬂied potato products separately. A dichotomous variable will be
created based on the national Five-a-Day recommendations, where any combination of answers was
accepted. Five servings per day or more was considered adequate.

Dieting behaviors will be considered as acceptable for exercise or eating less, but as not acceptable for

fasting, diet pill use, or vomiting/laxative use.

81,82

c Television viewing will be considered acceptable for less than or equal to 2 hours per day, based on

national recommendations.

d Physical activity cut-offs will be determined based on a combination of questions. The ACSM guidelines
were set at 20 minutes of vigorous activity three times per week, but new guidelines promote activity
daily.”84 A thorough literature review has shown that several studies use different cut-off points for the
determination of “active”.3°"”"‘”‘87 Our a priori designations are based on those established be Berrigan et
al: (3 x/wk vigorous or a combo of 5/wk moderate) = adherent.88

22

 

3.2.2.1 Physical Activity

Physical activity was categorized based on questions assessing both vigorous and
moderate activity levels. The American College of Sports Medicine (ACSM) guidelines
suggest adequate physical activity as 20 minutes of vigorous activity three times per
week. However, new national guidelines like Healthy People 2010 promote activity
daily83’84’89’90. A thorough literature review has shown that several studies used different
cut-off points for the determination of ‘active’30’82’85'87. Our a priori designations were
based on previous research using similar cut-off points88 and guidelines set by the
International Consensus conference on Physical Activity“. Other studies using data from
the YRBS also used the same categorization values for analysis25’69’87'88.

The questions pertaining to vigorous and moderate activity are:

“On how many of the past 7 days did you:
Q. 80 exercise or participate in physical activity for at least 20 minutes
that made you sweat and breathe hard, such as basketball,
soccer, running, swimming laps, fast bicycling, fast dancing, or .
, similar aerobic activities?
Q. 81 participate in physical activity for at least 30 minutes that did

not make you sweat or breathe hard, such as fast walking, slow
bicycling, skating, pushing a lawn mower, or mopping ﬂoors?”

Answers were coded as: “0-7 days/week”. I then created a dichotomous variable that
was deﬁned as meeting criteria for physical activity for vigorous 23 times per week (Q.
80), moderate 25 times per week (Q. 81), or a combination of vigorous and moderate for
25 times per week.
3.2.2.2 Television Viewing

Television viewing was included in this study due to the well-documented relationship

between sedentary behaviors including television viewing, exercise, and BM122’25’30’86.

23

The American Academy of Pediatrics recommends “limiting children’s total media time
(with entertainment media) to no more than 1-2 hours of quality programming per day”22.

Television viewing was assessed using the following question:

“Q. 83 On an average school day, how many hours do you watch TV?

. No TV on average school day
. Less than 1 hour per day

. 1 hour per day

. 2 hours per day

. 3 hours per day

. 4 hours per day

. 5 how's per day”

ﬁ¢MhWN~

I then created a dichotomous variable, and television viewing was considered acceptable
for less than or equal to two hours per day based on the guidelines set by the American
Academy of Pediatricszz.
3.2.2.3 Fruit and Vegetable Intake

Dietary intake of fruits and vegetables was based on ﬁve questions regarding the
intake of 100% ﬁ'uit juice, ﬁ'uit, carrots, green salad, and other vegetables“. In the
Michigan data set, the question regarding potato intake (Q. 76) was nOt used because it
did not address ﬂied potato products separately.

The questions, as they appeared on the MI YRBS, were:

“During the past 7 days, how many times did you eat/drink:

Q. 73 100% fruit juices such as orange juice, apple
juice, or grape juice?

Q. 74 fruit?

Q. 75 green salad?

Q. 77 carrots?

Q. 78 other vegetables?”

Responses were coded as:

6‘

Did not eat/drink
. 1-3 times

. 4-6 times

. 1 time per day

. 2 times per day

LII-buts):—

24

6. 3 times per day
7. 4 or more times per day”

In order to assess huh and vegetable intake, all responses were converted to times per
week. A response of ‘4 or more times per day’ was coded as 28 times per week. Based
on National F ive-per-day recommendations, any combination of answers where the
student had an intake of at least ﬁve servings of any fruit or vegetable met the
recommendation of ﬁve fruits and vegetables per day23 .

In a review of the current literature, we found that many studies also deﬁned adequate
intake at ﬁve or more servings per dayzs’“. Many studies allowed 100% juice to be
included in the total count of ﬁve servings per day that also used data ﬁom the
YRBSZé’w’s'. Using questions such as the ﬁve on the YRBS are acceptable to assess fruit
and vegetable intake, as established by Serdula et al, who compared similar questions
from the Behavioral Risk Factor Surveillance System (BRF SS) to other diet records”.
3.2.2.4 Dieting Behaviors

Dieting behaviors were included in our study to identify students with potential eating
disorders.

Questions regarding dieting behaviors are:

“During the past 30 days, did you:

Q. 68 exercise to lose weight or keep from gaining weight?
Q. 69 eat less food, fewer calories, or foods lower in fat to
lose weight or keep from gaining weight?
Q. 70 go without eating for 24 hours or more (also called fasting)
to lose weight or keep ﬁ'om gaining weight?
Q. 71 take any diet pills, powders, or liquids without a
doctor’s advice to lose weight or keep from gaining weight?
Q. 72 vomit or take laxatives to lose weight or keep from
gaining weight?”

Answers were coded for all ﬁve questions as “yes or no”. The students who answered

‘no’ to all ﬁve questions regarding dieting were considered to have no risky dieting

25

behaviors. For questions regarding exercise (Q. 68) or eating less (Q. 69), a response of
‘yes’ was also considered as having no risky dieting behaviors. For the questions
regarding fasting (Q. 70), diet pills / powders (Q. 71), and vomiting / laxative use (Q. 72),
a response of ‘yes’ to any one of these questions was considered a risky dieting behavior.
Other studies that have assessed dieting behaviors have also included fasting, diet pills,
and vomiting as extreme or unhealthy dieting behaviorsZ6’34’69’81’82.
3.2.3 Covariates
3.2.3.1 Sports Participation

Participation on sports teams was not used to determine compliance in physical
activity recommendations, however all health behaviors were stratiﬁed by sports team
participation to determine the relationship between the deﬁned healthy behaviors and
BMI in students who did and did not participate on sports teams.
Sports team participation was deﬁned by the question:

“Q. 86 During the past 12 months, on how many sports teams did you play?

1. Oteams
2. lteam
3. 2teams
4. 3ormoreteams”

I deﬁned a sports team participant as a student who had participated on one or more
teams in the last twelve months“.

3.2.3.2 Race

Race was determined by self-report as follows:

“Q. 4 How do you describe yourself?

1. American Indian / Alaska Native

2. Asian

3. Black or Aﬁican American

4. Hispanic or Latino

5. Native Hawaiian / other Paciﬁc Islander
6. White

26

7. Multiple — Hispanic
8. Multiple - Non-Hispanic”

For data analysis, we combined the groups to white (response #6), Black or African
American (response #3), Hispanic (responses #4 & #7), and other (responses #1, #2, #5,
& #8). All analyses were then stratiﬁed by race.

3.2.3.3 Grade in School

Grade in school was assessed as:

“Q. 3 In what grade are you?

1. 9"
2. lo‘“
3. ll'h
4. 12"
S. ungraded or other grade”

Responses for 9-12‘” grades were included in the ﬁnal analysis. Participants with the
response ‘ungraded’ (response #5) were excluded from analyses, and then all analyses

were stratiﬁed by grades 9-12.

3.3 Sample Characteristics
3.3.1 Description of Sample

The original data ﬁle contained observations from 3,630 participants ﬁom 43
Michigan schools. Students who reported having been pregnant were excluded from
analysis (n=l 31). The question was asked as “how many times have you been pregnant or
gotten someone pregnant?” I excluded any female participants that answered “1 time” or
“2 times”, as I had no way of identifying if they were currently pregnant. A total of 608
students (16.7%) were also excluded from further analysis due to missing exposures or

outcomes of interest.

27

Table 2: Sample Selection and Exclusion Criteria;

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Michigan YRBS, 2001
Sample Selection: N

Number of schools participating in state survey 43
Number of students who completed questionnaire 3630
Number of students excluded for pregnancy a 131
Number missing by variables of interest 608
Sex 0
Grade 47
Race 36
Sports Team Participation 108
Fruit / Vegetable Intake 154
Physical Activity 110
Dieting Behaviors 189
Television Viewing 92
BMI b 157
Number of students who completed all questions of interest 2891
TOTAL 2891

 

 

 

 

apregnancy determined by females responding that they had ever been pregnant
bBMI calculated using self-reported height and weight

Adolescents who were missing information on any of the variables were compared to
those who completed all questions of interest. Statistically signiﬁcantly more ninth
graders were missing variables than completed the survey (39.8% compared to 27.9%,
respectively), and statistically signiﬁcantly more black students were missing variables
than those that completed the survey (24.8% compared to 11.4%, respectively).
Differences were also noted between males and females, which approached statistical

Signiﬁcance.

28

Table 3: Population Demographics of Sample Analyzed
compared to adolescents with missing variables,
Michigan YRBS, 2001, weighted data

 

Adolescents with

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample used in Analysis
(n=2,891) Missing Variables
(n=608)
Independent
Variables n % 95% (:1a n % 95% CI
Sex:
Male 1436 51.4 49.4-53.3 337 55.6 51.3-59.9
Female 1455 48.6 46.7-50.6 271 44.4 40.1-48.7
Grade:
9 703 27.9 26.2-29.8 188 39.8 35.3-44.2
10 855 26.3 24.8-27.9 162 24.0 20.4-27.6
11 775 23.5 22.0-25.1 135 20.9 17 .4-24.4
12 558 22.2 20.5-23.9 76 15.3 11.9-18.7
Race:
White 2191 83.7 82.5-85.0 328 66.3 62.8-69.8
Black 323 11.4 10.3-12.6 125 24.8 21 .5-28.0
Hispanic 89 1.1 0.9-1.4 25 2.1 1.2-3.1
Other 288 3.7 3.2-4.1 94 6.8 5.3-8.3
School
Sports:
Participant 1774 61.4 59.5-63.3 300 58.5 53.7-63.3
Non- 1117 38.6 36.7-40.5 200 41.5 36.7-46.3
participant

 

 

Column percents add to 100% with missing variables removed
a95% CI = Conﬁdence Intervals

3.4 Statistical Analysis
3.4.1 Software Used in Analysis

SAS statistical software packages 8.1 and 9.1 were used for all data analysis.
Due to the complex cluster sampling design, weights and stratum were taken into account
for ﬁnal data analyses in SAS 9.1.3.

3.4.2 Weighting Factors .

29

The YRBS is weighted to allow the sample to be representative of the entire
population of publicly educated students, grades 9-12, in the state of Michigan. The
weighting factors are assigned based on a complex formula of base weights and
adjustments for non-responses”. The decision to use the weights is determined by three
factors: proper documentation, legitimate sampling methods, and an overall response
rate of 260%92 (Michigan’s overall response rate for 2001 = 73%).

3.4.3 Data Analysis by Aim
3.4.3.1 Aim 1

T 0 determine both the prevalence of individual and combined nutrition-related

behaviors, and the adherence to national recommendations regarding modiﬁable lifestyle
factors, self-reported, among a population-based sample of publicly educated Michigan
youth, grades 9-12.

Prevalence and 95% conﬁdence intervals were determined for each nutrition-related
health behavior, and stratiﬁed by grade, gender, race, and sports team. participation.
Prevalence and 95% conﬁdence intervals were then determined for students having at
least one, at least two, at least three, and all four nutrition-related health behaviors, also
stratiﬁed by grade, gender, race and sports team participation.
3.4.3.2 Aim 2

T o examine the association between nutrition-related health behaviors and overweight

/ at-risk for overweight, based on BMI calculated ﬁ'om self-reported height and weight,
among a population-based sample of publicly educated Michigan youth, grades 9-12.
BMI-for-age was applied to all subjects using a SAS program provided for public use

by the Centers for Disease Control and Prevention (CDC). Prevalence and 95%

30

conﬁdence intervals were determined for zero, one, two, three, and four nutrition-related
health behaviors by BMI-for-age, and stratiﬁed by gender, grade, race, and sports team
participation. BMI-for-age was then categorized into normal weight (BMI Stu-85m
percentile) and overweight / at-risk overweight (BMI 285‘“ percentile). Nutrition-related
health behaviors (NRHB) were then combined into two groups: 0-1 NRHB and 3-4
NRHB, and the prevalence of normal weight or overweight / at-risk overweight within
each group were then examined. Results were then sub-classed by gender, due to
differences between BMI-for-age among males and females, and stratiﬁed by grade, race,

and sports team participation.

31

CHAPTER 4:RESULTS
4.1 Paper for Publication
4.1.1 Abstract

Purpose: To examine the adherence to national health recommendations regarding
fruit and vegetable intake, physical activity, television viewing, and no risky dieting
behaviors, as well as the relationship between these lifestyle behaviors and the prevalence
of overweight / at-risk of overweight among Michigan high school students.

Methods: Data is from the Michigan Youth Risk Behavior Survey (YRBS), 2001, a
representative sample of publicly educated high school youth in the state, grades 9-12
(n=2,891). Nutrition-related health behaviors were self-reported, and body mass index-
for-age (BMI) was calculated based on national CDC growth charts. Results were
weighted using stratum and individual weights, allowing prevalence estimates and 95%
conﬁdence intervals to be generalized.

Results: In the state of Michigan, 9.4% (n=276) of adolescents participated in all four
deﬁned nutrition-related health behaviors (NRHB), and 7.7% (n=222) participated in all
four NRHB and maintained a healthy body weight (BMI-for-age s'h-ssth percentile).
Girls who participated in 3-4 NRHB were signiﬁcantly more likely to maintain a healthy
body weight than girls with only 0-1 NRHB (84.8%, 95% CI: 81.9-87.8 compared to
72.8%, 95% CI: 67.2-78.5, respectively). A similar trend was noted among boys (70.2%
compared to 59.4%, respectively).

Conclusion: Less than 10% of adolescents, grades 9-12, were adhering to national

recommendations regarding healthy lifestyle choices. Students who were adhering to 3-4

32

nutrition-related health behaviors were more likely to be maintaining a healthy body

weight than those only participating in 0-1 health behaviors.

KEY WORDS:
Adolescent overweight
Fruit and vegetable intake
Physical activity
Television viewing

Multiple health behaviors

4.1.2 Introduction

Pediatric overweight is crn'rently one of the leading health issues among youth in the
United Statesl. In 1999-2000, among adolescents aged 12-19 years, 15.5% were
considered overweights. In 2003-2004, 18% of children aged 6-17 were overweightz; and
among adolescents aged 12-19 years, 34.3% were either overweight, or at-risk of
overweight’. Childhood overweight has many possible consequences, both short-term
and across the lifespan”. Overweight in adolescence is associated with an increased risk
of developing pseudotumor cerebri, sleep disorders, and cholelithiasisl’G. Overweight /
obesity as a child or adolescent is also a predictor of adult overweight and obesity, and
can affect adult mortality and morbidity‘ 1, such as an increased risk for high blood
pressure (HTN), hyperlipidemia, and Type 2 Diabetes Mellitus”. Modiﬁable lifestyle
behaviors, like dietary intake, physical activity, and time spent in sedentary activities
have been associated with overweight in adolescents".

National recommendations address modiﬁable nutrition-related health behaviors such
as fruit and vegetable intake and leisure-time activities”. The 2005 Dietary Guidelines

for Americans recommends maintaining a healthy weight, daily physical activity, and

33

increased fruit and vegetable intake”. For children, they also recommend limiting the
amount of time spent being sedentary, and to increase physical activity”. In 2000, the
Centers for Disease Control and Prevention (CDC) recommended that adolescents should
participate in moderate or vigorous physical activity for at least twenty minutes at least
three times per week, and that physical activity be included as a daily or near-daily
practice”. Fruit and vegetable intake is recommended to be at least ﬁve servings per day,
and is marketed nationally as the “5-A-Day for Better Health” program”. Television
viewing for children is recommended by the American Academy of Pediatrics to be no
more than one or two hours per dayzz.

Studies have shown that physical activity, television viewing, and fruit and vegetable
intake are related to body mass index in childrenw‘éz. Patrick et al reported that
adolescents with lower levels of vigorous physical activity were more likely to be
overweight“). In a study assessing the effect of television viewing on BMI, Kaur et al
reported that increased hours of television viewing was positively assOciated with an
increased BMI”. Boynton-Jarrett et al reported that increased hours watching television
was inversely associated with fruit and vegetable intake“.

In the following study, we identiﬁed four areas of nutrition-related health behaviors to
examine: fruit and vegetable intake, physical activity, television viewing, and no risky
dieting behaviors. We seek to determine the prevalence of these nutrition-related health
behaviors in youth, to describe the demographics of students who are compliant with
these recommendations, and examine the association of these behaviors with overweight /
at-risk overweight among adolescents, grades 9-12, publicly educated in the state of

Michigan.

34

Information regarding the percent of children and adolescent’s meeting national health
recommendations and guidelines is needed to determine their compliance, and assist
policymakers, school ofﬁcials, and health educators in their attempts to address the health
of the nation’s children. This study seeks to contribute to the current lmowledge
regarding the associations of multiple nutrition-related health behaviors to overweight /
at-risk overweight among adolescents.

4.1.3 Methods
Description of Sample

Subjects for this study were participants in the 2001 Youth Risk Behavior Survey in
the state of Michigan. The Michigan YRBS sample from 2001 was gathered by the
Michigan Department of Education, as part of a nation-wide survey organized and
overseen by the Centers for Disease Control and Prevention (CDC). The study design is
a three-stage cluster sample, by area of the state, individual high schools, and then
individual classes from each school. The YRBS is then weighted to be representative of
all publicly educated high-school students in the state of Michigan based on a response
rate of at least 60%. The Michigan YRBS response rate for 2001 was 73%, therefore
sample weights were added to each individual observation to allow the sample to be
representative of publicly educated adolescents, grades 9-12, in the state.

The original data ﬁle contained observations from 3,630 participants hour 43
Michigan schools. Participants who reported having been pregnant were excluded from
analysis (n=131). Subjects who reported their grade as “Imgraded or other grade” were
also excluded ﬁom analysis. A total of 608 students excluded from further analysis due to

missing exposures or outcomes of interest. Adolescents who were missing information

35

on any of the variables were compared to those who completed all questions of interest.
Among those missing variables, signiﬁcantly more ninth graders were missing variables
than completed the survey (39.8% compared to 27.9%, respectively), and signiﬁcantly
more black students were missing variables than completed the survey (24.8% compared
to 11.4%, respectively). Our sample included only students who completed all questions
of interest (n=2,891).

Body Mass Index (BMI-for-age)

To assess the outcome of overweight or obesity among our sample, self-reported
height and weight were used to obtain each individual’s body mass index [BMI: weight
in kg / (height in m)2 ]. The CDC’s Growth Charts were used to assess BMI-for-age
based on percentile height and weight cut-offs”. The 2000 CDC Grth Charts are
gender-speciﬁc and age-speciﬁc. They were developed ﬁom national data sets
(NHANES I-III), and are recommended to be used as a screening tool to identify
overweight / at-risk overweight children and adolescents”. The CDC deﬁnes overweight
and at-risk of overweight based on BMI-for-age as: BMI 295th percentile as overweight,
and BMI 285tll percentile - 95th percentile as at-risk of overweight. Normal weight is
deﬁned as a BMI 25th-85ﬂ‘ percentile. Underweight is deﬁned as BMI-for-age <5th
percentile and children in this group were considered separately, as this may be indicative
of disordered eating or other health problems.

Physical Activity

Physical activity was categorized based on questions assessing both vigorous and

moderate activity levels. A thorough literature review has shown that several studies

used different cut-off points for the determination of ‘active’82’85‘87. Our a priori

36

designations were based on previous research using similar cut-off points and guidelines
set by the International Consensus conference on Physical activity” . For vigorous
exercise, the question was asked “on how many of the past 7 days did you exercise. . . ..for
at least 20 minutes that made you sweat and breathe hard. . .”; for moderate exercise, it
was “on how many of the past 7 days did you participate in physical exercise for at least
30 minutes that did not make you sweat or breathe hard. . . .”. Answers were coded for
each question as “0-7 days/week”. We then created a dichotomous variable that was
deﬁned as meeting criteria with either three days per week or more vigorous, ﬁve days
per week or more moderate, or a combination of ﬁve days per week vigorous and
moderate exercise.
Television Viewing

Television viewing was included in our study as a health-related behavior of interest
due to the well-documented relationship between sedentary behaviors including
television viewing, exercise, and BM122’25’30’87. Television viewing was assessed using
the following question: “on an average school day, how many hours do you watch TV?”
Answers were coded as 0-5 hours per day. Television viewing was considered acceptable
for less than or equal to two hours per day based on the guidelines set in 2001 by the
American Academy of Pediatricszz.
Fruit and Vegetable Intake

Dietary intake of ﬁ'uits and vegetables was based on ﬁve questions regarding the
intake of 100% fruit juice, fruit, carrots, green salad, and other vegetables. In the
Michigan data set, the question regarding potato intake was not used in the variable

created because it did not address fried potato products separately. The following

37

questions were asked: “During the past 7 days, how many times did you eat/drink. . ..
‘100% fruit juice’, ‘ﬁ'uit’, ‘green salad’, ‘carrots’, and ‘other vegetables’?”. Responses
were coded as none, times per week, or times per day. In order to assess fruit and
vegetable intake, all responses were converted to times per'week. A response of ‘4 or
more times per day’ was coded as 28 times per week. Based on National F ive-per-day
recommendations, any combination of answers where the student had an intake of at least
ﬁve servings of any fruit or vegetable met the recommendation of ﬁve fruits and
vegetables per day23 .
Risky Dieting Behaviors

Risky dieting behaviors were included in our study to identify students with potential
eating disorders in order to distinguish them from the population deﬁned as ‘healthy’.
The following questions were asked: “During the past 30 days, did you: ‘exercise’, ‘eat
less food’, ‘go without eating for 24 hours (fasting)’, ‘diet pills, powders, or liquids’, and
‘vomit or take laxatives’ to lose weight or keep from gaining weight?” Answers were
coded for all ﬁve questions as: “yes or no”. The students who answered ‘no’ to all ﬁve
questions regarding dieting were considered to have no risky dieting behaviors. For
questions regarding exercise or eating less, a response of ‘yes’ was also considered as
having no risky dieting behaviors. For the questions regarding fasting, diet pills/powders,
and vomiting/laxative use, a response of ‘yes’ to any one of these questions was
considered a risky dieting behavior.
Covariates:
We examined our nutrition-related variables by covariates of interest. These included

gender, grade in school, race, and sports team participation. Grade in school was used as

38

a demographic variable, and was assessed from the question: “In what grade are you?”
Responses were “9“”, “10m”, “11th”, “12“”, or “ungraded or other grade”. Responses for
9-12th grades were included in the ﬁnal analysis. Race was determined by the question:
“How do you describe yourself?” Responses included “1) American Indian / Alaska
Native, 2) Asian, 3) Black or African American, 4) Hispanic or Latino, 5) Native
Hawaiian / other Paciﬁc Islander, 6) White, 7) Multiple - Hispanic, and 8) Multiple —
Non-Hispanic”. For data analysis, we combined groups to white (response #6), Black or
Aﬁican American (response #3), Hispanic (responses #4 & #7), and other (responses #1,
#2, #5, & #8). Sports team participation was deﬁned by the question: “During the past 12
months, on how many sports teams did you play?” Answers were coded from “none” to
“three or more” teams. We deﬁned a sports team participant as a student who had
participated on one or more teams in the last twelve months.
Statistical Analysis

Prevalence and 95% conﬁdence intervals were determined for each nutrition-related
health behavior by grade, gender, race, and sports team participation. The same analyses
were then done for the number of students who had at least one, at least two, at least
three, and all four healthy behaviors. BMI-for-age was determined for all participants.
Then, the relationship between students who exhibited all four healthy behaviors and who
were also maintaining a healthy body weight was examined. The relationship between
number of nutrition-related healthy behaviors and BMI-for-age was also examined.
Results were stratiﬁed by gender due to differences between BMI-for-age for males and

females.

39

SAS statistical software packages 8.1 and 9.1 were used for all data analysis. To
account for the sample design, stratum and weights were used in the SAS analysis, and
then the design effect (deﬁ) was applied to all 95% conﬁdence intervals for the MI
YRBS, 2001 (deﬂ=2.2). Analyses run were proc frequencies, to determine prevalence of
health behaviors among each demographic variable, and by BMI-for-age.

4.1.4 Results

Among study participants, 51.4% were male and 48.6% were female (see Table 1).
There were slightly more participants in the ninth grade (27.9%) than in the twelﬁh grade
(22.2%). Study participants were 83.7% white, 11.4% black, 1.1% Hispanic, and 3.7%
deﬁned in the ‘other’ category. A high proportion of students participated on sports
teams (61.4%).

Fruit and Vegetable Intake

As seen in Table 2, 19.2% of subjects were meeting the recommendation for fruit and
vegetable intake (deﬁned as 25 servings per day). There were no signiﬁcant differences
by gender, grade, or race. A signiﬁcant difference was observed between students who
participated in school sports compared to non-participants (22.7%, 95% Conﬁdence
Interval (CI): 20.6-24.7 compared to 13.7%, 95%CI: 11.5-15.9, respectively).

Adequate Physical Activity

Signiﬁcant differences were seen across most demographic groups for adequate
physical activity (deﬁned as 23 times per week vigorous, 25 times per week moderate, or
_>_5 week vigorous and moderate). Males had higher prevalence rates than females
(78.3% and 67.1%, respectively). There was a decreasing trend in activity levels after the

9th grade. White students had a statistically signiﬁcant higher prevalence of adequate

4O

physical activity than black students (75.0% and 59.4%, respectively). Students who
participated in school sports had a prevalence of 84.0% (95% CI: 82.2-85.8), while
students who did not participate in school sports had a signiﬁcantly lower prevalence, at
55.1% (95% CI: 51.9-58.2). 9
No Risky Dieting Practices

Gender differences were noted for risky dieting behaviors. Males were more likely to
have no risky dieting practices than females (87.9%, 95% CI: 86.0-89.7 and 74.7%, 95%
CI: 72.3-77.1, respectively). Among individual dieting practices, females had a higher
prevalence of fasting (16.7%), diet pill/powder use (10.2%), and vomiting/laxative use
(8.5%) than did males (8.0%, 5.4%, and 3.5%, respectively; data not shown).
Television Viewing

Nearly 70% of our sample reported watching television for two hours or less per day.
Signiﬁcant differences were noted among white and black students, where 74.2% (CI:
72.3-76.1) of white students met the guideline for healthy television viewing, compared
to 36.7% (95% CI: 31.4-42.0) of black students. Students who participated in school
sports also had a higher rate of compliance to television viewing guidelines than did non-
participants (73.0% and 63.6%, respectively).
Compliance with Multiple Nutrition-Related Health Behaviors

Only 50.1% of the total sample were participating in at least three healthy behaviors,
and only 9.4% (n=276) were found to be compliant with all four nutrition-related health
behaviors (see Table 3). Among subjects who participated in at least one nutrition-
related health behavior, there were no signiﬁcant differences in prevalence by gender,

grade, race, or school sports participation. For students who participated in at least two

41

nutrition-related health behaviors, males had a higher prevalence than females (87.9%,
95% CI: 86.1-89.7 and 81.2%, 95% CI: 79.1-83.3, respectively). There were also
differences by race, where the prevalence of black subjects was signiﬁcantly lower than
white subjects (68.3%, 95% CI: 62.9-73.8 and 87.1%, 95% CI: 85.6-88.5, respectively).
Differences were also seen among students who participated in sports compared to those
who did not, where sports participants had a prevalence of 90.3% for at least two
nutrition-related health behaviors, while non-participants had a prevalence of 75.6%.
Among those subjects with at least three nutrition-related health behaviors, males had a
signiﬁcantly higher prevalence than females (55.6% and 44.3%, respectively), white
subjects had a signiﬁcantly higher prevalence than black subjects (51.8% and 23.0%,
respectively) and sports participants had a signiﬁcantly higher prevalence rate than non-
participants (60.7% and 33.3%, respectively). For students participating in all four health
behaviors, signiﬁcant differences were noted by race, again where white subjects had a
higher prevalence rate for all four behaviors (10.2%, 95% CI 8.9-10.2) than did black
subjects (3.9%, 95% CI 1.6-6.2), and among students who participated in school sports
(13.2%, 95%CI 11.4-14.8) compared to those who did not participate (3 .4%, 95% CI 2.3-
4.6).
BMI-for-age Distribution and Demographic Characteristics

Among our sample, BMI-for-age distribution did not follow a normal distribution
curve. 1.7% of students were considered underweight (BMI-for-age <5th percentile). Of
the remaining subjects, 72.7% were maintaining a normal weight, while 14.7% were
considered at-risk of overweight, and 10.9% were overweight (see Table 4). Females

reported a higher prevalence of normal weight than males (78.7% and 67.0%,

42

respectively), and had a lower prevalence than males for both at-risk for overweight
(11.9% and 17.4 %, respectively) and overweight (7.6 % and 14.0% respectively).
Differences by grade and by sports participation were both noted. White subjects had a
higher prevalence of normal weight than did black subjects (74.2%, 95% CI: 72.3-76.1
and 62.7%, 95% CI: 56.9-68.4, respectively). Conversely, blacks also had a higher
prevalence of at-risk of overweight (20.4%) and overweight (15.6%), compared to white
students (13.9% at-risk for and 10.1% overweight).
Nutrition-Related Health Behaviors and BM-for-age

We then examined prevalence of each NRHB and BMI-for-age (see Table 5a & 5b).
Among each individual NRHB, there was a signiﬁcant difference noted among white and
black students. Black students were signiﬁcantly more likely than white students to be in
the overweight / at-risk of overweight category for fruit and vegetable intake, physical
activity, television viewing, and no risky dieting practices. Due to small sample size, for
further analysis, categories of nutrition-related health behaviors (NRHB) were condensed
to zero to one NRHB, and three to four NRHB. We then compared subjects with 0-1
NRHB and 3-4 NRHB by BMI-for-age (see Table 6). For students with 2 NRHB, white
students were signiﬁcantly more likely to be maintaining a healthy body weight than
were black students (26.4%, 95% CI: 23.0-29.7 compared to 42.9%, 95% CI: 33.7-52.1,
respectively; see Table 6b). Among subjects exhibiting 0-1 NRHB, no differences were
noted by gender, grade, race or school sports participation among either BMI-for-age
category. Of subjects exhibiting 0-1 NRHB, 67.4% were maintaining a healthy weight,
and 30.1% were overweight/ at-risk of overweight. Among subjects with 3-4 NRHB,

signiﬁcant differences were noted between males and females in both BMI-for-age

43

categories. Females with 3-4 NRHB were more likely to be a normal weight than males
with 3-4 NRHB (84.8%, 95% CI: 81.9-87.8 and 70.2%, 95% CI: 66.7-73.6%,
respectively). Females with 3-4 NRHB were therefore also less likely to be overweight /
at-risk of overweight than males with 34 NRHB (13.1%, 95% CI: 10.3-15.9 and 28.6%,
95% CI: 25.2-32.0, respectively). Girls participating in 3-4 NRHB were more likely to
be maintaining a healthy weight than girls with 0-1 NRHB (84.8%, 95% CI: 81 .9-87.8
and 72.8%, 95% CI: 67.2-78.5, respectively). A similar trend, though not statistically
signiﬁcant, was noted among boys.

Given the differences BMI-for-age by gender, the covariates of grade, race and sports
team participation were then examined separately among males and females (data not
shown). Among males, there were no signiﬁcant differences between students
participating in 0-1 NRHB when compared to those with 3-4 NRHB, nor were there any
noted differences by BMI-for-age, grade, race or school sports participation. However,
due to small sample size, conﬁdence intervals were very large.

4.1.5 Discussion

Our results show that a high percentage of adolescents in the state of Michigan are not
in compliance with national recommendations regarding nutrition-related health
behaviors. Among individual nutrition-related health behaviors, 72.8% of subjects were
compliant with physical activity guidelines, 69.4% were compliant with television
viewing recommendations, and only 19.2% reported an adequate intake of fruits and
vegetables, including 100% fruit juice. Students who participated on sports teams were
signiﬁcantly more likely to be compliant with each recommendation. Only 9.4% of

subjects were compliant with all four nutrition-related health behaviors (N RHB), of

which 79% were additionally maintaining a healthy body weight. A total of 7.7% of the
population was both compliant with all four NRHB and maintaining a healthy body
weight. Among females, there was an association between the number of NRHB and
BMI-for-age. Female students who participated in 3-4 NRHB were more likely to be
maintaining a normal weight than those with only 0-1 NRHB (84.8% compared to 72.8%,
respectively). A similar trend that did not meet statistical signiﬁcance was also noted for
boys.

When examining individual nutrition-related health behaviors, signiﬁcant differences
were noted in behaviors related to physical activity among demographic characteristics.
Males had a signiﬁcantly higher prevalence rate for physical activity than did females
(78.3% compared to 67.1%, respectively), which has been similarly noted in other
research28’93. In 1998, using data ﬁom NHANES III, Andersen et al reported a gender
difference in vigorous physical activity”. Among their sample, 86% of boys, aged 14-
16, reported participating in vigorous physical activity 23 times per Week, compared to
only 65% of girls of the same age”.

Racial differences in television viewing were also noted. The prevalence of white
students who watched 52 hours per day was 74.2%, compared to a rate of 36.7% for
black students. This trend has been noted in other research25’93. Using the national 1999
YRBS, Lowry et al reported that 57.2% of subjects watched television for 52 hours per
day; the prevalence of white students watching 52 hours of television per day was
signiﬁcantly higher than the prevalence among black students (65.8% compared to

26.3%, respectively)25.

45

Differences were also noted in nutrition-related health behaviors related to dietary
intake by demographic characteristics. Among our study population, sports participants
had a signiﬁcantly higher prevalence of eating 25 ﬁ'uits and vegetables per day than did
non-participants (22.7% compared to 13.7%, respectively). A similar ﬁnding was
reported by Baumert et al, who reported that 47% of adolescent athletes, deﬁned as
participants of organized sports outside of gym class in grades 9-12, ate at least one
serving of ﬁ'uits and vegetables daily, compared to 40% of non-athletes (p<0.0005)94.
There was a signiﬁcant difference noted in dieting practices by gender in our sample.
The prevalence of males with no risky dieting practices was signiﬁcantly higher than the
prevalence among females (87.9% compared to 74.7%, respectively). In a study
examining weight control behaviors among adolescents who were trying to control / lose
weight, researches reported a similar trend for gender”.

When examining the prevalence of compliance to multiple health behaviors, one study
similar in theory to ours was identiﬁed, though the study is examining some different
health behaviors. In 2004, using a sample derived ﬁom a health plan in the Midwest,
Pronk et al examined four behaviors, physical activity, no tobacco use, diet quality and
body weight among adolescents aged 13-17”. Of that study population, 78.9% had a
BMI-for-age 385'Ill percentile”, similar to our ﬁndings of 72.7% with a BMI-for-age
S85th percentile. A total of 31.2% of adolescents in the Pronk study were compliant with
all of their deﬁned healthy behaviors (adequate physical activity, no smoking, a high
quality diet, and a BMI S85th percentile)”, in contrast to our study, where only 9.4% of
subjects were compliant with all four NRHB (adequate physical activity, television

viewing 52 hours per day, adequate fruit and vegetable intake, and no risky dieting

46

behaviors). In addition, a total of 7.7% of the population was both compliant with all
four NRHB and maintaining a healthy body weight.

Research looking at patterns of healthy lifestyle behaviors has been published among
adult subjects. In a study using data from the BRFSS, Reeves and Rafferty identiﬁed
only 3% of the population nationally was compliant with all four of their deﬁned health
behaviors [fruit and vegetable intake (25 per day), physical activity (230 minutes 25
times per week), no tobacco use, and BMI (18.5-25)]37. Reeves and Rafferty also
reported ﬁndings for the state of Michigan, with only 3% of the population participating
in the same health behaviors96. Berrigan et al reported similar results when examining
ﬁve health behaviors [fruit and vegetable intake (25 per day), physical activity (_>_3 times
per week vigorous or 25 times per week moderate), no tobacco use, moderate alcohol
consumption (5 2 drinks per day for males or $1 drink per day for females), and dietary
fat intake (<30% of total calories)] using data from NHANES III“. In their study, 5% of
the population was compliant with their deﬁned health behaviors“. ‘

Strengths and Limitations

A signiﬁcant limitation of our study was the fact that the YRBS is self-reported data.
In a study examining self-reporting for physical activity, it was noted that there are
signiﬁcant reporting differences when using different reporting tools”. One study
identiﬁed also looked at the questions regarding ﬁ'uit and vegetable intake on the BRFSS
(similar to those on the YRBS), and found that the self-reported intake reported was
similar to intake assessed using multiple diet recalls or diet records”. There has been
considerable research looking at self-report versus measured height and weight, and most

studies have found that there was underreporting of both when using self-reported

47

numbers79‘80’97'99. Some research has suggested that this trend is more prevalent in
females and among subjects that are overweight80’98’99. In our study, prevalence of
females that were overweight / at-risk of overweight was considerably lower than the
rates seen in the male subjects. These results may have been affected by the above trends
concerning self-report, whereby the differences between genders may be a remnant of a
reporting issue.

Another signiﬁcant limitation of this study was the unavailability of information
regarding socioeconomic status (SES). The YRBS does not ask questions regarding
household income, level of parental education, neighborhood information, or access to
food. Therefore, we were unable to examine the associations of nutrition-related health
behaviors and BMI-for-age with gender, grade, or race within the framework of SES, all
of which have been reported elsewhere"”93 ,100-102.

A potential limitation to this study includes the signiﬁcant differences among subjects
who completed the survey when compared to students with missing Variables. These
differences were signiﬁcant for both grade and race, and approaching signiﬁcance for
gender. Our sample is representative only of students publicly educated in the state of
Michigan, so results cannot be generalized to all adolescents in the state, as there may be
signiﬁcant difference among risk behaviors when comparing students educated in

'03. Also, the survey is a cross-sectional design, which does not allow

different settings
temporality to be determined.
The strengths of this study include its ability to address adherence to national health

recommendations among a representative sample of publicly educated adolescents, and to

compare the prevalence of overweight / at-risk of overweight among subjects who are

48

adhering to the deﬁned NRHB. Adherence to national health recommendations for
individual behaviors identiﬁed were similar to research discussed elsewhere. Of
signiﬁcant importance is the relatively low number of students who are participating in
these health behaviors and maintaining a healthy body weight. This is reﬂective of the
growing concern of childhood and adolescent overweight and obesity, and the long-term
health effects associated with both lifestyle choices and overweight and obesity in
adulthood.

Our results suggest that public health interventions should be focused on determining
the barriers to compliance to national health recommendations to help increase the
prevalence of nutrition-related healthy behaviors among adolescents in the state of
Michigan. Students at-risk should be identiﬁed, and speciﬁc strategies implemented to
help these students make positive nutrition-related healthy lifestyle choices. Another
issue of importance to clinicians is the prevalence rate of overweight and at-risk of
overweight among Michigan adolescents, as this could have a signiﬁcant impact on their

health and well-being across the lifespan.

49

Table 1: Population Demographics of Sample Analyzed.
Michigan YRBS, 2001, weighted data.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Adolescents Included
in Analysis
(n=2,891)

Independent
Variables n % 95% (:13
Sex:

Male 1436 51.4 49.4-53.3

Female 1455 48.6 46.7-50.6
Grade:

9 703 27 .9 26.2-29.8

10 855 26.3 24.8-27.9

1 l 775 23.5 22.0-25.1

12 558 22.2 20.5-23.9
Race:

White 2191 83.7 82.5-85.0

Black 323 11.4 10.3-12.6

Hispanic 89 1 .1 0.9-1 .4

Other 288 3.7 3.2-4.1
School Sports:

Participant 1774 61 .4 59.5-63.3

Non- 1 1 17 38.6 36.7-40.5

participant

 

 

 

Column percents add to 100% with missing variables removed
a95% CI = Conﬁdence Intervals

50

Sad 8a .98: N 9 .880 8 .85 $2 8m 035303 383280 on EB was»; 85323. o
.8: 33883538? 8 .8: =8 8% .waumam ”£3283 9508 97.3

e
822%“ H 38.88.: v—En .8 3:80 a .8 308w? {<me an «o smut—om .3
35:83... 085 .8 comma Pa Eouucwﬁou term a 80 .3288. .353» 8% :83 Ba maﬁa ooh: $8— 8 .0338 05688 8 288»? 5::

53:8 .852m 35. 8:08:88.— monzoEnw 3:282 £8528 .8 828588 a 8 women gag—use 3 EB $8-80 .0328 Foam—E 0

$3.35 oocoucaoona a
.8833 38238 33 0.85 .8 >8 8n mwﬁtom 3E .3380»? 8.8 one
.829. :02» .8950 .315 .ooETEb £2: .8 8.85 05 98.5on 82328 o>c 8 women on E? 838%? 98 38b .8 8.8.5 £29

5 e\emo e\e

_U Rena ﬂ.
0 98);:va

_0 $3 .x. n _0 $3 ..\° =
80:85

£38
0

  
 
  

e s can \ meZom mmv bong—U
8:8

  

com .38 $5828<
momomiowomaao oEquoEoD 3 m8M>msom 580$ 8038M I BUSES/Co coco—«>2; ”N 03$.

51

0.0.3085 00000580 $00 -I- _U 0
.Awqumaozmﬁa 00500: 050083050000 0.830000
0.505 000: 00 000 $00 000 0800 0w 00363 0203008 “808088 + 08803 0003 80 00:5 0M 8 .80888 0003 80

00:5 0N 0.8803 0003 .00 00:5 0M .8000 003500 820000 0000 000 00:38» 0M 00000 208003 000 02.0 ”0830000 000002 0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.0-0.0 0.0 :e 0.00-0.00 0.00 000 0.00-0.00 0.00 000 0.00-0.00 0.00 0000 80.000.00.82

0.070.— _ 0.00 000 0.00-0.00 0.00 000— 0. 000.00 0.00 0000 0.00-0.00 0.00 000— 80000000
“3800 80000

0.2-0.0 0.2 00 0.00-0.00 0.00e 00— 0.00-0.00 0.00 000 0.00-0.00 0.00 000 0050

0.0.0.0 _.0 0 0.00-0.00 0.00 00 0.00-0.00 0.00 _0 0.007000 0.00 00 08000:.—

0.0..0._ 0.0 0 _ 0.00- _ .00 0.00 00 0.00-0.00 0.00 000 0.00-0.00 0.00 000 0005

0.0 70.0 0.00 000 0.00-0.00 0.00 00— _ 0.00-0.00 0.00 300 0.00-0.00 0.00 .000 0003
”0000

0070.0 0.0 00 0.00-0.00e 0.00 000 0.00-0.00 0.00 000 0.007500 0.00 000 0—

0.0 70.0 0.0 00 0.00-0.3e 0.00 _00 0.00-0.00 0.00 000 0.00-0.00 0.00 000 0 0

0.0 70.0 0.0 00 0.00-0.00e 0.00 000 0.00-0.00 0.00 000 0.00-0.00 0.00 000 0.

0.0700 0.0 v0 0.00-0.00 000 000 0.00-000 :0 000 0.00-0.00 0.00 000 0
“000.5

00.00 0.0 0 _ _ 0.00-0. 0 0 0.3V :0 0.00- _ .00 0. _0 000 0 0.00- _ .00 0.00 003 0.080...

0.02.0 0.00 000 0.00-0.00 0.00 000 0.00-.00 0.00 0000 0.00-0.00 0.00 00.2 202
n 80:00
0.2-3 _ 0.0 0 e00 0.00-0.3. _ 33 5: 0.00.0.0» 3.00 _ 30 000-30 _ 0.00 _ $00 080

00 $00 .x. 0 _0 $00 ..\o 0 _0 $00 e\.. 0 0 5 $00 ..\e 0
0830000 050000 0830000 050000 0005 0830000 05—000 030 0 830000 050000 08 0.800000
.55 =< 0000. .< 0000. 2. 7.00— 0< ,

 

 

 

 

 

 

 

.300 03503 .800 .053 .3002:
.070 000000 300000800. 0880 005000800000 050000800— 00 08300000

580: 00000000 I 85502 85 :0 8 .005 0000— 00 .025 00000 00 .08 00000 00 .8 00000305 ”0 0000.0.

52

00000200 a00v 8 a00A 000-000-0200 00 005-000 00 0008388 8.0 0.00.30 000 6080200 a00m 000.80-02m 00 000000 00 0008388 0300080

.3800 00380 0.80 05 050: 00000000 00 E3 who-80 0500080 000 Nashua 000003 000 000000 00:82.20.“ 8 00000 00.5 928 000—0 0008 0000—

0
0:085 880080 0.3 8 ..

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.2-0.: 3.2 _02 3.2-0.2 3.2 7.: 0003.000 0.020: $0.30 :12 0500000982
0.2-0.0 _00 :2 _00300 3.20000 3.000.010.0200::04: _00 _0m 50002000
380008000
0.2-0.0 0.: R 0.2-0.0 0.2 00 0.00-0.00 0.: 000 0.0-0.0 00 m .008
mom-0.00 20 2 0.8.0.0 0.: : 0.00-0.00 0.00 mm 0.0-0.0 0.0 0 25000
0.2-0.: 0.2 cm 0.00-0.2 0.00 8 0.00-0.00 0.00 000 0.0-0.0 00 0 0300
0:00 2: 000 0.2-0.2 0.2 50 30.000 0.00 002 0.0-0.0 0.0 00 203
000M
0.2-0.0 0.0 00 0.2.0.0 0: 8 0.00-0.00 0.00 000 0.0-0.0 0.0 2 2
0.2-0.0 0.0 00 0.2-0.0 0.: 00 0.00-0.00 $0 000 0.0.3 0.0 2 :
0.3-0.2 0.2 02 0.8.0.2 0.: E 20-0.00 000 000 0.0-0.0 E S 2
0.2.0.0 0.: on 0.2.0.: 0.2 0: 0.00-0.00 0.00 02. 0.0-0.0 00 ﬂ 0
0080
0.0-0.0 00.0 _0: _0072: _0.: _02 $00.30 _Siai _00-2 :0 _00 20800
0.03.2 :1: _30 $20.: :0: _000 38-3.0 ES C000 00.0-0.0 _00 :0 202
02:50
0.2-0.0 7.2: _0: 0.00-0.2 :0; 3:. 000.000 :00720 0.0-S :.L% 08.0
50000 .x. 0 5&00 .x. 0 5300 .x. 0 05.x.00 ..\o 0 ...08008.0
02000080 _FmoN 02000080 500v 8 500A 0 000000000 500 00 50 00000080 50v

 

 

 

 

.070 000000 0000000003.. 0880 000000880800 000000008000 00 .00—000080 000-80035 ”0 0000.0

.000 020003 .800 .000.» 50032

 

53

w—dtoun: ounoﬁﬁﬂoo e\ona ”—0 a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

£33. .85.:

 

 

3.3.: 03.50am!» c5. :5...”—

 

ﬁwﬁinate _EE.8 _ 2o _ Sq _332 _Sm _3 :3d% :8 _2: “592532
33.8 _22 mm». #35,? _ o3 _ om: _ 32.: _EN _S 3.3m: :2 T: EasEé
“tam—065%
23:: EN 3‘ wowns mi 5 93-2: 5.: : 934$ 08 8 .25
0.8-0.8 in 2 SE: v.3 on 21.3. :3. m 33.: 3% w 25%:
$2.3. 9: Q 53.3 a; as 33.3 ”.3 a $2.3 Nam ; :85
0.8.5.8 3m 52 32.2 35 82 :63: as a 32.2 n: 3 3:;
nooum
_.m~-m.£ :3 S 23.? a: E 23.2 92 8 33.8 on a a
33.: in E qzéﬁ 3: o; 33.2 t: 3 SEQ 35 v2 :
32.8 i; :2 33.8 ta. 5. $2: 33 S 32.8 2: 2. 2
92-0.8 3” m: 33.8 2: mg 33.2 .3 mm 33% 9% E o
”DUSO
5502 :2 _3: _3333615 2.3-x: :2 E. :35: EQCNN 22$
33.8 :2me 343.8 _ is _ up 3.2-98 _m: _S 3.9-2» :5 :2 2s:
“ho—250
23.8 :2 _ am 0.36: Gala”: $3.8 :VNZS 22$ _Nﬁ 3? EA;
5:? .x. = 6.x? .x. = 6&8 .x. a L0 :3 .x. =
beuoaau
Ew_o>»._o>0 ho xmt¢< Ems? 72:52 532330 5 V3.72 £303 _SEoZ
3.839: + 30.5 _> :83 En 3:5 mm 8
62.58.: x025 “on 3:5 mm
538%? x83 Ba 86: va 63v \ mwﬁtom mmv

 

 

.98 82%; .58 am”; .3222
.NTm $3le Scoomﬂoﬁi 98.5 83380820 oEgauonoD 3 owaéomAEm was

33:3. 3%: a. £85 05%»; Ba “we”: 225$ £30m Bﬁomeeﬁaz 3:233 am 03¢

54

mates 853:8 :3 ”_u
_ .

     

ﬂ _
6 $3 a $3 .x. 5 .33 .x. no $3 _x. a

8 5
wage; usage—eh. 82.3.5 Ema c2
.88 com

.Q& 83% manage—81‘ macaw 35:888an oEamumoEoQ 3 owaﬁoﬁuzm EB
@5265 56323 mm 805an @505 Emma 36 ﬂogﬁom HERE BEEMéoEbsz 3:232: 5m 28¢

55

:32? 8a 8a mEovBm Eﬂozouﬁ. 3:83
ﬂogugom .23: 3.a.o¢-:o=E=Z Tm Ea To 55?» $2: 55 m3— o. 25 ﬁﬂoigo 8 “7.1.3 28 E203 $8.5: $8.5 3522— 3mm
mEESE oonuuccoo $3 NC a

£3223 £506 Eu: 0: EB $530? 56322 5:53 30.993 .8135 038%? v5 :3...— EoSEom 5.3:

- .4 -_. me ﬂ

5 $3 _0 $3 o\o : _0 $3

no

“22223 5.8: 833-8552 E 33:23. 5.3: E:§-§=E=z E.

    
 

5% Sow .m _ -o mEoom2ow<
moamtaoﬁﬁo oEaSonoQ E ommLomAEm Ea ﬂoggom 5.3: 36$Méou£=2 6 2an

    

56

Table 6b: Addendum to Table 6: showing only those students with
two Nutrition-Related Health Behaviors and BMI-for-age
by Demographic Characteristics among Adolescents
grades 9-12. Michigan YRBS, 2001, weighted data.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Two Health Behaviorsa
Normal Weight At-risk or Overweight
Category n % 95% CI” n % 95% CI
Total 702 I 69.6 | 66.5-72.6 292 | 2901260320
Gender :
Male 301 64.4 59.8-69.0 164 34.0 29.4-38.6
Female 401 74.4 70.4-78.3 128 24.3 20.4-28.1
Grade:
9 160 64.6 58.0-71.3 76 34.9 28.3-41.6
10 186 61.8 55.9-67.7 111 37.2 31 .3-43.0
1 1 200 74.3 68.9-79.8 67 23.5 18.2-28.8
12 156 79.0 73.3-84.8 38 18.5 13.1-24.0
Race:
White 524 72.1 68.7-75.5 193 26.4 23.0-29.7
Black 82 55.9 46.7-65.1 61 42.9 33.7-52.1
Hispanic 21 60.0 43.7-76.4 15 40.0 23.6-56.3
Other 75 74.4 64.8-83.9 23 24.8 15.3-34.3
School Sports:
Participant 382 71.0 66.8-75.2 148 27.2 23.1-31.3
Nonjartigant 320 68.0 63.5-72.5 144 30.9 26.5-35.4

 

 

 

 

 

 

a Health Behaviors: Fruit and vegetable intake, physical activity, television viewing, and no risky dieting
behaviors

b CI= 95% Conﬁdence Intervals

57

CHAPTER 5: CONCLUSIONS
5.1 Summary of Findings
My results show that the prevalence of adolescents in the state of Michigan that
comply with national recommendations regarding individual healthy lifestyle behaviors is
varied, and among combined multiple health behaviors is very low. Among individual
nutrition-related health behaviors, 72.8% of subjects were compliant with physical
activity guidelines, 69.4% were compliant with television viewing recommendations, but
only 19.2% had an adequate intake of fruits and vegetables. Sports team participants
were signiﬁcantly more likely to be compliant with these three recommendations. In
addition, males had a higher prevalence of no risky dieting behaviors than did female
participants (87 .9% compared to 74.7%, respectively). Among this study population,
10.9% were overweight, and 14.7% were at-risk of overweight. Of the sample, 9.4%
were compliant with all four nutrition-related health behaviors (NRHB). A total of 7.7%
of the population was both compliant with all four NRHB and maintaining a healthy body
weight. Among females, there was an association between the number of NRHB and
BMI-for-age. Female students who participated in 3-4 NRHB were more likely to be
maintaining a normal weight than those with only 0-1 NRHB (84.8% compared to 72.8%,
respectively). This difference was also seen for the total sample ((76.5% compared to
67.4%, respectively), but was not noted among male subjects.
5.2 Comparison of Findings to Prior Literature
When examining individual nutrition-related health behaviors, signiﬁcant differences
were noted in level of physical activity by gender. Males had a signiﬁcantly higher

prevalence rate for physical activity than did females (7 8.3% compared to 67.1%,

58

respectively), which has been noted in other research28’93. In 1998, using data from
NHANES III, Andersen et al reported a gender difference in vigorous physical activity,
with 86% of boys aged 14-16 participating 23 times per week, compared to only 65% of
girls of the same age. Signiﬁcant differences among white and black students were also
observed. Among white students, 75.0% were reporting adequate physical activity, while
only 59.4% of black students reported the same. This has been reported in other
studies'o“, however that may be affected by school environment105 . Another trend noted
in our study was the decreased prevalence of physical activity with increased grade,
which was approaching statistical signiﬁcance when comparing 9th and 12th graders. The
prevalence rate of adequate physical activity among 9til graders was 78.7% (95% Cl:
742-332), while among 12th graders it was 69.7% (95% CI: 64.0-75.4). Other studies
have found a similar inverse relationship: as grade in school increases, physical activity
levels decreasem’m.
Racial differences were noted for television viewing in this study. The prevalence of
white students who watched 52 hours per day was 74.2%, compared to a rate of 36.7%
for black students. This trend has been noted in other research25’93. Using the national
1999 YRBS, Lowry et al reported that 57.2% of the sample watched television for 52
hours per day, with the prevalence among white students considerably lower than among
black students (65.8% compared to 26.3%, respectively)”. The YRBS only asks about
television; it does not address other media like computer use and video games as seen in
other studies93’m7.

Differences were also noted in ﬁ'uit and vegetable intake in this study. Sports

participants had a signiﬁcantly higher prevalence of eating 25 fruits and vegetables per

59

day than did non-participants (22.7% compared to 13.7%, respectively). A similar
ﬁnding was reported by Baumert et al, who reported that 47% of adolescent athletes,
deﬁned as participants of organized sports outside of gym class in grades 9-12, ate at
least one serving of fruits and vegetables daily, comparedto 40% of non-athletes
(p<0.0005)94.

There was a signiﬁcant difference noted in no risky dieting practices by gender in this
sample. The prevalence of males with no risky dieting practices was signiﬁcantly higher
than the prevalence among females (87.9% compared to 74.7%, respectively). In a study
examining weight control behaviors among adolescents who were trying to control / lose
weight, researches reported a similar trend for gender”.

Overweight among children and adolescents has increased over the past several
decades”. My study population followed this trend: 10.9% were overweight, and 25.6%
were either overweight or at-risk of overweight. Overweight among males in this study
was 14.0% and among females was 7.6%. Nationally, reported preValence rates in 2003-
2004 of overweight males, aged 12-19, were at 18.3% and females of the same age were
at 16.4%3.

When examining the prevalence of participation in multiple nutrition-related health
behaviors (NRHB), signiﬁcant differences were noted by demographic characteristics.
By gender, differences were noted among males and females when examining prevalence
of adhering to multiple NRHB: males were more likely than females to be participating
in at least 2 NRHB (87.9% compared to 81 .2%, respectively) and more likely to be
participating in at least 3 NRHB (55.6% compared to 44.3%, respectively). This trend

was also seen among sports participants, who were more likely to be participating in at

60

least 2, at least 3, and all 4 NRHB than were non-sports participants. No other studies
looking at the same four deﬁned NRHB were identiﬁed for the purposes of comparison.

One study looking at multiple health behaviors in adolescents was identiﬁed, but for
the purposes of comparison, the health behaviors are not all the same. In 2004, using
subjects extracted from a health plan in the Midwest, Pronk et a1 examined four healthy
behaviors among adolescents, aged 13-17 years: physical activity (deﬁned as moderate
activity 230 minutes 25 days per week or vigorous activity 220 minutes 23 times per
week), no tobacco use, diet quality (using Recommended Food Score), and healthy body
weight (BMI-for-age S85th percentile)”. Of Pronk’s study population, 78.9% were
maintaining a healthy weight”, similar to my ﬁndings of 72.7%. A total of 31.2% of
adolescents in the Pronk study were compliant with all of their deﬁned healthy behaviors
(adequate physical activity, no smoking, high diet quality, and healthy body weight)”, in
contrast to my study, where only 9.4% of subjects were compliant with all four NRHB
(fruit and vegetable intake 25 per day, adequate physical activity, television viewing _<_2
hours per day and no risky dieting behaviors). In addition, only 7.7% of my total study
population was both participating in all four NRHB and maintaining a healthy body
weight (BMI-for-age 25th — 85th percentile).

Research looking at patterns of healthy lifestyle behaviors has been published among
adult subjects. In a study using data from the BRF SS, Reeves and Rafferty identiﬁed
only 3% of the population nationally was compliant with all four of their deﬁned health
behaviors [ﬁ'uit and vegetable intake (_>_5 per day), physical activity (30 minutes at least 5
times per week), no tobacco use, and BMI of 18.5-25]”. Reeves and Rafferty also

reported ﬁndings for the state of Michigan, with only 3% participating in the same health

61

behaviorsg6. Berri gan et al reported similar results when examining ﬁve health behaviors
using data from NHANES 11188. In their study, 5% of the population was compliant with
their deﬁned health behaviors [ﬁ'uit and vegetable intake (25 per day), physical activity
(23 times per week vigorous or 25 times per week moderate), no tobacco use, moderate
alcohol consumption (51 drink per day for females or 5?. drinks per day for males), and
dietary fat intake (330% of total calories)].
5.3 Strengths and Limitations

A signiﬁcant limitation of this study was the fact that the YRBS is self-reported data.
In a study examining self-reporting for physical activity, it was noted that there are
signiﬁcant reporting differences when using different reporting tools”. One study
identiﬁed also looked at the questions regarding ﬁ'uit and vegetable intake on the BRF SS
(similar to those on the YRBS), and found that the self-reported intake reported was
similar to intake assessed using multiple diet recalls or diet records”. There has been
considerable research looking at self-report versus measured height and weight; most
studies have found that there was underreporting of both when using self-reported
numbers79’80’97‘99. Some research has suggested that this trend is more prevalent in
females and among subjects that are overweight80’98’99. In my study, prevalence of
females that were overweight / at-risk of overweight was considerably lower than the
rates seen in the male subjects. These results have likely been affected by the above
trends concerning self-report, especially among females.

Another signiﬁcant limitation of my study was the unavailability of information
regarding socioeconomic status (SES). The YRBS does not ask questions regarding

household income, level of parental education, neighborhood information, or access to

62

food. Therefore, I was unable to examine the associations of nutrition-related health
behaviors and BMI-for-age with gender, grade, or race within the framework of SES, all
of which have been reported elsewherel3'93’lmm.
A potential limitation to this study includes the signiﬁcant differences in subjects who
completed the survey when compared to students with missing variables. These
differences were signiﬁcant for both grade and race, and approaching signiﬁcance for
gender. Also important to note: this sample is representative only of students publicly
educated in the state of Michigan, so results cannot be generalized to all adolescents in
the state, as there may be signiﬁcant differences among risk behaviors when comparing

students educated in different settings, like private or home schools103

. Also important to
note is that the very nature of a cross-sectional study does not allow temporality to be
determined.

The strengths of this study include its ability to address adherence to national health
recommendations among a representative sample of publicly educated adolescents, and to
compare the prevalence of overweight / at-risk of overweight among subjects who are
adhering to the deﬁned NRHB. From a public health standpoint, this information can be
used to help assess current education needs, decide on appropriate strategies to target
adolescents, and help identify other risk factors that are affecting the health of the
population. Another strength with the YRBS is that it is collected in public schools, rather
than at home, and this method of data collection has been shown to produce a higher and

108

more accurate prevalence of risky behaviors .

5.4 Suggestions for Future Research

63

This study suggests many avenues for further research into the relationship between
current health recommendations and compliance to them, and their relationship to BMI-
for-age in adolescents. Further research should take into account the above limitations,
like SES, family environment, and include computer use and video games in addition to
television viewing. Also interesting would be further examination of the relationship of
sports team participation to nutrition-related health behaviors.

From a clinical standpoint, further research needs to examine not only the relationship
between nutrition-related health behaviors and overweight among children and
adolescents, but must also focus on tools for use in an office setting. Clinicians would
beneﬁt from tools developed to help them identify at-risk youth, and to have guidelines
for intervention to ensure standardization of care.

From a public health standpoint, further research should examine not only the
relationship of these behaviors and overweight, but also the effectiveness of current
health guidelines, as there are many students who are not currently compliant with many
recommended health behaviors. Both state-wide governmental ofﬁces and school
officials would beneﬁt from better educational tools for children and adolescents to help
increase awareness and compliance with these healthy behaviors.

From an economic standpoint, the burden of overweight and obesity among children is
already measurable in healthcare dollars. In 1997-1999, an estimated $127 million
dollars was spent on hospital admissions due to overweight / obesity among childrenm.
As the rates of overweight / obesity among children and adolescents rise, so will the risk
of long-term health issues, health care costs, and lost days of productivity at work, which

will together place a large economic burden on the United States in years to come.

Research that can address these nutrition-related health behaviors, their association with
overweight among adolescents, and can identify effective education tools to increase

compliance will go a long way to lessening these burdens.

65

APPENDIX

Table 1: Deﬁning Risky Dieting Behaviors: Fasting, diet pills, and/or vomiting to lose
weight; compared to students who are either not dieting, or are eating less
and/or exercising to lose weight: Michigan YRBS, 2001, weighted data

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Dieting Behaviors: Total Population Females Males
n % 95% Cle n % 95% CI 11 % 95% Cl

Risky dieting:

Fasting a 364 12.2 11.0-13.5 249 16.7 14.7-18.7 115 8.0 6.5-9.6

Diet pills b 226 7.7 6.9-8.8 148 10.2 8.6-11.9 78 5.4 4.1-6.6

Vomiting c 173 5.9 5.0-6.8 122 8.5 7.0-10.1 51 3.5 2.4-4.5
No risky dietin :

Not dieting 2346 81.5 799-83 .0 1084 74.7 72.3-77.1 1262 87.9 86.0-89.7

 

a Fasting = students who are fasting (go without eating for 24 hours or more) to lose weight

b Diet pills = students who are using diet pills, powders, liquids (without doctor’s advice) to lose

weight

c Vomiting = students who are vomiting or taking laxatives to lose weight

d Not dieting = students who are either not dieting, or are eating less and/or exercising to lose weight
° 95% CI - Conﬁdence Intervals

66

 

88: mm .395»? x83 hon 8E: mN 653 bits Roam—E Rev hem mmﬂtom mm 8.35 938%? Ba :3". "E0323 >28: a

$3.8:— 8:03:00 l _0 $3
.Awﬁaﬁoimin “233.: games—huge: Eazﬁon
@506 bar e: can 2% .6..— Ezo: Nw wake? ”Sago—3 ”3.832: + 395w? x83 can 85: nN .5 68.58.: :83 ham

9

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

«2.5.2 :51: 736% _Ee TE Iguana 3.3. _S :43: :3 _2 aeﬁoeaeoz
32.5 :31: inseam 362219855 43a _m 32.3% I? T: assented
“mtomm _oosom
33$ ”.2 : 33.3. was mm - o: e - 9% N 550
$3.8 own w ”.842 Em a - 3a _ - a: _ gene:
33.2 ”.8 mm 33.3. eee 8 93-3 e? m $3.9 3% a Em
32.2 23 R 2.2.8 v.8 a: 33.2 ”.3“ a 33.8 NS 2 an;
523*
21.8 SH MN 22.3 wee we - 3e _ - Sam _ N.
$3.2 EN em 32% 23 we 3.3.: «.mm m Evie NS w :
32.8 «.2 no 32% 3e 2: 36-3w 3m e 33.90. 58 o S
22:: SN em 25% 36 3 313 0.3 a 87$: 35 a o
“OUNHO
3.3.2 _Neia Siege :2 3M: Kasai :2 _: $.56; _eee _2 eased
:32 _eeQee $5-3m :52: 2.23.: Cele _eawai :3 3 2a:
”.5250
33.x 7.515. Shame 33:3 33.2 :2 _: _ﬁeae ZS _mN 53
6:3 .x. a 6:3 as. a 6.x? as a n6 :3 as. e
bouoaau
£303.30 no xmta< Emma? _SEoZ E30530 no xmtn< Haws? 13:02
hertz—om 5.3: ago a. Eegaim 5.5: PEN
San 8:363 . SON .mmy; :me22 .NT@ 322w mEoomﬂow< wcoﬁe mocmtouowano
oEmEmoEoD 3 owséomézm was Boggom _Emom Ban—um 53532 28 no PEN 5MB magnum an 2an

67

Table 2b: Students with two, three, or all four Nutrition— Related Health Behaviors and BMI-for-age by Demographic Characteristics

among Adolescents grades 9-12. Michigan YRBS, 2001, weighted data

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Two Health Behaviors a Three Health Behaviors Four Health Behaviors
Normal Weight At—risk or Overweight Normal Weight At-risk or Overweight Normal Weight At-risk or Overweight

Category n % 95% CI n % 95% CI n % 95% CI n % 95% CI n % 95% c1 n % 95% CI
Total 702 | 69.6 | 66.5-72.6 292 I 29.0 I 26.0—32.0 872 I 75.9 I 73.2-78.6 257 I 22.7 I 20.1-25.3 222 I 78.9 I 73.5-84.2 47 | 18.6 | 13.4-23.8
Gender:

Male 301 I 64.4 I 59.8-69.0 I 164 I 34.0 I 29.4—38.6 I 430 I 69.3 I 65.4-73.3 | 180 I 29.3 I 254-331 I 121 | 73.5 | 65.8-81.2 I 39 I 25.9 I 18.2-33.6

Female 401 I 74.4 I 704-783 I 128 I 24.3 | 20.4-28.1 | 442 I 84.4 I 812-877 I 77 I 14.2 I 11.0-17.3 I 101 I 86.7 I 806-928 I 8 I 8.0 | 3.5-12.4
Grade:

9 160 64.6 58.0-71.3 76 34.9 28.3-41.6 214 73.1 67.7-78.6 74 25.7 20.3-31.1 51 78.8 67.6-90.0 15.4 5.2-25.6

10 186 61.8 55.9-67.7 111 37.2 31.3-43.0 233 73.7 68.5-78.9 76 25.6 20.5-30.7 59 72.2 61.4-83.0 26.5 15.9-37.0

11 200 74.3 68.9—79.8 67 23.5 18.2-28.8 249 80.4 75.8-85.1 58 18.1 13.5—22.6 65 81.5 72.8-90.3 15.7 7.8-23.5

12 156 79.0 73.3-84.8 38 18.5 13.1-24.0 176 77.2 71.5-83.0 49 20.4 14.9-26.0 47 83.7 71.7-95.7 7 16.3 4.3-28.3
Race:

White 524 72.1 68.7-75.5 193 26.4 230-297 726 76.4 73.6-79.2 206 22.2 19.4-24.9 181 78.8 73.1-84.6 40 18.4 129-239

Black 82 55.9 46.7-65.1 61 42.9 33.7-52.1 44 70.1 57.4-82.9 20 29.9 17.1-42.6 10 76.4 54.9-97.8 3 23.6 2.2-45.1
Hispanic 21 60.0 43.7-76.4 15 40.0 23.6—56.3 20 61.9 42.1-81.7 10 32.2 12.5-51.9 3 100 100 - - -

Other 75 74.4 64.8-83.9 23 24.8 15.3-34.3 82 76.0 67.1-85.0 21 22.7 14.0-31.5 28 80.7 72.8-88.6 4 19.3 11.4-27.2
School Sports:
Participant 382 I 71.0 | 66.8-75.2 | 148 27.2 23.1-31.3 | 634 I 77.1 I 74.0-80.2 | 173 21.4 | 18.4-24.4 I 193 I805 | 74.8-86.1 36 | 17.0 I 115-224
Non participant 320 | 68.0 | 635-725 | 144 I 30.9 26.5-35.4 238 I 73.0 I 67.8-78.2 | 84 26.0 I 20.9-31.2 I 29 I 69.0 | 55.1—82.9 11 I 28.5 | 156-41.4

 

 

a . - n I u u o -
Healthy behaviors: Fruit and vegetable intake 25 serv1ngs per day; phy81cal act1v1ty elther Z3 t1mes per week v1gorous, ZS tunes
per week moderate, or 25 times per week vigorous + moderate; television viewing :32 hours per day; and no risky dleting
behaviors (fasting/laxative use/diet pills/vomiting).

95% CI — Conﬁdence Intervals

68

 

 

ﬂag:— oououucoo 15 $3
.aéESEE .238: 3533536 2333
maﬁa b2.— o: 98 Saw Hon 35: NW 350? 56338 ”88308 + 32cm? :83 hon BE: mN .5 .8889: x33 hon

no.5. mN Japan; x33 .3 8:5 mm .656 £36m 1363“. San hon 35:3 mN 8.85. 0330»? v5 «3.5 ”£2553 3:3: a

a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

33.8 _ ”.3 _ 8 _ 33w; _ is _ m2 _ $3.8 _ Em _ G C.$-_.% H 5% Co «geogm .52
”2-0.8 :ﬁgoﬂ _358 _2: 3:. 3.01.2 _Em :N $25.3 _ﬂm _ov Eases
”mtodw _oocom
32.: EN 2 23.3 2% _m 33.8 2; 2 Nﬁédm 3m 2 .25
Egé 92 m 32.8 3% w is; 5.3 o - - o 252:
9915 9% E 92.9% q: 3 32am 0.: 2 22.? in mm :85
333 SN 2: 33.8 NE a; «mama 3A a 35.0% 0% $ 833
50g
33.2 :3. 3 5:426 2K as $3.2 2m 2 53.3 36 M: N.
22.2 mg 3... N. 5-9% am 8. .33.: 2m 2 32.3 ”.3 mm :
”34.2 «.2 $ 35% <8 N2 32% a? R $3.3 New 2 2
$3. R 3a _m 33.8 ”do m: 33.8 an t 32.8 0% R o
”DUN—O
33.2 P93 is 35.8 L S: 4 _mm 33.8 _Nsm _E 9%-”; 3% _2: 8.22 :85
6 :3 .x. a 6 $3 .x. = 5 :3 .x. a 35 $3 : =
Ecuauav
EmEEo>O go “3.7% 2303 22502 £303.55 ho VEE< Ema? _EEoZ

 

 

 

 

.2253 5.8: EEéézEzz Sam 3 8:;

 

w 2222.05 5.3: USES—acct“: Z :0 3 PEN

 

 

 

8% “Sims? ,Som .mmy; sawmaomz .NTo 392w
832 83863“ mace“ 8558230 oEmﬁonoQ 3 ommﬁwzﬁm c5. ﬁcEﬁom ”Baum BEBMEoEE—Z ”m 2an

69

mﬁtouﬁ ooaoumaoo I _0 3mm

Awqa_Eo>\m=a $.38: gag—\wﬁumﬂv who—>33
mucus gm: 0: 23 S8 .6; 3:2. NW 9:305 Era—>28 28038 + map—ow; :83 non 8:5 mm .5 3808:. xooB hon

8:5 nN .mzohow_> v.33 hon 85: mN 655 @358 _moﬁba San hon mwﬁiom mN 83:: 038%? was :8“—

n

225:3 3:8: a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

33.: _Embm __.£-N.: It :2 _Eméom :2 T? 33.3% :§ In. “ﬁgogmcoz
3:; _2: T; 3.8-03 :5 :3 _332 I: :N :353 i: _8 East“;
“mtomm_oo:om
<83 3; 2 33.8 in a 5.2.3 92 m 33.: «a a 550
$23 a: m gown: NS 2 335 SN m 357mm :5 w 25%:
towns 3a a “3.92 v: 8 23.2 EN 2 $3.8 QR S :85
37; ed a 33.3 53 m: 23.2 EN a. 33.8 55 02 323
”comm
323 E : $3.; 25 2: 31am <2 3 Siam 2% _m 2
$73 Na S 239 E a: 33.: 9mm 2 $33.8 mm 3 :
98-; a: 3 33.: ”.3 o: 33$ EN 8 v3-23 v: E 2
33.2 3: NM 23.? mi N2 $2.: 0.3 2 933% _ﬁ 2 o
Human—U
$72: _ 3 _ a 33.; :3 Hg 23.2 _ 2m _8 32.3 _mﬁ _mom museum ”EB
6:8 .x. = 5:3 .x. 6x3 .x. a “.6 £3 x. =
5.5930
£395.30 ho xmr3< Ems? 72:52 2302230 ho V3.72 EMS? _mEoZ
€£>asom 5.3: toga—omécmxbaz .Ech 3 99:; a 32253“ 5.3: nova—oméozwzaz 2.0 3 PEN

 

 

 

 

58 82m? $8 any; 8322 .N To 88% 838$
Eoom20v< macaw moumtﬂoﬁmnv oﬁmﬁmoﬁoﬂ 3 uwm¢omézm Ea ﬂora—«om £133 BEEMEonﬁsz ”v 2an

70

10.

ll.

12.

13.

15.

l6.

l7.

l8.

BIBLIOGRAPHY

Must A, Strauss RS. Risks and consequences of childhood and adolescent obesity. Int J Obes
Relat Metab Disord 1999;23 Suppl 2:82-11.

Forum on Child and Family Statistics. America's Children 2006: In brief: Key National
lndicator's of Well-Being, 2006. ‘

Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of
overweight and obesity in the United States, 1999-2004. JAMA 2006;295(l3):1549-55.

Troiano RP, Flegal KM. Overweight children and adolescents: description, epidemiology, and
demographics. Pediatrics 1998;101(3 Pt 2):497-504.

Ogden CL, Flegal KM, Carroll MD, Johnson CL. Prevalence and trends in overweight among US
children and adolescents, 1999-2000. JAMA 2002;288(l4): 1728-32.

Strauss R. Childhood obesity. Curr Probl Pediatr l999;29(l): 1-29.

Brownell K and Fairburn C, Eating Disorders and Obesity: A Comprehensive Handbook; second
edition. New York: Guilford Press, 2002.

Hagarty MA, Schmidt C, Bernaix L, Clement IM. Adolescent obesity: current trends in
identiﬁcation and management. J Am Acad Nurse Pract 2004;l6(11):481-9.

Hassink S. Problems in childhood obesity. Prim Care 2003;30(2):357-74.

Childhood Obesity in the United States: Facts and Figures. Institute of Medicine of the National
Academies, ed, 2004.

Deckelbaum RJ, Williams CL. Childhood obesity: the health issue. Obes Res 2001;9 Suppl
422398-2438.

Jolliffe D. Extent of overweight among US children and adolescents Born 1971 to 2000. Int J
Obes Relat Metab Disord 2004;28(1):4-9.

Miech RA, Kumanyika SK, Stettler N, Link BG, Phelan JC, Chang VW. Trends in the association
of poverty with overweight among US adolescents, 1971-2004. JAMA 2006;295(20):2385-93.

Drewnowski A, Specter SE. Poverty and obesity: the role of energy density and energy costs. Am
J Clin Nutr 2004;79(l):6-l6.

Cummins S, Macintyre S. Food environments and obesity-neighbourhood or nation? Int J
Epidemiol 2006;35(l):100-4.

Nelson MC, Gordon-Larsen P, Song Y, Popkin BM. Built and social environments associations
with adolescent overweight and activity. Am J Prev Med 2006;3 1(2): 109-17.

Nutrition and Your Health: Dietary Guidelines for Americans
Aim for Fitness, Build a Healthy Base, Choose Sensibly for good health. US Dept of Agriculture
and US Dept of Health and Human Services, ed. Vol. 5th Edition, 2000.

2005 Dietary Guidelines Committee Report: Executive Summary. ADA Times. Vol. Volume 2,
2004 September/October, pages 9-10.

71

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

US Dept of Health and Human Services and US Department of.Agriculture Dietary Guidelines for
Americans, 2005. Vol. 6th Edition US. Government Printing Ofﬁce, Washington DC, January
2005.

US Dept of Health and Health. The National Survey of Children‘s Health 2003. In: Health
Resources and Services Administration Maternal and Child Health Bureau, ed US. Department of
Health and Human Services; Rockville, Maryland, 2005. .

Centers for Disease Control and Prevention. Promoting Better Health for Young People through
Physical Activity and Sports: A Report to the President; Fall 2000.

American Academy of Pediatrics: Children, adolescents, and television. Pediatrics
2001;107(2):423-6.

Subar AF, Heimendinger J, Patterson BH, Krebs-Smith SM, Pivonka E, Kessler R. Fruit and
vegetable intake in the United States: the baseline survey of the Five A Day for Better Health
Program. Am J Health Promot l995;9(5):352-60.

Sussman MP, Jones SE, Wilson TW, Karin L. The Youth Risk Behavior Surveillance System:
updating policy and program applications. J Sch Health 2002;72(1):13-7.

Lowry R, Wechsler H, Galuska DA, Fulton JE, Kann L. Television viewing and its associations
with overweight, sedentary lifestyle, and insufﬁcient consumption of fruits and vegetables among
US high school students: differences by race, ethnicity, and gender. J Sch Health
2002;72(10):413-21.

Story M, Neumark-Sztainer D, Sherwood N, Stang J, Murray D. Dieting status and its relationship
to eating and physical activity behaviors in a representative sample of US adolescents. J Am Diet
Assoc l998;98(10):1127-35, 1255.

Proctor MH, Moore LL, Gao D, Cupples LA, Bradlee ML, Hood MY, Ellison RC. Television
viewing and change in body fat ﬁ'om preschool to early adolescence: The Framingham Children's
Study. Int J Obes Relat Metab Disord 2003;27(7):827-33.

Andersen RE, Crespo CJ, Bartlett SJ, Cheskin LJ, Pratt M. Relationship of physical activity and
television watching with body weight and level of fatness among children: results from the Third
National Health and Nutrition Examination Survey. JAMA 1998;279(12):938-42.

Eisenmann JC, Bartee RT, Wang MQ. Physical activity, TV viewing, and weight in US. youth:
1999 Youth Risk Behavior Survey. Obes Res 2002;10(5):379-85.

Pate RR, Heath GW, Dowda M, Trost SG. Associations between physical activity and other health
behaviors in a representative sample of US adolescents. Am J Public Health l996;86(11):1577-81.

Patton GC, Selzer R, Coffey C, Carlin JB, Wolfe R. Onset of adolescent eating disorders:
population based cohort study over 3 years. BMJ l999;318(7186):765-8.

Krowchuk DP, Kreiter SR, Woods CR, Sinal SH, DuRant RH. Problem dieting behaviors among
young adolescents. Arch Pediatr Adolesc Med l998;152(9):884~8.

Calderon LL, Yu CK, Jambazian P. Dieting practices in high school students. J Am Diet Assoc
2004; 104(9): 1369-74.

Neumark-Sztainer D, Hannan PJ, Story M, Perry CL. Weight-control behaviors among adolescent
girls and boys: implications for dietary intake. J Am Diet Assoc 2004;104(6):913-20.

72

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

Ford ES, Ford MA, Will JC, Galuska DA, Ballew C. Achieving a healthy lifestyle among United
States adults: a long way to go. Ethn Dis 2001;] l(2):224«3l.

Kim S, Popkin BM, Siega-Riz AM, Haines PS, Arab L. A cross-national comparison of lifestyle
between China and the United States, using a comprehensive cross-national measurement tool of
the healthfulness of lifestyles: the Lifestyle Index. Prev Med 2004;38(2): 160-71.

Reeves MJ RA. Healthy lifestyle characteristics among adults in the United States, 2000. Arch
Intern Med 2005;165(8):854-7.

Centers for Disease Control and Prevention. BMl--Body Mass Index: About BMI for Children
and Teens. www.cdc.gov, accessed 11/6/06.

Koplan JP, Liverman CT, Kraak VI. Preventing childhood obesity: health in the balance:
executive summary. J Am Diet Assoc 2005;105(l):l3 1-8.

Centers for Disease Control and Prevention. Overweight and Obesity: Childhood Overweight:
Overweight Prevalence. 2006. www.cdc.gov, accessed 12/8/06.

America's Children: Key National Indicators of Well-Being. Health Indicators, 2003.

Crawford PB, Story M, Wang MC, Ritchie LD, Sabry ZI. Ethnic issues in the epidemiology of
childhood obesity. Pediatr Clin North Am 2001;48(4):855—78.

Dietz WH. Health consequences of obesity in youth: childhood predictors of adult disease.
Pediatrics l998;101(3 Pt 2):518-25.

Cook S, Weitzman M, Auinger P, Nguyen M, Dietz WH. Prevalence of a metabolic syndrome
phenotype in adolescents: ﬁndings from the third National Health and Nutrition Examination
Survey, 1988-1994. Arch Pediatr Adolesc Med 2003;157(8):821-7.

Schwirnmer JB, Burwinkle TM, Varni JW. Health-related quality of life of severely obese children
and adolescents. JAMA 2003;289(14):1813-9.

Williams J, Wake M, Hesketh K, Maher E, Waters E. Health-related quality of life of overweight
and obese children. JAMA 2005;293(l):70-6.

Schonfeld-Warden N, Warden CH. Pediatric obesity. An overview of etiology and treatment.
Pediatr Clin North Am l997;44(2):339-61.

Dietz WH. Childhood weight affects adult morbidity and mortality. J Nutr l998;128(2
Suppl):4l 18-414S.

Bray GA. Contemporary Diagnosis and Management of Obesity and the Metabolic Syndrome.
Vol. third edition. Newtown, Pennsylvania: Handbooks in Health Care.

Styne DM. Childhood and adolescent obesity. Prevalence and signiﬁcance. Pediatr Clin North Am
2001;48(4):823-54, vii.

Guo SS, Wu W, Chumlea WC, Roche AF. Predicting overweight and obesity in adulthood ﬁom
body mass index values in childhood and adolescence. Am J Clin Nutr 2002;76(3):653-8.

Freedman DS, Khan LK, Serdula MK, Dietz WH, Srinivasan SR, Berenson GS. The relation of
childhood BMI to adult adiposity: the Bogalusa Heart Study. Pediatrics 2005;] 15(1):22-7.

73

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

65.

66.

67.

68.

69.

Must A, Jacques PF, Dallal GE, Bajema CJ, Dietz WH. Long-term morbidity and mortality of
overweight adolescents. A follow-up of the Harvard Growth Study of 1922 to 1935. N Engl J Med
1992;327(19): 1350-5.

Krebs NF, Jacobson MS. Prevention of pediatric overweight and obesity. Pediatrics
2003;] l2(2):424-30.

Centers for Disease Control and Prevention. Overweight and Obesity: Childhood Overweight:
Contributing Factors. 2006. www.cdc.gov, accessed 12/8/06.

Warden NA, Warden CH. Biological inﬂuences on obesity. Pediatr Clin North Am
2001 ;48(4):879-91.

Dietz W. Factors associated with childhood obesity. Nutrition 1991 ;7(4):290-1.

Birch LL, Fisher JO. Development of eating behaviors among children and adolescents. Pediatrics
l998;101(3 Pt 2):539-49.

Birch LL, Davison KK. Family environmental factors inﬂuencing the developing behavioral
controls of food intake and childhood overweight. Pediatr Clin North Am 2001;48(4):893-907.

Patrick K, Norman 6], Calfas KJ, Sallis JF, Zabinski MF, Rupp J, Cella J. Diet, physical activity,
and sedentary behaviors as risk factors for overweight in adolescence. Arch Pediatr Adolesc Med
2004;158(4):385-90.

Boynton-Jarrett R, Thomas TN, Peterson KE, Wiecha J, Sobol AM, Gortmaker SL. Impact of
television viewing patterns on fruit and vegetable consumption among adolescents. Pediatrics
2003;112(6 Pt l):1321-6.

Kaur H, Choi WS, Mayo MS, Harris KJ. Duration of television watching is associated with
increased body mass index. J Pediatr 2003;]43(4):506-11.

Storey ML, Forshee RA, Weaver AR, Sansalone WR. Demographic and lifestyle factors
associated with body mass index among children and adolescents. Int J Food Sci Nutr
2003;54(6):491-503.

Pangrazi RP, Beighle A, Vehige T, Vack C. Impact of Promoting Lifestyle Activity for Youth
(PLAY) on children's physical activity. J Sch Health 2003;73(8):317-21.

Myers L, Strikmiller PK, Webber LS, Berenson GS. Physical and sedentary activity in school
children grades 5-8: the Bogalusa Heart Study. Med Sci Sports Exerc 1996;28(7):852-9.

5-a-Day for Better Health. Foundation National Cancer Institute and the Produce for Better Health
Foundation, ed.

Krebs-Smith SM, Cook A, Subar AF, Cleveland L, Friday J, Kahle LL. Fruit and vegetable
intakes of children and adolescents in the United States. Arch Pediatr Adolesc Med
l996;150(1):8l-6.

Field AE, Austin SB, Taylor CB, Malspeis S, Rosner B, Rockett HR, Gillrnan MW, Colditz GA.
Relation between dieting and weight change among preadolescents and adolescents. Pediatrics
2003;] 12(4):900-6.

Lowry R, Galuska DA, Fulton IE, Wechsler H, Kann L. Weight management goals and practices

among US high school students: associations with physical activity, diet, and smoking. J Adolesc
Health 2002;3 1(2): 133-44.

74

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

85.

Brener ND, Kann L, Kinchen SA, Grunbaum JA, Whalen L, Eaton D, Hawkins J, Ross JG.
Methodology of the youth risk behavior surveillance system. MMWR Recomm Rep 2004;53(RR-
12): 1-13.

Michigan Department of Education. 2001 Michigan Youth Risk Behavior Survey. August 2002.

Strauss RS. Comparison of measured and self-reported weight and height in a cross-sectional
sample of young adolescents. Int J Obes Relat Metab Disord l999;23(8):904-8.

Centers for Disease Control and Prevention. Overview of the CDC Growth Charts: Online
Module. www.cdc.gov, accessed 3/1/07.

Ogden CL, Carroll MD, F legal KM. Epidemiologic trends in overweight and obesity. Endocrinol
Metab Clin North Am 2003;32(4):74l-60, vii.

Use and Interpretation of the CDC Growth Charts: An Instructional Guide. National Center for
Disease Prevention and Health Promotion, 2001. www.cdc.gov, 1/31/05.

Freedman DS, Khan LK, Serdula MK, Dietz WH, Srinivasan SR, Berenson GS. Inter-
relationships among childhood BMI, childhood height, and adult obesity: the Bogalusa Heart
Study. Int J Obes Relat Metab Disord 2004;28(1):10-6. '

Mei Z, Grammar-Shawn LM, Pietrobelli A, Goulding A, Goran MI, Dietz WH. Validity of body
mass index compared with other body-composition screening indexes for the assessment of body
fatness in children and adolescents. Am J Clin Nutr 2002;75(6):978-85.

Pietrobelli A, Faith MS, Allison DB, Gallagher D, Chiumello G, Heymsﬁeld SB. Body mass
index as a measure of adiposity among children and adolescents: a validation study. J Pediatr
l998;132(2):204-10.

Morrissey SL, Whetstone LM, Cummings DM, Owen LJ. Comparison Of self-reported and
measured height and weight in eighth-grade students. J Sch Health 2006;76(10):512-5.

Brener ND, McManus T, Galuska DA, Lowry R, Wechsler H. Reliability and validity of self-
reported height and weight among high school students. J Adolesc Health 2003;32(4):281-7.

Pesa JA, Turner LW. Fruit and vegetable intake and weight-control behaviors among US youth.
Am J Health Behav 2001 ;25(l):3-9.

Middleman AB, Vazquez I, Durant RH. Eating patterns, physical activity, and attempts to change
weight among adolescents. J Adolesc Health l998;22(1):37-42.

American College of Sports Medicine position stand. The recommended quantity and quality of
exercise for developing and maintaining cardiorespiratory and muscular ﬁtness in healthy adults.
Med Sci Sports Exerc l990;22(2):265-74.

US Department of Health and Human Services: Ofﬁce of Public Health and Science. Healthy
People 2010 Objectives: Draft for Public Comment. Washington DC: US Department of Health
and Human Services, September 15, 1998.

Hu FB, Manson JE, Stampfer MJ, Colditz G, Liu S, Solomon CG, Willett WC. Diet, lifestyle, and
the risk of type 2 diabetes mellitus in women. N Engl J Med 2001;345(l l):790-7.

75

86.

87.

88.

89.

90.

9].

92.

93.

94.

95.

96.

97.

98.

99.

100.

10].

Pratt M, Macera CA, Blanton C. Levels of physical activity and inactivity in children and adults in
the United States: current evidence and research issues. Med Sci Sports Exerc 1999;31(1]
Suppl):8526-33.

Levin S, Lowry R, Brown DR, Dietz WH. Physical activity and body mass index among US
adolescents: youth risk behavior survey, 1999. Arch Pediatr Adolesc Med 2003;157(8):8 16-20.

Berrigan D, Dodd K, Troiano RP, Krebs-Smith SM, Barbash RB. Patterns of health behavior in
US. adults. Prev Med 2003;36(5):615-23.

Sarkin JA, Nichols JF, Sallis JF, Calfas KJ. Self-report measures and scoring protocols affect
prevalence estimates of meeting physical activity guidelines. Med Sci Sports Exerc
2000;32(1): 149-56.

Barlow SE, Dietz WH. Obesity evaluation and treatment: Expert Committee recommendations.
The Maternal and Child Health Bureau, Health Resources and Services Administration and the
Department of Health and Human Services. Pediatrics 1998;102(3):E29.

Serdula M, Coates R, Byers T, Mokdad A, Jewell S, Chavez N, Mares-Perlman J, Newcomb P,
Ritenbaugh C, Treiber F, et al. Evaluation of a brief telephone questionnaire to estimate fruit and
vegetable consumption in diverse study populations. Epidemiology 1993 ;4(5):455—63.

Centers for Disease Control and Prevention. 2005 Youth Risk Behavior Surveys: State and Local
Youth Risk Behavior Surveys Weighting Procedures. 2005.

McMurray RG, Harrell J S, Deng 8, Bradley CB, Cox LM, Bangdiwala S]. The inﬂuence of
physical activity, socioeconomic status, and ethnicity on the weight status of adolescents. Obes
Res 2000;8(2): 130-9.

Baumert PW, Jr., Henderson JM, Thompson NJ. Health risk behaviors of adolescent participants
in organized sports. I Adolesc Health l998;22(6):460-5.

Pronk NP, Anderson LH, Crain AL, Martinson BC, O’Connor PJ, Sherwood NE, Whitebird RR.
Meeting recommendations for multiple healthy lifestyle factors. Prevalence, clustering, and
predictors among adolescent, adult, and senior health plan members. Am J Prev Med 2004;27(2
Suppl):25-33.

Prevalence of healthy lifestyle characteristics--Michigan, 1998 and 2000. MMWR Morb Mortal
Wkly Rep 200] ;50(35):758-6] .

Shapiro JR, Anderson DA. The effects of restraint, gender, and body mass index on the accuracy
of self-reported weight Int J Eat Disord 2003;34(l): 1 77-80.

Elgar FJ, Roberts C, Tudor-Smith C, Moore L. Validity of self-reported height and weight and
predictors of bias in adolescents. J Adolesc Health 2005;37(5):371-5.

Abraham S, Luscombe G, Boyd C, Olesen I. Predictors of the accuracy of self-reported height and
weight in adolescent female school students. Int J Eat Disord 2004;36(1):76-82.

Chang VW, Lauderdale DS. Income disparities in body mass index and obesity in the United
States, 1971-2002. Arch Intern Med 2005;165(l8):2l22-8.

Wang Y, Zhang Q. Are American children and adolescents of low socioeconomic status at

increased risk of obesity? Changes in the association between overweight and family income
between 197 ] and 2002. Am J Clin Nutr 2006;84(4):707-16.

76

102.

103.

104.

105.

106.

107.

108.

109.

Fahlman MM, Hall HL, Lock R. Ethnic and socioeconomic comparisons of ﬁtness, activity levels,
and barriers to exercise in high school females. J Sch Health 2006;76(]):12-7.

Grunbaum JA, Lowry R, Kann L. Prevalence of health-related behaviors among alternative high
school students as compared with students attending regular high schools. J Adolesc Health
2001;29(5):337-43.

Adams J. Trends in physical activity and inactivity amongst US 14-18 year olds by gender, school
grade and race, 1993-2003: evidence from the youth risk behavior survey. BMC Public Health
2006;6z57.

Richmond 'I'K, Hayward RA, Gahagan S, Field AE, Heisler M. Can school income and
racial/ethnic composition explain the racial/ethnic disparity in adolescent physical activity
participation? Pediatrics 2006;] 17(6):2 158-66.

Nelson MC, Neumark-Stzainer D, Harman PJ, Sirard JR, Story M. Longitudinal and secular trends
in physical activity and sedentary behavior during adolescence. Pediatrics 2006;] l8(6):e1627-34.

Roberts DF. Media and youth: access, exposure, and privatization. J Adolesc Health 2000;27(2
Suppl):8-l4.

Kann L, Brener ND, Warren CW, Collins JL, Giovino GA. An assessment of the effect of data
collection setting on the prevalence of health risk behaviors among adolescents. J Adolesc Health
2002;3 l(4):327-35.

Overview of the IOM‘s Childhood Obesity Prevention Study. In: Institute of Medicine of the
National Academies, ed, 2004.

77

\Iliililljﬂlliﬂlliljlllllglllliixl