University of Southern California Dissertations and Theses

Adolescent cigarette use and psychological distress: relation to adult obesity risk

(USC Thesis Other)

Adolescent cigarette use and psychological distress: relation to adult obesity risk

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
ADOLESCENT CIGARETTE USE AND PSYCHOLOGICAL DISTRESS: RELATION
TO ADULT OBESITY RISK

by

Guneet Kaur

A Dissertation Presented to the
FACULTY OF GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PREVENTIVE MEDICINE)

May 2008

Copyright 2008 Guneet Kaur

ii
Dedication

To my children, Angad and Arjun, for all their unconditional love and laughs. To
my husband for his support. To my parents, sister, and younger brother for their constant
love and support in so many ways.

iii
Acknowledgements

There are a number of people to thank for their support over the years, and here
are only a few of them. I would like to thank my advisor, Dr. Mary Ann Pentz, for her
copious amount of support, guidance, and direction at all times. I am proud to be her
student!! I would also like to thank Dr. Chih-Ping Chou for providing me the opportunity
to work closely with him over these years and gain an understanding as well as get
comfortable with various advanced statistical procedures. I am honored to have known
him and worked with him on a one-to-one level. I also would like to thank Dr. Donna
Metz for providing me with the opportunity to work with her on a research project and
for providing excellent guidance all along. Dr. Steve Sussman, agreed to be a member on
the committee at a moment’s notice and has been a constant source of encouragement and
advice. I greatly appreciate and thank him for that. I also greatly appreciate Dr. Carol
Prescott’s commitment to being the outside member of my dissertation committee and for
her detailed and extensive feedback on the proposal as well as the dissertation. I want to
thank her for her involvement at every stage in this process beginning with the qualifiers.
I would also like to thank Marny Barovich, Roberta Kenny, Eric Wang, Valentina
Andreeva and Lei Duan for their friendship and help in so many ways through the
process. Last but not the least, I would like to thank my husband and children for their
never-ending love and support which kept me going through these years. I would like to
thank my parents and in-laws for their countless prayers inspite of being overseas, and
my sister and brother for their countless phone calls and moral support.
iv
This research was funded by the National Institute of Drug Abuse, Grant # F31
DA022186-01, awarded to Guneet Kaur (Parent Grant # 5R01 DA 0110366, PI: Mary
Ann Pentz).

v
Table of Contents

Dedication ii

Acknowledgements iii

List of Tables vii

List of Figures viii

Abstract ix

Chapter 1: Introduction 1
Obesity: Prevalence and risk factors 1
Review of Literature 2
Limitations of existing studies 16
Overview of current study 21

Chapter 2: Methods 26
Study design 26
Sample 27
Measures 28

Chapter 3: Study 1 32
Introduction 32
Methods 36
Results 42
Discussion 47

Chapter 4: Study 2 52
Introduction 52
Methods 54
Results 62
Discussion 67

Chapter 5: Study 3 72
Introduction 72
Methods 75
Results 80
Discussion 88

vi

Chapter 6: Discussion 92
Brief summary of findings 92
Overall implications 99
Strengths and limitations 102
Overall conclusions 105
Future research 109

References 110

Appendices 135
Appendix A: Complete structural model 135
Appendix B: Model with both maladaptive and adaptive coping 136
136

vii

List of Tables

Table 1: List of measures used at each developmental period 31

Table 2: Equivalence of sample without BMI vs. with BMI 43

Table 3: Summary descriptive statistics and correlations 44
Table 4: Summary results of tests of measurement invariance 66
for males and females

Table 5: Mean differences on outcomes for adaptive and 81
maladaptive coping groups

Table 6: Summary descriptive statistics and correlations for coping items 82
Table 7: Summary results of tests of measurement invariance for low 88
vs. high adaptive coping

viii
List of Figures

Figure 1: Measurement model (Kansas City) 45

Figure 2: Structural model (Kansas City) 47

Figure 3a: Measurement model (Males) 63

Figure 3b: Measurement model (Females) 64

Figure 4: Final structural model 67

Figure 5a: Measurement model (Adaptive coping) 83

Figure 5b: Measurement model (Maladaptive coping 84

Figure 6a: Structural model (Adaptive coping) 85

Figure 6b: Structural model (Maladaptive coping) 86

Appendix A: Complete structural model 135

Appendix B: Model with both maladaptive and adaptive coping 136
ix
Abstract

Cigarette use, depression, and obesity are major health threats to society. The
three studies of this dissertation examined the inter-relations of cigarette use,
psychological distress, physical activity, self-rated health, coping, and body mass index
(BMI) among subjects aged 11-34 years in a drug abuse prevention trial, the Midwestern
Prevention Project (MPP), in Kansas City and Indianapolis.
Study 1 examined the longitudinal relations between cigarette use in
adolescence to cigarette use, distress, physical activity, and self-rated health, and BMI in
adulthood in Kansas City. Results from this study suggest that adolescent cigarette use
was associated with continuing cigarette use, high distress, and lower BMI in adulthood.
Findings from Study 1 suggest that prevention efforts need to counteract the weight-
control ‘benefits’ of smoking with a focus on healthy ways to loose or maintain weight.
Study 2 examined gender differences in the interrelations of cigarette use,
distress, physical activity, self-rated health, and BMI longitudinally, from adolescence to
adulthood in the ‘stacked’ dataset (included both Kansas City and Indianapolis). Tests of
measurement invariance demonstrated that overall data were invariant across males and
females. Out of a total of 25 parameters tested, only two covariances and two regression
weights were different between the two groups. Results from Study 2 suggest that
universal prevention programs be augmented with separate components for males and
females that address the varying correlations and paths that were different between the
two groups.
Study 3 examined the roles of maladaptive and adaptive coping as a
mediator/moderator in the distress-BMI relation in the stacked dataset (included both
x
Kansas City and Indianapolis). Structural equation modeling showed that maladaptive
coping did not operate as a mediator in the distress-BMI relation. Tests of measurement
invariance demonstrated that data were not invariant across low adaptive and high
adaptive coping groups, with 10 out of a total of 15 factor loadings tested different
between the two groups. Results from Study 3 demonstrate a moderator effect of adaptive
coping in the relations between constructs studied.
Overall, findings from the dissertation will inform the development of prevention
program that target multiple constructs instead of focusing on a single construct.

1
Chapter 1: Introduction

Obesity: Prevalence and risk factors
Overweight and obesity have become major public health problems, resulting in
multiple medical consequences such as increased risk for coronary heart disease, high
blood pressure, high blood cholesterol level, type 2 diabetes mellitus, gallbladder disease,
osteoarthritis, and overall mortality (Maffeis 2001). In addition, the prevalence of
overweight in children and adolescents

and obesity in adults in the United States has
increased over

several decades (Ogden 2006). Between 1980 and 2004 the prevalence of
obesity increased from 15% to 33% among adults and the prevalence of overweight in
children increased from more than 6% to 19% in the United States (Ogden 2007). The
significant psychosocial and health problems associated with obesity (Dietz 1998), as
well as the significant costs of obesity-related treatment (Wellman 2002), argue strongly
for identification of targets at earlier developmental periods. It is logical then that
potentially modifiable risk factors that are strongly predictive of obesity even before it
starts should be explored and identified in the prevention of obesity.
According to the energy balance model of adiposity, weight gain occurs if caloric
intake exceeds energy expenditure (Rosenbaum 1997). Thus, the two traditional proximal
risk factors shown to be determinants of obesity include low levels of physical activity
(Berkey 2000; Proctor and Ellison 2003) and unhealthy dietary practices such as
overeating (Prentice 2001) and eating patterns (e.g. consumption of sweetened beverages,
sweets, and meats) (Nicklas 2003). However, results from obesity prevention efforts that
have utilized physical activity enhancement and improving nutritional intake have been
2
mixed (Thomas 2006). Researchers have asked whether there are other novel risk factors
in adolescence that make the development of obesity more likely (McElroy 2004). If
identified, such variables might serve as useful targets for preventive efforts and provide
insight into other possible preventative approaches to overweight.

Review of literature
This section summarizes existing research on relations between each of these risk
factors, specifically cigarette use, psychological distress, physical activity, and self-rated
health to obesity risk. Additionally, the relation of these risk factors with each other is
also reported.

Cigarette use and Obesity
Adolescent studies
An extensive literature has demonstrated links between smoking and body weight,
specifically weight concern, in adolescents. Existing studies have shown a positive
association between concern and dissatisfaction with body weight and smoking
(Neumark-Sztainer 2006), particularly among adolescent females (French 1995; Grigg
1996; Potter 2004). Findings from existing studies support a strong positive association
between smoking and weight concern in adolescent females (Potter 2004). Two
hypotheses have been proposed to explain this smoking-weight concern relationship in
adolescence(French 1995; French 1996). First, in response to societal pressure to be thin
(Wiseman 1998), adolescent females may initiate smoking as a weight control measure
(Tomeo 1999). Many adolescents

believe that smoking is an effective weight control
3
method, and

girls tend to adhere to this belief more than boys do (Camp 1993; Klesges
1997). Secondly, according to the problem behavior theory (Jessor 1991),
experimentation with smoking and engaging in unhealthy dietary practices, such as eating
junk food and abstaining completely from eating can be thought of as part of clustering of
risk factors during adolescence (Burke 1997; Escobedo 1997; DuRant 1999). These
youth risk behaviors cluster because these behaviors serve similar functions for
adolescents such as independence from adults, rebelliousness, etc (French 1996; Spruijt-
Metz 1999). From a problem-behavior perspective, adolescent cigarette smoking as a
normative transgression could be motivated by goals such as rejecting the norms of
conventional society, affirming membership in a peer group, asserting independence from
parents, or being seen as more mature (Turbin 2000). Specifically, smoking is legal only
for adults and hence, initiation and experimentation during this period may be a
convenient way for adolescents inclined towards deviance or rebellion to engage in risk-
taking or anti-social behavior (Simons-Morton 1999).
Compared to the extensive number of studies examining the weight concern-
smoking association, relatively few studies have examined the relation between actual
body weight and smoking in adolescents. While a positive association between smoking
and body weight (i.e. increased smoking was associated with higher body weight ) for
both genders has been suggested both cross-sectionally (Townsend 1991; Freedman
1997; Robinson 1997; Klesges 1998), and longitudinally (Cooper 2003), a non-
significant association in both males and females between these behaviors has also been
found primarily cross-sectionally (Page 1993; Crawley 1995; Crisp 1998; Crocker 2001;
Strauss 2001). A few studies found an association between these two variables only in
4
one gender. While two cross-sectional studies have found a positive relation between
smoking and body weight only in males (Gofin 1982; Tomeo 1999), two longitudinal
studies reported a negative association between smoking and weight (Austin 2001) as
well as a positive association between smoking initiation and overweight status only in
females (Cawley 2004). The study conducted by Austin and Gortmaker in 2001 included
a total of 932 6
th
and 7
th
graders. After a follow-up of 20 months from baseline, smoking
initiation at baseline was inversely associated with body weight at follow-up. The study
conducted by Cawley and colleagues in 2004 included a total of 9,022 adolescents aged
12-16 years measured at baseline and at three follow-ups spanning 4 years.
These differences in study findings may be related to differences in measurements
of key variables, specifically BMI. Though BMI was consistently used to determine
weight status, however, while some studies measured height and weight objectively to
compute BMI (Townsend 1991; Crawley 1995; Freedman 1997; Crisp 1998), others used
self-reported height and weight provided by the subjects (Page 1993; Robinson 1997;
Klesges 1998; Crocker 2001). Though self-reported BMI is an efficient way of obtaining
information about BMI, however, self-reported data tends to underestimate weight and
overestimate height, thus underestimating BMI (Basterra-Gortari 2007; Larsen 2008).
Thus, in the adolescent population, some evidence supports a positive relation
between smoking and body weight. However, not all studies have found an association
between these variables. Several authors propose that a positive association between
smoking and body weight among adolescents may indicate that some overweight
adolescents initiate smoking to control weight (Halek 1993; Crisp 1998; Klesges 1998).
5
The association of youth smoking behaviors with other behaviors related to higher body
weight, such as physical inactivity and unhealthy eating behaviors, also may be a
contributing factor to positive relation between smoking and weight (Potter 2004).

Adult studies
The finding of a positive relation or no relation between body weight and smoking
among adolescents is inconsistent with findings for adults, which have generally
suggested that body weight and current smoking status are inversely associated (i.e.
smoking results in lower body weight). Findings from cross-sectional studies have found
that current adult smokers have tended to weigh less than nonsmokers in both males and
females (Bamia 2004), only in males (Akbartabartoori 2005), and only in females (John
2005). Additionally, evidence from two prospective studies have found smoking
initiation to be associated with reduced weight gain (Lissner 1992; Klesges 1998) While
the study conducted by Lissner and others in 1992 included only females (N=1,291), the
second study (Klesges 1998) included both males and females (N=5,115). Non-smokers
weighed more than smokers in both genders though the relation between smoking and
body weight differed by ethnicity and was found only in Black adults and not in Whites.

Depression and Obesity
Negative affect has been posited as a risk factor for obesity (Hoppa 1981).
According to the affect regulation model, dysphoric individuals eat in an effort to provide
comfort or distraction from negative emotions, which results in increased risk for weight
gain (Stice 2005). Additionally, obesity may develop as a result of reduced interest and
6
enjoyment of physical activity, increased appetite for energy-rich comfort foods, and due
to medication for depression treatment (Fava 2000; Dixon 2003). Conversely, the social
stigmatization associated with obesity is believed to engender chronic embarrassment,
shame and guilt, all of which may lead to affective disorders (Friedman 1995).

Adolescent studies
Few prospective studies exist to help establish the causal direction of depression
predicting obesity. In three longitudinal studies which included only females, while one
study reported a non-significant association between depression and follow-up obesity
(Bardone 1998), the other two females-only studies found that association between
depression and later obesity risk was significant (Franko 2005), whereby depression in
adolescence predicted obesity in young adulthood. In longitudinal studies which included
both males and females, both a positive relation (i.e. existence of baseline depression was
associated with increased risk of obesity at follow-up) between depression and BMI in
emerging adulthood (Pine 2001; Goodman 2002)at 1-year follow-up (Goodman 2002)
and short-term body weight change in mid-adulthood (Hasler 2005) as well as a non-
significant relation at adulthood (Pine 1997) was reported. In longitudinal studies which
included both gender as well as offered some comparisons across gender, a positive
relation between depression and BMI was suggested only in females, whereby late
adolescent depression predicted adult obesity (Richardson 2003; Hasler 2005)and
childhood depressive symptoms predicted both increased weight gain (Hasler 2005;
Anderson 2006) as well as increased occurrence of obesity in adulthood (Hasler 2005).
However, for males, existing studies have either shown a lower magnitude of association
7
between depression and body weight (Hasler 2005) as compared to females or no
association between these two variables (Anderson 2006). Thus, findings from these
studies suggest that the association between body weight and depressive disorders differs
between women and men.

Adult studies
The majority of studies conducted on adults have focused on the relation between
depression and weight gain/chance cross-sectionally (Young 1990; Carter 2000) and
longitudinally (Noppa 1981; Young 1990; DiPietro 1992; Barefoot 1998; Carter 2000).
In the study conducted by Barefoot and colleagues, males and females who reported high
levels of depressive symptoms gained less weight than their non-depressed counterparts
if they were initially lean, but more if they were initially heavy. In another longitudinal
study, positive depression-weight gain associations were reported in males (i.e. baseline
depression predicted weight gain at follow-up) and positive depression-weight loss
associations were found in females (i.e. baseline depression lead to follow-up weight
loss) (DiPietro 1992). One study specifically examined the association of depression with
obesity in adults (Roberts 2003). This study found a non-significant association between
depression at baseline and obesity at 5-year follow-up in both males and females (Roberts
2003).

Cigarette use and Depression
An extensive literature has demonstrated linkages between cigarette use and mental
health problems and the co-morbidities between the two have been well documented.
8
Three broad explanations for this association

have been proposed. First, according to the
―self-medication‖ hypothesis, it has been suggested that major depression

plays a causal
role in smoking—increasing the risk for

smoking initiation and the progression to regular
and heavy

smoking (Kendler 1993). This argument focuses on the chemical properties of
nicotine and suggests

that smokers use nicotine to medicate their

depressed mood
especially in adults (Breslau 1998). In the adolescent population, however, there has not
been much support for the ‗self-medication‘ hypotheses for smoking initiation (Difranza
in press). Second, long-term exposure to nicotine itself may influence neurobiological
systems that have an etiological role in depression (Pomerleau 1984; Carmody 1989),
thereby suggesting smoking as a risk factor for subsequent depression. Third, another
commonly posited hypothesis in the literature

suggests that the comorbidity between
smoking and depression may be non-causal i.e. common genetic and family
environmental factors, or individual environmental vulnerabilities, or risk factors may
predispose to both

smoking and depression (Hughes 1986; Breslau 1993; Kendler 1993).

Adolescent studies
Since both cigarette use and depression have been shown to predict each other in
research, longitudinal studies conducted on adolescents are categorized here in two ways:
first, depression as a risk factor for smoking, and second, smoking as a risk factor for
depression.

9
Depression as a risk factor for smoking
While some studies which included both males and females have reported a non-
significant association between baseline depression and follow-up smoking in the
combined sample (Brook 1998; McGee 1998; Wu 1999; Goodman 2000; Brook 2004),
other studies found that depression at baseline predicted initiation of smoking at follow-
up in the total sample (Brown 1996; Patton 1998; Wang 1999; Windle 2001; Fergusson
2003; Rodriguez 2005). Ebeling and colleagues (Ebeling 1999) found a significant
association between depression and regular smoking at follow-up only for females. In
studies which included only females as the sample, a similar significant relation between
depression and smoking was also found (Bardone 1998). Other research has shown that
males with higher depressive scores were more likely to have tried smoking over the
follow-up period (Killen 1997) and more likely to have used cigarettes to cope with
depressed mood as compared to females (Repetto 2005).

Smoking as a risk factor for depression
The majority of longitudinal studies conducted on adolescents have found a
significant positive association between baseline smoking and depressive disorders at
follow-up (Brown 1996; Brook 1998; Brook 2002; Brook 2004; Galambos 2004) (Wu
1999; Goodman 2000; Patten 2001; Windle 2001). Existing studies have found smoking
as a significant predictor of later depression in both males and females (Kandel 1986;
Kandel 1986; Choi 1997; Duncan 2005) as well as in females only (Steuber 2006). The
study conducted by Steuber and Danner in 2006 on adolescents in grades 7-12 found that
10
females were more likely to be depressed at 1-year follow-up compared to males if they
started smoking recently.

Adult studies
Similar to studies conducted in the adolescent population, longitudinal research
among adults has also suggested a bi-directional causal path between smoking and
depression whereby smoking predicts depression and depression predicts later smoking.
Existing research has shown current male and female smokers at baseline to have the
higher risk of depression at follow-up (John 2004). Findings from prospective studies
have also suggested a positive association between a history of nicotine dependence at
baseline and follow-up major depressive disorders (Breslau 1993; Breslau 2001).
However, a non-significant relation between early smoking and later depression has also
been noted in adults (Aneshensel 1983).
Further, depression at baseline has also shown to be associated with increased
follow-up smoking in existing research (Ismail 2000). In the study conducted by Ismail &
colleagues (2000) (Ismail 2000), the relationship between past mental disorders and
increased smoking at follow-up differed according to the age-groups with the strongest
association for the youngest (16-21 years) and the oldest (51-75 years) age groups.
Evidence from longitudinal studies has also reported a non-significant association
between prior depression and later smoking (Aneshensel 1983).
Although these studies reflect a complex mix of temporality, cause, and effect,
however, overall findings from these studies provide evidence for a significant
association between depression and cigarette use in adolescents and adults.
11
Self-rated health and its relation with other constructs studied
Self-rated health is an important indicator of general well-being, which
encompasses physical, mental, and social well-being (Okosun 2001). It has been studied
with reference to its ability to predict morbidity and mortality (Appels 1996; Idler 1997;
McGee 1999; Ahmad 2005) and risk behaviors such as smoking, exercise, sleep, body
weight, and alcohol consumption in both adults (Segovia 1989; Gregor 2006) as well as
adolescents (Geckova 2001; Vingilis 2002; Sleskova 2005). Not only it is an easily
administered and inexpensive measure (Manora 2000), it is also a multi-dimensional and
holistic measure that potentially represents an overall cognitive or attitudinal set that
could drive change in behavior.
Limited evidence from existing studies has shown poor self-rated health to be a risk
factor for obesity primarily among older adults (Kaplan 2003). The relation between
subjective rating of health and obesity risk has been relatively less examined in the
adolescent and the young adult population though findings from a recent cross-sectional
study have found poor self-rated health to be associated with increased body mass index
(Lim 2007).The majority of studies in literature have examined poor self-rated health as a
consequence of existing overweight and obese status. In these cross-sectional studies,
obese subjects were found to report poorer subjective health as compared to normal
weight subjects (Ferraro 1995; Ostbye 1995; Cairney 1998; Okosun 2001; Guallar-
Castillon 2002).
Other research has shown depression to be associated with poor self-rated health
(Choi 1997; Mulsant 1997; Gazmararian 2000; Chi 2005) and depression to be predictive
of decline in self-related health again primarily among older adults (Han 2001; Han
12
2002). Existing studies have also shown nicotine dependence and psychiatric disorders to
contribute to the longitudinal prediction of low self-rated health (Vingilis 2002; John
2005). Further, individuals reporting poor self-rated health have been shown to exhibit
reduced physical activity patterns compared to those reporting very good health (Norman
2002).

Physical activity and its relation with other constructs studied
The positive association between reduced physical activity and increased risk of
obesity and weight gain is well-established in the literature (Haapanen 1997; Ball 2001;
Drøyvold 2004; Catenacci 2007). Additionally, existing research has shown both
smoking status to be inversely related to frequency of physical activity, whereby smokers
were less likely to participate in exercise than non-smokers (Wilson 2005), as well as
cigarette use to have no association with physical activity (Nerin 2004). Further, studies
have also shown that there is significant positive association between reduced physical
activity levels and poor self-reported health (Ferraro 1995; Han 2002).
Previous data also support an association between physical activity and major
depression disorders and participation in physical activity has been shown to be
associated with positive mood (Williamson 2001), greater psychological well-being
(Calfas 1994; Kirkcaldy 2002), life satisfaction, and cognitive functioning (Stewart
1994). This association has been suggested to work bi-directionally with depression
resulting in low energy thereby leading to reduced physical activity levels (Studer 2001)
and conversely, reduced or lack of physical activity increasing the risk of mental
disorders both cross-sectionally (Goodwin 2003; Brosnahan 2004) and longitudinally
13
(Farmer 1988; Camacho 1991; Lampinen 2000). Taken together, previous findings
overwhelmingly suggest that physical activity is beneficial for physical and mental
health.
.
Coping as a mediator/moderator in the depression-obesity relationship
The widely accepted definition of coping put forward by Lazarus and Folkman
(Lazarus 1984) describes coping as ―continually changing behavioral and cognitive
efforts to manage external and/or internal demands that are appraised as exceeding the
individual‘s resources‖. It has been suggested that the person‘s cognitive appraisal of the
stimulus, and associated emotional reaction, determine the coping methods selected by
the individual (Lazarus 1984). Overall, coping styles have been distinguished into more
direct and adaptive coping modes (i.e., problem-focused, or active coping) and more
indirect and generally maladaptive coping efforts (emotion-focused, or avoidance coping)
(Compas 2001). While active coping involves both behavioral and cognitive efforts to
directly address the stressful problem, avoidance oriented coping attempts to reduce
tension by escapist behaviors and thoughts. Of the two coping domains identified, active
coping seems to represent more mature ‗adaptive‘ coping efforts, which are hypothesized
to be more beneficial to health. They include utilizing social resources for emotional
support or for discussion of problems, engaging in cognitive relaxation activities that
reduce tension, or engaging in physical activity that can improve self-esteem or
productively channel internal conflict (Broder 1995). The second type of coping domain,
avoidance coping, focuses on less mature responses and ‗maladaptive‘ coping efforts to
stress or problems including antisocial activities such as getting angry, causing trouble, or
14
doing something bad, and avoidance activities including the use of substances (cigarettes,
drugs, or alcohol), and ignoring the problem by hanging out with friends or having sex
(Broder 1995).

Relation of maladaptive coping with depression and obesity
Overall in research, maladaptive coping strategies have been shown to be associated
with increased depressive symptoms (Ebata 1994; Chan 1995; Herman-Stahl 1995;
Seiffge-Krenke 2000) and higher levels of avoidant coping have been reported in
depressed adolescents (Seiffge-Krenke 1993; Ebata 1994; Lewinsohn 1994; Chan 1995).
To elucidate further, Chan in 1995 found that adolescents diagnosed as depressed and
anxious were all characterized by frequent use of withdrawal or avoidance, independent
of gender. Others suggest that depressive symptomatology among adolescents is
accompanied by higher levels of passive and avoidant coping but lower levels of active
and approach coping styles (Ebata 1994; Herman-Stahl 1995). Conversely, existing
research has also suggested that avoidant coping styles, such as aggression and
withdrawal, may be considered as a risk factor for developing or increasing depressive
symptoms (Herman-Stahl 1995; Seiffge-Krenke 2000; Murberg 2005). Specifically, in
the study conducted by Herman-Stahl, Stemmler, & Petersen in 1995, it was found that
approach-oriented copers reported the fewest depressive symptoms, whereas avoidant
copers reported the most. Adolescents who changed from using approach oriented coping
to avoidant coping during the first year showed a significant increase in depressive
symptoms, whereas adolescents who switched from avoidant to approach-oriented coping
during the same time period showed a decrease in depressive symptoms.
15
A significant positive relation between maladaptive coping strategies (such as
avoidance and emotion-focused) and eating pathology (especially bulimic pathology) has
also been noted in previous studies (Troop 1994; Yager 1995; Nagata 2000). Specifically,
in terms of body weight, in a study conducted on weight gain in a college population,
women in the body mass index (BMI) gain group were more likely to use maladaptive
coping behaviors (Adams 2007). Additionally, in the field of obesity, maladaptive
component of emotion-focused coping, such as ―wishful thinking‖, has been shown to be
associated with increased obesity (Ryden 2001) and a reduction in emotion-focused
coping has been found to lead to weight loss (Ryden 2003).

Relation of adaptive coping with depression and obesity
In the mental health literature, adaptive coping strategies (such as developing
social skills, social networks, social support and practicing relaxation/calming exercises)
have been effectively utilized in reducing depression among children and adolescents
(Essau 2005) as part of the ‗cognitive‘ and ‗cognitive-behavioral‘ therapies (Clarke 1995;
Wood 1996; Brent 1997; Franklin 1998; Lamb 1998; Merry 2004; Possel 2004). In the
field of obesity prevention, problem-focused coping strategies, such as ―social trust‖ and
―fighting spirit‖, have been used in designing effective intervention programs and large
weight reductions have been associated with significant improvements in these strategies
(Ryden 2001; Ryden 2003).

16
Limitations of existing studies on obesity
1. Identification of other risk factors in obesity prevention
Findings from existing obesity prevention programs have been modest and have not
shown major significant effects on dealing effectively with weight problems in children
and adolescents (Boon 2005; Jakicic 2005; Thomas 2006) and adults (Wing 1999) and
have focused on either dietary restrictions, or physical activity or a combination of both.
Out of the 57 randomized controlled trials reviewed by Thomas in 2006, while one study
showed clinically significant results in terms of improving nutrition (Nicklas 1998),
another study demonstrated clinically significant group differences in decreasing physical
inactivity (Robinson 1999). In terms of both improving nutrition and increasing physical
activity, only 1 study found clinically significant differences on both outcomes (Luepker
1996).
An alternative approach to prevention of overweight and obesity involves the
identification of other potentially modifiable risk factors (Hasler 2005). Researchers have
asked whether there are other novel risk factors in adolescence that make the
development of obesity more likely (McElroy 2004). If identified, such variables might
serve as useful targets for preventive efforts and provide insight into other possible
preventative approaches to overweight. For instance, both cigarette smoking and major
depression are highly treatable. Thus, demonstrating a relation between prior depression
and/or smoking and later body weight might encourage prevention efforts to consider the
benefits of treating smoking and depression for weight-related morbidity and mortality.
These efforts will be further strengthened by the examination of other indicators such as
17
self-rated health and their relation with later obesity risk. As reported above, poor self-
rated health has shown to be a risk factor for obesity (Kaplan 2003).

2. Gender differences in the relations between these risk factors
The roles that these predisposing risk factors serve over developmental periods,
may differ for males and females, which would suggest that at minimum, prevention
strategies address male vs. female differences at risk. Gender differences in cigarette use,
depressive behavior, and physical activity, have been demonstrated in existing studies.
Specifically, in terms of cigarette use, body image and weight concerns have been
demonstrated to be important factors in the initiation of smoking in adolescent females
(Potter 2004), while these factors are less common for males (French 1995) and (Field
2001). Further, in existing studies male and female adolescents have been found to
endorse different reasons for wanting to quit smoking with males citing performance-
related issues (e.g. to improve athletic performance) and females commonly stating
health reasons for wanting to quit smoking (Aung 2003). Similarly, in the adult
population, males and females have been shown to differ in their reasons to smoke with
women smoking for more tension reduction/relaxation, stimulation and social reasons as
compared to men (Berlin 2003).
Gender differences in depressive behavior beginning in adolescence have been
suggested, whereby girls are more likely to be depressed than boys (Allison 2001),
attributed to the changes in hormones, societal expectations and experiences that
accompany puberty and menarche in females (Born 2002). Finally, gender has also been
shown to have an effect on physical activity. Findings from existing studies have
18
suggested that men and women exercise for different reasons, in relation to their eating
and related attitudes, with men exercising for positive goals (e.g., fitness), whereas
women exercising to escape perceived negative consequences (e.g., weight gain) (Penas-
Lledo 2002). These suggest that potential gender differences merit further investigation in
order that future prevention programs are better tailored to meet gender needs.
However, existing studies are limited by their gender focus either on a single
construct (cigarette use, distress, physical activity) or on a specific developmental period.
Research focusing on the existence of gender differences in these relations will have
important public health implications for the development and targeting of gender-specific
obesity prevention strategies.

3. Focus on single relations
Previous studies have focused on single relations, i.e. depression and cigarette use,
depression and obesity, or cigarette use and obesity, and the simultaneous relations
between these constructs has been studied primarily cross-sectionally in adults (Cohen
1991; Istvan 1992; John 2005). Longitudinal studies in adults have looked at the relation
between weight gain, depression, nicotine dependence, and smoking cessation and
relapse (Killen 1996; Borrelli 2002). In the adolescent population, studies have focused
on the simultaneous relation between cigarette use, distress, and weight concern,
specifically body dissatisfaction instead of actual body weight (Stice 2003; Franko 2005).
Additionally, there is little information in the literature on how the relations among these
constructs differ and change according to the developmental stage of an individual.

19
4. Identification of potential mediators or moderators
Research needs to draw attention to other cognitive-behavioral and psychosocial
constructs such as coping styles which may link these complex relations between
smoking, depression and body weight by operating as potential mediators and moderators
in these relationships. As noted above, previous studies have shown that adaptive coping
strategies were linked with better psychological adjustment, whereas maladaptive coping
strategies have been shown to be associated with increased depressive symptoms (Ebata
1994; Chan 1995; Herman-Stahl 1995; Seiffge-Krenke 2000). In terms of body weight,
existing research has shown weight gain to be associated with the use maladaptive coping
behaviors (Adams 2007). Additionally, a reduction in emotion-focused coping (a type of
maladaptive coping) has been found to lead to weight loss and problem-focused coping
strategies (a type of adaptive coping) have been used in designing effective obesity
intervention programs (Ryden 2003).
‗Moderator‘ (tests whether the association between obesity and depression is
conditional upon the presence or criterion level of a third variable) and ‗mediator‘ (tests
whether some third variable causally bridges the association between body weight and
depression) models have been generally underutilized in these studies. To assess the
relative importance of potential moderator and mediator variables, the focus should be on
the ones which are modifiable in the cigarette use-distress-obesity relationship (Faith
2002). Determining whether coping acts as a mediator or moderator in the smoking,
distress, obesity relationships could have important implications for promoting or
decreasing particular types of coping as part of a prevention program. For example, if
adaptive coping acted as a moderator then sub-groups of high and low adaptive coping
20
individuals may be identified for more targeted prevention or a prevention program might
be constrained to one or the other of these groups e.g. a selected or indicated prevention
program. Alternatively, if adaptive coping acted as a mediator then it might be addressed
in a universal prevention program that benefits everyone.

5. Lack of a theoretical framework
A final limitation of exiting studies is the explicit or implicit lack of any theoretical
model as the basis for understanding the cigarette use-obesity and depression-obesity
relations. Theoretical models that conceptualize the causal path between cigarette use,
depression and obesity risk hold great potential and can be explored in future longitudinal
studies. As noted above, this could involve the examination of theoretical constructs such
as coping styles as potential mediators and moderators. Additionally, the etiological
relations between cigarette use, depression and obesity can be understood by focusing on
various types of influences, which link these three constructs. These theories can be
classified as operating at the person-level (S), situation-level (S), and environmental level
(E) (Pentz 1999). Of these, person-level (P) theories, which comprise of intrapersonal
constructs and operate at the individual level might be most instructive towards
explaining the cigarette use-depression-obesity risk epidemiological relations due to their
direct and immediate effect on actual behavior, pertinent with respect to smoking,
depression and obesity. Person-level theories are those in which intrapersonal influences
or characteristics such as cognitions, personality and genetic makeup play a key role in
the performance of a behavior.
21
In the current study, one of the major P-level constructs studied is maladaptive
coping. Even though coping has been theoretically viewed as a
psychosocial/interpersonal construct (social coping theory), however, the items which
maladaptive coping comprises of in this study are intrapersonal in nature. In the present
study, maladaptive coping is comprised of items of negative affect (―I get angry, I say or
do unpleasant things‖), drug use (―I smoke a cigarette/drink alcohol‖), and avoidance (―I
withdraw‖), which can be largely considered to be intrapersonal or individualistic style
of coping. Adaptive coping is the only interpersonal level or situation-level (S) construct
in the current study. In this study, it is comprised of items representing social support (―I
talk to my mother/father‖), physical activity (―I go out and play sports‖), and relaxation
(―I try to calm myself‖), which involve the interaction of the individual with the social
environment.

Overview of current study
The current research is an epidemiological study of the longitudinal relations of
early adolescent cigarette use, to continued cigarette use, subjective health, coping and
psychological distress in emerging adulthood, and finally, to BMI in early adulthood. The
overall objective of this proposal is to firstly, examine the relation between both
psychological distress (composed of anxiety, depression, and somatic symptoms items)
and cigarette use with adult BMI, and secondly, to examine reduced physical activity
both as a consequence of adolescent psychological distress and as a risk factor for
22
subsequent adult distress. These relations will be further evaluated in light of a)
moderating role of gender, b) moderating/mediating role of coping behavior, and c) the
relation between subjective rating of health and BMI.
This study will add to previous research by (1) simultaneously examining the
longitudinal relations of cigarette use, psychological distress, self-rated health, coping,
and physical activity to later BMI, (2) evaluating gender differences in the relations
between these constructs, and (3) assessing the potential mediating and/or moderating
role of coping in these relations.

Proposed hypotheses
The following hypotheses will be tested in the current study:

H1: Cigarette use, Psychological Distress, and BMI: Longitudinally, cigarette use and
psychological distress will have significant relations with early adulthood BMI.
Specifically, it is hypothesized that cigarette use will have a negative association with
BMI in adulthood (i.e. increase in cigarette use will result in decrease in BMI) and
distress will have a positive association with adult BMI (increase in distress will result in
increase in BMI).

H2: Gender as a moderator: Gender will moderate the cigarette use-distress-BMI
relations such that there will be stronger associations of distress and cigarette use with
BMI for females compared to males. Based on findings from existing studies, it is
23
hypothesized that females will have stronger negative cigarette use-BMI and positive
distress-BMI relations as compared to males.

H3: Subjective rating of health: Poor subjective rating of health will have a significant
relation with BMI at early adulthood. Based on existing findings in older adults as
reported above, it is hypothesized that self-rated health will be inversely associated with
adult BMI whereby poor self-rated health will lead to increased BMI in adulthood. One
possible explanation of the negative association between self-rated health and obesity
could be that self-rated health as an indicator is known to reflect underlying diseases and
symptoms and a poor score on this indicator could signify risk to various diseases
including obesity risk.

H4a: Coping as a mediator: Maladaptive or non-productive coping strategies (composed
of drug use, avoidance, and negative affect coping items) at mid-emerging adulthood will
mediate the beginning of emerging adulthood distress-early adulthood BMI relation. It is
hypothesized that distress at the beginning of emerging adulthood will have a positive
relation to maladaptive coping at mid-emerging adulthood (i.e. increase in distress will
lead to increase in maladaptive coping skills) and maladaptive coping will subsequently
have a positive association with adulthood BMI (i.e. increase in maladaptive coping will
result in increased BMI). This positive association between maladaptive coping and
increased BMI could be directed by the fact that maladaptive coping is associated with
poor impulse control and decision-making and this could possibly come into play in
terms of poor food choices thereby increasing obesity risk.
24
H4b: Coping as a moderator: Adaptive or productive coping strategies (composed of
physical activity, support, and relaxation items) will moderate the distress-BMI relation.
It is hypothesized that there be stronger positive associations between psychological
distress and BMI in individuals with low adaptive coping skills as compared to
individuals with high adaptive coping skills.

H5: Psychological distress and Physical activity: Distress at beginning of emerging
adulthood will have a negative path to physical activity at mid-emerging adulthood,
which will then have a significant negative association with distress at early adulthood. It
is hypothesized that increased distress at the beginning of emerging adulthood will lead
to reduced physical activity at mid-emerging adulthood, resulting in increased distress at
early adulthood.

To test these five hypotheses, three studies will be conducted as part of the
dissertation. These three studies will focus on examining the inter-relations beginning in
early adolescence between cigarette use, psychological distress, physical activity, self-
rated health, coping (both adaptive and maladaptive), to BMI in adulthood. An attempt
will also be made to evaluate whether these relations operate differently in males and
females and whether there are any potential mediators or moderators which can be
identified in these relations. Collectively, findings from these three studies will assist in
identifying what constructs possibly function as risk factors to the other constructs
studied, which can then be targeted effectively in future prevention programs, possibly at
earlier developmental periods. Additionally, if the relations between these constructs
25
studied are potentially different in the two genders, then design and development of
gender-specific prevention programs will become imperative. Finally, identification of
which type of coping (adaptive or maladaptive) operates as a mediator or moderator in
these relations will have strong implications for prevention research, i.e., both types of
coping may be considered in prevention, but coping that shows moderator effect may
suggest that sub-groups of adolescents may be identified for more targeted prevention,
whereas the other type of coping might be addressed in a universal prevention program
that benefits everyone.
The next chapter (Chapter 2) describes the sample, research methods, and measures
assessed in the longitudinal study. Next, Chapters 3, 4, and 5 present the research
hypotheses, analytic plans, results and discussions for Studies 1, 2, and 3 respectively.
Finally, Chapter 6 summarizes the overall findings, limitations, and implications for
future research.

26
Chapter 2: Methods

This chapter describes the research and measurement design as well as the measures
utilized in the current study in detail. The analyses conducted for each of the three studies
is discussed in their respective chapters.

Study design
The current study uses data collected as part of a long-term follow-up of a large
drug abuse prevention trial, the Midwestern Prevention Project (MPP), which began in
Kansas City, Missouri in Fall 1984 and a three-year lagged replication in Indianapolis,
Indiana in Fall 1987 (Pentz 1989). The research design of the MPP is a two-group design,
with random assignment of schools to a program or control condition. The measurement
design is longitudinal, with baseline, 6-month follow-up, and annual follow-ups through
high school (12
th
grade), and 18
th
month follow-ups thereafter.
The intervention component of MPP is a comprehensive multi-faceted program
comprising of five components namely, school, parent, community organization, health
policy, and mass media programs for adolescent drug use prevention. Detailed
description of the design and intervention programs are reported elsewhere (Pentz 1989;
Pentz 1997).

27
Sample
Beginning with the original panel of 1606 participants in Kansas City and 3413 in
Indianapolis, planned sub-sampling over multiple grant periods resulted in 425 and 717
subjects in Kansas City and Indianapolis respectively, in the current 2003-05 wave (total
N=1142). The population was representative of the Kansas City and Indianapolis
populations at the time that study was initiated (i.e. 1984 in Kansas City and 1987 in
Indianapolis). Currently, sixty percent of this sample is female. Based on self-reported
ethnicity, current sample is currently 83% White, between 27-34 years, and fifty four
percent are in the low SES strata. Each of the three studies conducted for the dissertation
included a specific sample size. Study 1 was conducted on the Kansas City population
and was restricted further to only those subjects who had complete BMI data (N=414).
Analyses for Study 2 was conducted by ‗stacking‘ the early adulthood data (mean
age=30.06 years) for Kansas City and Indianapolis and restricting it to those subjects who
had complete BMI data (N=1139). In the current study, ‗stacking‘ refers to combining
waves of same grade/age cohort data corresponding to each developmental period studied
for each site (Crawford 2003). Study 3 was also conducted on the ‗stacked‘ data for both
sites, however, unlike Study 2 it was not restricted to only those participants who had
complete BMI data, thereby resulting in a total sample of 5019 subjects.
To evaluate changes in relations between these constructs over time and also based
on when these constructs were measured, four waves of data were selected for the
current study, which represent four distinct developmental periods, namely early
adolescence, beginning of emerging adulthood, mid-emerging adulthood, and early
adulthood (Pentz 2006).Retention rates for the early adolescence period were 100% for
28
Kansas City and 100% for Indianapolis, 77.9% and 69.8% for beginning of emerging
adulthood, 68.6% and 74.9% for mid-emerging adulthood, and 83.4% and 64.2% for the
early adulthood period for Kansas City and Indianapolis, respectively.

Measures
Measures were administered in schools during the early adolescent period and a
combination of mail, phone and web surveys were used thereafter for the remaining three
developmental periods. A random item-sampling strategy was used to incorporate a
larger set of items than would be possible with one form (Graham 1984). Participants
were randomly assigned within classroom and school to one of the four forms. Two-
thirds of each form had the same drug use and demographic items. One of the four forms
included the psychological distress, coping, subjective rating of health, physical activity,
and BMI items which were used in the current study. Previous analyses showed no
differences in experimental group or demographic characteristics of participants who
received any particular form.
Since the MPP was designed as a drug abuse prevention study, survey items
concentrated on drug abuse, including cigarette use. As participants moved from early to
mid-adolescence, items reflecting early adolescent relations were replaced on some forms
of the survey with items relevant to mid-late adolescent smoking, including psychological
distress and coping. By emerging adulthood, smoking or lack of it may be part of a larger
developing healthy lifestyle and the inclusion of subjective health, physical activity and
finally, BMI items reflect this.
29
A table illustrating the developmental periods/waves when each construct was
studied in the current study is provided at the end of the methods section. The cronbach
alpha‘s reported for each construct below are based on the early adulthood stacked data
for both sites and restricting to only those subjects who had complete BMI data.
In the present study, cigarette use items were used at multiple waves (early
adolescent, beginning of emerging adulthood, and mid-emerging adulthood). In the early
adolescent period, cigarette use items measured lifetime (1=one puff-6=more than 5
packs), monthly (1=none-6=more than 1 pack), weekly (1=none-6=more than 1 pack),
and current usage (1=used to smoke but don‘t-7=a pack a day or more)?‖ (Pentz 1989)
and (Johnston 2003). Cigarette use items in both the beginning and mid-emerging
adulthood waves measured lifetime, monthly, weekly, and past 24 hours and the range for
all these items (except for the lifetime use item) was the same (1=none-7=more than 10
packs). The Cronbach alpha for the cigarette use items was .87 in early adolescence, .95
in the beginning of emerging adulthood period, and .95 in the mid-emerging adulthood
period.
Psychological distress was used at the beginning of emerging adulthood (i.e.
when it was first measured in the MPP study) and at the mid-emerging adulthood wave in
the current study. Items measuring distress were originally derived from the Hopkins
Symptom Checklist (Derogatis 1973; Hansell 1991) and represent a 13-item scale, which
include distinct constructs of anxiety (4 items; ―I feel nervous or anxious, I feel I have
nothing to look forward to, I think seriously of a way to hurt myself, I am no good for
anything at all (1=almost never-4=almost always)‖; Cronbach alpha: .70), depression (6
items; ―I feel I don‘t want to do anything, I worry about things in my life, I am tired
30
during the day, I feel lonely, I feel sad, I find it hard to keep my mind (1=almost never-
4=almost always)‖; Cronbach alpha: .77), and somatic symptoms (3 items; ―I feel dizzy
or lighthearted, I get headaches, I have pains in my heart or chest (1=almost never-
4=almost always)‖; Cronbach alpha:.70). The Cronbach alpha for the entire distress scale
was .84.
Coping was used at the mid-emerging adulthood wave and was measured with 24
items from the Wills Coping Inventory (Wills 1996) (Cronbach alpha=.89). Factor
analysis on the entire coping construct resulted in two distinct sub-constructs of adaptive
coping (7 items; ―I work it off with physical exercise, I go out and play sports, I get
information that is needed to solve the problem, I talk with my mother or father, I talk
with one of my friends, I try to calm myself, I try deep breathing (1=almost never-
4=almost always)‖; (Cronbach alpha: .81) and maladaptive coping (10 items; ―I tell
myself that the problem is not worth getting upset about, I wait and hope things will get
better with time, I try not to think about it, I withdraw, I get mad at people, I sleep more, I
get away from people, I say or do unpleasant things, I smoke a cigarette to relax, I drink
alcohol/drugs to feel better (1=almost never-4=almost always)‖ (Cronbach alpha:.86).
Physical activity (CDC 2005) items were added to surveys in 1993-94 in Kansas
City and 1994-95 in Indianapolis. In the current study, it was used only at the mid-
emerging adulthood developmental period. The following 2 items were used in the
current analyses ―How many days a week do you exercise (1=0 days on most weeks-6=5
days a week)?‖; ―How many times a week do you exercise so hard that your heart
breathes fast or you breathe hard (1=I don‘t exercise at all-8=6 or more times a week)?‖
(Cronbach alpha: 90).
31
Subjective rating of health (SAMHSA 1997,1998) was used only at the mid-
emerging adulthood wave and was measured by 1 item ―Compared to other people your
age, how would you rate your health (1=poor, 2=average/fair, 3=good, 4=very good,
5=excellent)?‖.
Body mass index (BMI), the primary outcome, was measured at only the early
adulthood wave and was used as a continuous variable in the current analyses. Self-
reported weight (in pounds) and height (both in feet and inches) were converted first into
kilograms (weight) and meters (height) and then used for computation of BMI (weight in
kilograms/(height in meters)
2
(CDC 2005).

Table 1: List of measures used at each developmental period
Early adolescence
(Mean=12.06 yrs)
Beginning of emerging
adulthood
(Mean=19.47 yrs)
Mid-emerging adulthood
(Mean=23.22 yrs)
Early adulthood
(Mean=30.06 yrs)
Cigarette use Cigarette use Cigarette use Cigarette use
Psych. distress Psych. distress Psych. distress
Physical activity
Coping
Subj rating of health
BMI

As mentioned in the beginning of this chapter, the data analyses conducted for each
of the three studies is discussed in their respective chapters (Chapter 4, 5 and 6).

32
Chapter 3: Study 1

Introduction

Cigarette smoking and obesity are major health threats to society. While smoking
is a proven cause of respiratory and cardiovascular disorders, as well as a wide range of
cancers (Boyle 1997; Bartal 2001; USDHHS 2004), co-morbidities associated with
obesity include elevated blood pressure, dyslipidemia, insulin resistance, and type 2
diabetes (Deckelbaum 2001). Despite the detrimental health effects associated with both
smoking and obesity, trends show that these behaviors generally start and increase during
the adolescent years (Escobedo 1990; Warren 1997; Gilpin 1999; Kimm 2002; Kaur
2003). Further, both these health behaviors share common developmental triggers,
including early pubertal changes (such as increased breast size, age of menarche, body
curviness in girls and age of peak height velocity in boys), which have been found to be
an important risk factor for later cigarette use (Lanza 2002; Chung 2005) and obesity
(Van Lenthe 1996; Power 1997; Biro 2001).
An extensive literature has demonstrated links between smoking and body weight,
specifically weight concern, in adolescents. Existing studies have shown that concern and
dissatisfaction with body weight are important factors in initiation of tobacco use,
particularly among adolescent females (French 1995; Grigg 1996; Potter 2004). Findings
from existing studies support a strong positive association between smoking and weight
concern in adolescent females (Potter 2004). Two hypotheses have been proposed to
explain this smoking-weight concern relationship in adolescence (French 1995; French
1996). First, in response to societal pressure to be thin (Wiseman 1998), adolescent
33
females may initiate smoking as a weight control measure (Tomeo 1999). Secondly,
according to the problem behavior theory (Jessor 1991), experimentation with smoking
and engaging in unhealthy dietary practices, such as eating junk food and abstaining
completely from eating can be thought of as part of clustering of risk factors during
adolescence (Burke 1997; Escobedo 1997; DuRant 1999). These youth risk behaviors
cluster because these behaviors serve similar functions for adolescents such as
independence from adults, rebelliousness, etc. (French 1996). From a problem-behavior
perspective, adolescent cigarette smoking as a normative transgression could be
motivated by goals such as rejecting the norms of conventional society, affirming
membership in a peer group, asserting independence from parents, or being seen as more
mature (Turbin 2000). Specifically, smoking is legal only for adults and hence, initiation
and experimentation during this period may be a convenient way for adolescents inclined
towards deviance or rebellion to engage in risk-taking or anti-social behavior (Simons-
Morton 1999). Dietary practices, such as abstinence from food and skipping meals, have
also been shown to represent as asserting one‘s own will above their parents, testing
personal boundaries and challenging authority (Spruijt-Metz 1995).
Compared to the extensive number of studies examining weight concern-smoking
association, relatively few studies have examined the relation between actual body weight
and smoking in adolescents. While a positive association between smoking and body
weight for both genders has been suggested both cross-sectionally (Townsend 1991;
Freedman 1997; Robinson 1997; Klesges 1998) and longitudinally (Cooper 2003), a non-
significant association between these behaviors has also been found (Page 1993; Crawley
1995; Killen 1997; Crisp 1998; Crocker 2001; Strauss 2001). On the other hand, in
34
adults, body weight and smoking have been shown to be inversely associated in both
genders (Lissner 1992; Klesges 1998; Bamia 2004). Thus, there is scant and inconclusive
information about the smoking-weight relation across different populations.
Besides obesity, previous studies have also shown cigarette use to be associated
with other constructs such as psychological distress, physical activity, and self-rated
health, and individually each of these has been found to be associated with obesity. While
linkages between cigarette use and distress, particularly depression, are well established
in adolescence (Windle 2001; Fergusson 2003; Brook 2004; Duncan 2005) and adulthood
(Ismail 2000; Breslau 2001), there is also emerging evidence to suggest that depression
may be implicated in the development of later obesity, especially in females (Richardson
2003; Franko 2005; Hasler 2005; Stice 2005). Evidence from limited studies has shown
cigarette use to contribute to physical impairments over time (Hansell 1991) and, in
particular, have a direct effect on poor self-rated health (Vingilis 1998). Further, both
physical inactivity and poor health perceptions have been suggested to predict later
obesity risk (Ball 2001; Kaplan 2003). Certain demographic factors have also been
identified as risk factors for body mass index (BMI) including socio-economic status
(SES) and race/ethnicity both in adolescents (Storey 2003; Forshee 2004) as well as in
adults (Jeffery 1991; Seo 2006).
The course of cigarette use has been shown to vary across periods of development
and be driven by different factors. Specifically, cigarette use has been shown to increase
sharply in early adolescence followed by a maturational pattern resulting in a flattening
or even decline by the 20‘s, the period referred to as emerging adulthood (ages 19-24)
(Kandel 1992; Chen 1995). Cigarette use in early adolescence may represent
35
experimentation compared to a means to cope with anxiety in adulthood. Relatively little
is known about how these developmental benchmarks might operate in changing the
relations between cigarette use and other constructs such as distress, and obesity.
These previous studies have several limitations. First, most of these smoking-
weight studies are limited in the cross-sectional nature of their design and temporal
relations (if not causal) cannot be inferred from their findings. Another limitation is that
previous studies have focused on single relations, i.e. depression and cigarette use,
depression and obesity, or cigarette use and obesity, and the simultaneous relations
between these behaviors has only been studied cross-sectionally in adults (Cohen 1991;
Istvan 1992). Finally, there is little information in the literature on how the relations
between cigarette use and other health behaviors differ and change according to the
developmental stage of an individual.
The purpose of the present study was to examine simultaneous longitudinal
relations between cigarette use in early adolescence to continuing cigarette use,
psychological distress, physical activity, and subjective rating of health in emerging
adulthood, and finally, to body mass index (BMI) in early adulthood in Kansas City. In
this study, it is hypothesized that cigarette at early adolescence would predict both
cigarette use and psychological distress at the beginning of emerging adulthood. Further,
cigarette use and distress at the beginning of emerging adulthood would predict cigarette
use, physical activity, and subjective rating of health at mid-emerging adulthood, which
in turn would predict body mass index (BMI) at early adulthood. To evaluate changes in
relations between these constructs over time and also based on when these constructs
were measured, four waves of data were selected for the current study, which represent
36
four distinct developmental periods, namely early adolescence (11-13 years), beginning
of emerging adulthood (17-20 years), mid-emerging adulthood (22-24 years), and early
adulthood (30-34 years).

Methods
Study background
The current study uses data collected as part of a long-term follow-up of a large
drug abuse prevention trial, the Midwestern Prevention Project (MPP), which began in
Kansas City, Missouri in Fall 1984 (Pentz 1989). Data were drawn from a panel of
students who represented 96% of entering class of middle (6
th
) or junior high (7
th
) in
eight schools in Kansas City. These 8 schools were randomly drawn from a larger pool of
50 based on identified school feeder patterns between middle and high school. The
research design of MPP is a two-group design, with random assignment of schools to a
program or control condition. Measurement design is longitudinal, with baseline, 6-
month follow-up, and annual follow-ups through high school (12
th
grade), and 18
th
month
follow-ups thereafter. The MPP study includes 16 waves of data collection across four
theoretically distinct developmental periods (Junior High, High School, Emerging
Adulthood, and Early Adulthood) representing 23 years, from ages 11 through 34 (Pentz
et al., 2006). The intervention component of the MPP is a comprehensive multi-faceted
program comprising of five components namely, school, parent, community organization,
health policy, and mass media programs for adolescent drug use prevention. Detailed
description of the design and intervention programs are reported elsewhere (Pentz 1993).

37
Sample and sample quality
Details of this sample are reported previously (Pentz 1989). Sampling for the panel
changed three times in this study. First, from the original panel of 1606 Kansas City
participants, a sub-sample of 1002 were randomly selected to be followed into adulthood
due to cost of follow-up considerations, since subjects were spread all over the world by
that point. The specific sample size of 1002 was selected based on power calculations to
detect a sustained effect of at least .20 with .80 power. Second in 1996, this sample was
expanded to include additional subjects from the original panel, thereby raising the
eligible sample to 1167 in Kansas City. Third, in 2000, with a change in data collection
procedures and organization, only 565 subjects could be located. Previous analyses
showed no differences in demographic or drug use characteristics at baseline in 1984-85
between the 565 subjects who could be located and the other 602 who could not be
contacted. However, for the 1999-2000 wave (i.e. before the change in data collection
procedures and organization occurred), there were significantly more males and weekly
smokers in the sample that could not be located (N=602) as compared to the sample that
was located (N=565). Of these 565 subjects, 140 could not be surveyed due to the closure
of the measurement period and need to be measured at a future time point.
Thus, in the current wave (2003-05), subjects are a total of 425 adults in Kansas
City ranging between 30-34 years of age. Sixty-four percent of this sample is currently
female. Based on self-reported ethnicity, sample is currently 85% White and
approximately 8% African-American. Forty seven percent of the current sample is of low
SES. For the current analyses, the sample was restricted to only those participants who
38
had complete data on BMI, measured only in the early adulthood wave. Of the 425
subjects at this wave, these subjects comprise a total of 414.
In terms of the sample quality, the attrition pattern as well as the bias associated
with this attrition has been examined by Fan and colleagues (Fan 2002). In this study, it
was found that cigarette and marijuana users were less likely to be retained over time
though there was no significant interaction between substance users and intervention
condition.

Measures
Measures were administered in schools during the early adolescent period and a
combination of mail, phone and web surveys were used thereafter for the remaining three
developmental periods. For all the variables used in the analyses (except the
demographic, height and weight variables, which were used in the computation of BMI)
non-users or those respondents who responded "never" or "not at all" etc., were coded as
0 in the dataset.
Cigarette use was the only construct used at multiple waves (early adolescent,
beginning of emerging adulthood, and mid-emerging adulthood) in the present study. In
the early adolescent period, cigarette use items measured lifetime (1=one puff-6=more
than 5 packs), monthly (1=none-6=more than 1 pack), weekly (1=none-6=more than 1
pack), and current usage (1=used to smoke but don‘t-7=a pack a day or more)?‖ (Pentz
1989; Johnston 2003). Cigarette use items in both the beginning and mid-emerging
adulthood waves measured lifetime, monthly, weekly, and past 24 hours and the range for
all these items was the same (1=none-7=more than 10 packs). However, the lifetime
39
cigarette use item from the mid-emerging adulthood period was not included in the
current analyses as it was a dichotomous variable. The other cigarette use variables in all
the waves studied were categorical. The Cronbach alpha for the cigarette use items was
.94 in early adolescence, .96 in the beginning of emerging adulthood period, and .97 in
the mid-emerging adulthood period.
Psychological distress was used at the beginning of emerging adulthood wave in
the current study. Items measuring distress were originally derived from the Hopkins
Symptom Checklist (Derogatis 1973; Hansell 1991) and represent a 13-item scale, which
include distinct constructs of anxiety (4 items; ―I feel nervous or anxious, I feel I have
nothing to look forward to, I think seriously of a way to hurt myself, I am no good for
anything at all (1=almost never-4=almost always)‖; Cronbach alpha: .73), depression (6
items; ―I feel I don‘t want to do anything, I worry about things in my life, I am tired
during the day, I feel lonely, I feel sad, I find it hard to keep my mind (1=almost never-
4=almost always)‖; Cronbach alpha: .80), and somatic symptoms (3 items; ―I feel dizzy
or lighthearted, I get headaches, I have pains in my heart or chest (1=almost never-
4=almost always)‖; Cronbach alpha:.68). The Cronbach alpha for the entire distress scale
was .87.
Physical activity (CDC 2005) was used only at the mid-emerging adulthood wave
and was measured by 3 items at this wave ―How many days a week do you exercise (1=0
days on most weeks-6=5 days a week)?‖; ―How hard would it be for you to run 20
minutes without stopping (1=impossible-5=very easy)?‖; ―How many times a week do
you exercise so hard that your heart breathes fast or you breathe hard (1=I don‘t exercise
at all-8=more than once a day)?‖ (Cronbach alpha: 82).
40
Subjective rating of health (SAMHSA 1997,1998) was used only at the mid-
emerging adulthood wave and was measured by 1 item ―Compared to other people your
age, how would you rate your health (1=poor, 2=average/fair, 3=good, 4=very good,
5=excellent)?‖.
Body mass index (BMI), the primary outcome, was measured at only the early
adulthood wave and was used as a continuous variable in the current analyses. Self-
reported weight (in pounds) and height (both in feet and inches) were converted first into
kilograms (weight) and meters (height) and then used for computation of BMI (weight in
kilograms/(height in meters)
2
(CDC 2005).
Five demographic variables were used as covariates in the analyses: a. gender,
coded 1=male, 2=female; b. baseline ethnicity, coded 1=white, 2=non-white; c. baseline
age; d. baseline socio-economic (SES) status which was measured as mother or father in
a professional occupation, coded 0=low SES, 1=high SES; and e. current SES status,
coded 0=low SES, 1=high SES. Intervention condition was coded 0=control, 1=program.
To minimize loss of cases due to missing SES, an imputation strategy in SAS (SAS,
1999-2001) was employed whereby if the current SES was missing then the SES from the
previous wave was used to replace the missing value.

Data Analyses
Data analysis was conducted in three steps. Using SAS, exploratory factor analysis
(EFA) was conducted on the items measuring cigarette use, psychological distress, and
physical activity. An individual item was retained if it had a factor loading higher than
.50. As expected, all items representing cigarette use in each of the three waves loaded on
41
one factor. Similarly, the three items measuring physical activity also loaded on one
factor. Items measuring cigarette use and physical activity were treated as single
indicators in the analyses. EFA on the psychological distress items resulted in three
separate factors, namely depression, anxiety, and somatic symptoms. Thus, three
composite scores constituting depression, anxiety, and somatic symptoms were created
for the distress construct. Secondly, the factors and variables obtained from EFA were
subjected to confirmatory factor analysis (CFA) to yield a measurement model. In the
CFA model, correlational paths between all the constructs were included as free
parameters while covarying for baseline age, baseline ethnicity, baseline SES, gender,
intervention condition, and current SES.
In the third step of data analysis, a structural model was generated from the
measurement model. Compared to the CFA model, the structural model was more
specifically defined for testing wave-to-wave relationships and hypothesis by including
directional paths between these constructs while controlling for the covariates. To
capitalize on all available data points, AMOS 5.0 (Arbuckle 2003) was used for
estimation of the models, with full information maximum-likelihood (FIML) missing
data imputation. According to (Wothke 1998), ―The FIML method uses all of the
information of the observed data, including mean and variance for the missing portions of
a variable, given the observed portion(s) of other variables‖. To improve model fit for
the two models, cross-sectional correlation matrices between the variables at each wave
studied were generated in SAS (SAS 1999-2001). Both theoretical meaningfulness and
significant correlations at each wave were considered to be included as correlated errors
42
between those indicators in the AMOS models. Based on the output generated in AMOS,
non-significant correlated error terms were deleted to obtain more parsimonious models.

Results
Sample representativeness and descriptive statistics
Table 2 illustrates descriptive statistics as well as means on constructs used and
tested in each developmental period leading up to BMI in the present study for the
sample without BMI (n = 1193) and with BMI (n = 413). As is evident, there were more
females, whites, and higher psychological distress scores in the sample with BMI.
However, those without BMI had significantly higher scores on cigarette use at baseline
than those with BMI.
Bivariate correlations and summary descriptive statistics for the complete set of
measures used in modeling are shown in Table 3.

43
Table 2: Equivalence of sample without BMI vs. with BMI

Sample without BMI Sample with BMI
N=1193 N=413
% (SE) % (SE)
Intervention 56.3 (1.44) 55.9 (2.44)
White 80.3 (1.15) 88.4 (1.57)**
Female 45.2 ( 1.44) 64.2 (2.36)***
Low SES 61.2 (1.41) 53.9 (2.45)
Mean (SD) Mean (SD)
Early Adolescence (age:11-13 yrs)
Cigarette use 3.67 (4.61) 2.71 (.4.18)**
Beg. of Emerging Adulthood (age:17-20 yrs)
Cigarette use 9.09 (7.96) 8.65 (7.79)
Psychological distress 20.53 (5.75) 22.21 (5.59)*
Mid-Emerging Adulthood (age 22-24 yrs)
Cigarette use 6.91 (5.52) 6.40 (5.47)
Physical activity 10.58 (4.68) 10.65 (4.35)
Subjective rating of health 3.25 (1.22) 3.39 (1.05)
Mid-Emerging Adulthood: Cigarette use= composite of 3 items (monthly, weekly, past 24 hours; range:0-19); Physical activity=composite of 3 items
(range:0-19); Subjective rating of health=1 item (range:0-5)
PROC TTEST in SAS was used to generate means, standard deviations, and significance levels for the continuous variables.
Early Adolescence: Cigarette use=composite of 4 items (lifetime, monthly, weekly, present smoking; range:0-25);
composite of 13 items (range: 0-40);
Baseline variables were used for computation of intervention, white, female, and low SES proportions.
Chi-square in SAS was used to generate proportions, and signficance levels for dichotomous variables
Note: +p <.10;*p <.05;**p <.01;***p <.001
Beginning of Emerging Adulthood: Cigarette use=composite of 4 items (lifetime, monthly, weekly, past 24 hours; range:0-28); Psychological distress=
Sample without BMI=Subjects present both at baseline and having a missing BMI value
Sample with BMI=Subjects present both at baseline and having a non-missing BMI value

44
Table 3: Summary descriptive statistics and correlations

Mean, Standard Deviation, and variable 1 2 3 4 5 6 7 8 9
Mean 2.70 8.65 12.16 5.56 4.49 6.40 10.65 3.39 26.15
Standard Deviation 4.18 7.79 3.22 1.82 1.80 5.47 4.35 .97 5.64
Early Adolescence (age: 11-13 yrs)
1. Cigarette use .40* .07 .09 .02 .37* -.14* -.03 -.005
Beg. of Emerging Adulthood (age: 17-20 yrs)
2. Cigarette use .40* .08 .15 .16 .69* -.06 -.05 -.12*
3. Depression .07 .08 .61* .37* -.05 -.16 -.10 .01
4. Anxiety .09 .15 .61* .42* .20* -.19 -.27* .08
5. Somatic symptoms .02 .16 .37* .42* .14 -.24* -.18 -.03
Mid-Emerging Adulthood (age:22-24 yrs)
6. Cigarette use .37* .69* -.05 .20* .14 -.09 -.08 -.06
7. Physical activity -.15* -.06 -.16 -.19 -.24* -.08 .47* -.22*
8. Subjective rating of health -.03 -.05 -.10 -.27* -.18 -.08 .47* -.19*
Early Adulthood (age: 30-34 yrs)
9. BMI -.005 -.12* .01 .08 -.03 -.06 -.22* -.19*
Physical activity=composite of 3 items (range:0-19); Subj. rating of health=1 item (range:0-5).
Depression=composite of 6 items (range:0-21); Anxiety= composite of 4 items (range:0-11); Somatic symptoms=composite of 3 items (range:0-12).
Note: *p < .05; Early Adolescence : Cigarette use= composite of 4 items (lifetime, monthly, weekly, current use; range=0-25).
Beginning of Emerging Adulthood: Cigarette use=composite of of 4 items (lifetime, monthly weekly, past 24 hours use; range=0-28);
Mid-emerging adulthood: Cigarette use=composite score of 3 items (monthly, weekly, past 24 hours; range:0-19);
Early Adulthood : BMI (range:16.6-48.65).

Measurement model
The measurement model (Figure 1) demonstrates the results of the CFA model
( 
2
=396.933, df=212, NFI=.933, CFI=.967, RMSEA=.046). There were no significant
covariances between cigarette use in early adolescence to either subjective rating of
health in mid-emerging adulthood or to BMI in the early adulthood period. Further, there
was no association between cigarette use in the beginning of emerging adulthood and
subjective rating of health in mid-emerging adulthood. Finally, psychological distress in
beginning of emerging adulthood period was not significantly related to BMI in early
adulthood.
Additionally, results from the CFA model indicated that all the three composite
indicators of depression (.507), anxiety (.620), and somatic symptoms (.714) should load
on an overall psychological distress construct. Further, there were significantly moderate
correlations between each of these composite scores (depression-anxiety: r= .61***;
depression-somatic symptoms: r= .37***; anxiety-somatic symptoms: r= .42***).
45
However, since these three indicators have been discussed separately in the research
literature, separate structural models were generated with each indicator at the beginning
of emerging adulthood wave. Based on a one-tailed test, results suggested that depression
had no significant relation with cigarette use, physical activity, and subjective rating of
health at mid-emerging adulthood. On the other hand, anxiety indicator at the beginning
of emerging adulthood had significant relations with cigarette use, physical activity, and
self-rated health at mid-emerging adulthood (p<.05). The composite indicator of somatic
symptoms had a marginal significant relation with physical activity and self-rated health
(p<.10) and a non-significant relation with cigarette use at mid-emerging adulthood.

Figure 1: Measurement model (Kansas City)

Early Adulthood
(Age: 30-34 yrs)
Mid-Emerging adulthood
(Age:22-24 yrs)
Early Adolescence
(Age: 11-13 yrs)
Beginning of Emerging Adulthood
(Age: 17-20 yrs)
Model fit:
NFI=.933; CFI=.967
c2 (df)=396.933 (212)
RMSEA=.046
N=414
+p<.10; *p<.05;**p<.01;***p<.001 (one-tailed)
Model adjusted for baseline age, ethnicity, gender, intervention condition, baseline SES, and current SES.
BMI
Subj. rating
of health
Psych.
distress
Phy.
activity
Cig. use
Cig. use Cig. use .461***
.786***
-.277*
-.336***
-.142*
.637***
.294*
.450***
-.125+
-.327*
-.161*
-.181**
-.095+
-.177*
.342**
-.092+
.188 *
Early Adulthood
(Age: 30-34 yrs)
Mid-Emerging adulthood
(Age:22-24 yrs)
Early Adolescence
(Age: 11-13 yrs)
Beginning of Emerging Adulthood
(Age: 17-20 yrs)
Model fit:
NFI=.933; CFI=.967
c2 (df)=396.933 (212)
RMSEA=.046
N=414
+p<.10; *p<.05;**p<.01;***p<.001 (one-tailed)
Model adjusted for baseline age, ethnicity, gender, intervention condition, baseline SES, and current SES.
BMI
Subj. rating
of health
Psych.
distress
Phy.
activity
Cig. use
Cig. use Cig. use .461***
.786***
-.277*
-.336***
-.142*
.637***
.294*
.450***
-.125+
-.327*
-.161*
-.181**
-.095+
-.177*
.342**
-.092+
.188 *
46
Structural model
Figure 2 ( 
2
=413.402, df=218, NFI=.930, CFI=.965, RMSEA=.047) demonstrates
the results of the structural model on the constructs of interest. As indicated by the model,
only cigarette use, and physical activity in the mid-emerging adulthood period have a
significant relation to the primary outcome, BMI. There is no significant path from
subjective rating of health and BMI. As hypothesized, there are significant paths between
psychological distress in the beginning of emerging adulthood period to cigarette use,
physical activity, and subjective rating of health present in the mid-emerging adulthood
period. However, there are no significant paths between cigarette use in the beginning of
emerging adulthood to either physical activity or subjective rating of health in the mid-
emerging adulthood. Additionally, the hypothesized correlational paths between cigarette
use and physical activity and cigarette use and subjective rating of health in the mid-
emerging adulthood period are also not significant. Further, cigarette use in early
adolescence is significantly related to both cigarette use and psychological distress in the
beginning of emerging adulthood. There is also a very strong correlation between
physical activity and subjective rating of health in the mid-emerging adulthood period.
Only the significant paths based on one-tailed tests have been shown in both
Figures 1 and 2.

47
Figure 2: Structural model (Kansas City)

Discussion
The basic objective of this study was to examine the simultaneous longitudinal
relations between cigarette use in early adolescence to continuing cigarette use,
psychological distress, physical activity, subjective rating of health, and BMI in
adulthood. The structural equation model (SEM) generated for this study suggests that
adolescent cigarette use contributes to continuing cigarette use, high distress, and reduced
BMI in adulthood. However, there is no direct relation between cigarette use and physical
activity or cigarette use and subjective rating of health. Indirectly, cigarette use in
adolescence leads to high psychological distress which in turn predicts cigarette use, loss
of physical activity, and poor self-rated health in mid-emerging adulthood. This lack of
Early Adulthood
(Age:30-34 yrs)
BMI
Mid-Emerging adulthood
(Age:22-24 yrs)
Subj. rating
of health
Psych.
distress
Phy.
activity
Early Adolescence
(Age:11-13 yrs)
Beginning of
Emerging Adulthood
(Age:17-20 yrs)
Cig. use
Cig. use Cig. use
Model fit:
NFI=.930; CFI=.965
2 (df)=413.402 (218)
RMSEA=.047
N=414
.476***
.736***
+p<.10; *p<.05;**p<.01;***p<.001 (one-tailed)
Model adjusted for baseline age, ethnicity, gender, intervention condition, baseline SES, and current SES.
-.369*
-.281*
-.439***
-.152**
.593***
.214+
.265*
.190*
Early Adulthood
(Age:30-34 yrs)
BMI
Mid-Emerging adulthood
(Age:22-24 yrs)
Subj. rating
of health
Psych.
distress
Phy.
activity
Early Adolescence
(Age:11-13 yrs)
Beginning of
Emerging Adulthood
(Age:17-20 yrs)
Cig. use
Cig. use Cig. use
Model fit:
NFI=.930; CFI=.965
2 (df)=413.402 (218)
RMSEA=.047
N=414
.476***
.736***
+p<.10; *p<.05;**p<.01;***p<.001 (one-tailed)
Model adjusted for baseline age, ethnicity, gender, intervention condition, baseline SES, and current SES.
-.369*
-.281*
-.439***
-.152**
.593***
.214+
.265*
.190*
48
physical activity was associated with increased BMI in early adulthood. Pessimistic self-
rated health perceptions in the current study were found not to signal later obesity risk in
adulthood.
Though gender differences in the relations between these constructs was not the
focus of the current study, however, gender was included as a covariate in both the CFA
and SEM models. Based on a two-tailed test, gender was a significant predictor of
cigarette use (β=-.116, p=.010) and physical activity (β=-.216, p=.009) in the mid-
emerging adulthood wave, and BMI (β=-.202, p<.0001) in the early adulthood period in
the SEM model. These results suggest that females have higher cigarette use, lower
physical activity levels, and lower BMI as compared to males. Existing studies have
shown males and females to differ on reasons for smoking, physical activity, and
depression (Ge 2001; Born 2002; Penas-Lledo 2002; Berlin 2003). Future research will
further investigate structural relations differences in the overall model due to gender.
The findings of the present study should be considered relative to results of
previous studies. First, the intention of this study was to extend beyond previous limited
longitudinal research that had focused on the relationship between smoking and actual
body weight by evaluating the effect of cigarette use measured across multiple
developmental periods to BMI in adulthood. Results from the current study demonstrate
that cigarette use beginning in early adolescence and continuing into mid-emerging
adulthood had an inverse relationship with BMI in early adulthood. This finding supports
previous research conducted on adults which have found smokers to have lower BMI
than non-smokers (Lissner 1992; Klesges 1998; Bamia 2004). Second, the present study
intended to evaluate the relations between cigarette use and other constructs, including
49
distress, physical activity, and subjective rating of health simultaneously over time with
adult BMI. In terms of these relations, cigarette use was shown to have a synergistic
relation with psychological distress over time. In the current study early cigarette use,
which is often thought to be experimental and socially driven in early adolescence,
predicted later psychological distress, which in turn was related to increased cigarette use
in adulthood. Previous research has shown both smoking (daily and heavy) and nicotine
dependence and substance use disorders in general to be co-morbid with psychiatric
disorders such as major depression and anxiety (Breslau 1995; Kandel 1997; Kandel
1999; Johnson 2004) among adolescents and young adults.

The present study differs from previous findings in that cigarette use neither
predicted physical activity nor had a significant cross-sectional correlation with physical
activity. In the existing literature, smokers have been shown to exhibit lower levels of
physical activity as compared to non-smokers (Klesges 1992). Rather in the current
study, cigarette use had an indirect effect on physical activity mediated through changes
in distress. Further, there was no significant relation between subjective rating of health
and BMI in adulthood. Though past research has shown poor self-rated perceptions to
serve as early indicators for later obesity risk, this association has been found mainly in
older adults (Kaplan 2003). Overall, however, the findings of this study support previous
studies, and have important implications for designing future prevention programs that go
beyond cigarette use and target multiple risk behaviors.
The current results must be considered in light of study limitations. The first
limitation is that measurement of BMI is limited only to the early adulthood period since
50
drug abuse prevention rather than obesity was the focus of the original parent MPP study.
In the literature, overweight in childhood/adolescence has been shown to be a strong
predictor of overweight/obesity in adulthood (Guo 2000). Hence, adolescent overweight
status cannot be used as a covariate in analyses due to limitations of the MPP dataset.
A second limitation concerning the comprehensiveness of measures is that a single
item was used to measure subjective rating of health. Respondents were asked to how
good or bad their health had been on a 5-point scale from poor to excellent. Though the
use of a single-item question measuring self-rated health has warranted caution however,
self-rated health has also been demonstrated to be a valid measure for measuring health
status in different ethnic cultures (Chandola 2000; Agyemang 2006) and in adolescents
(Geckova 2001) as well as adults (Gregor 2006).
A third limitation, common to many prevention studies, is that the study relied
primarily on self-report survey data. Specifically, with reference to BMI measurement,
which is computed using self-reported height and weight, it has been suggested that these
self-report statistics often yield inaccurate information, as men are more likely to
overestimate their height (Gunnel 2000; Spencer 2002) while women have been shown to
underestimate their weight (Niedhammer 2000; Brener 2003). However, as a
benchmark diagnostic test for measurement of body weight, BMI has been used
extensively in the obesity field (Gortmaker 1999; Neumark-Sztainer 2003; Storey 2003).
Regardless of study limitations, results from this study suggest that long-term
cigarette use is predictive of later psychological distress and BMI in adulthood. The
developmentally staggered results suggest that such prevention programs need to focus
on cigarette use prevention in early adolescence, followed with components that address
51
psychological distress thereby increasing physical activity levels in adulthood to prevent
later obesity in adulthood. Since continuing cigarette use from adolescence to adulthood
has been shown in this study to have a protective effect on later BMI, prevention efforts
need to de-emphasize and counteract the weight-control benefits of smoking with a focus
on healthy eating, exercise and healthful ways to lose or maintain weight.

52

Chapter 4: Study 2
Introduction
Obesity is a major public health problem with 27.6% of men and 33.2% of women
estimated to be obese in 1999-2002 in the United States (Baskin 2005). Literature shows
that while major proximal risk factors for obesity are lack of physical activity or
sedentary behavior (Jago 2005) and unhealthy eating patterns (Nicklas 2001), other
factors might serve as predisposing risk factors, particularly for physical activity
including depression (Goodwin 2003; Hasler 2005) and cigarette use (Klesges 1990).
However, the roles that these risk factors serve over developmental periods, may
differ for males and females, which would suggest that at minimum, prevention strategies
address male vs. female differences at risk. Specifically, in terms of cigarette use, body
image and weight concerns have been demonstrated to be important factors in the
initiation of smoking in adolescent females, while these factors are less common for
males (French 1995) and (Field 2001). Further, in existing studies male and female
adolescents have been found to endorse different reasons for wanting to quit smoking
with males citing performance-related issues (e.g. to improve athletic performance) and
females commonly stating health reasons for wanting to quit smoking (Aung 2003).
Similarly, in the adult population, males and females have been shown to differ in their
reasons to smoke with women smoking for more tension reduction/relaxation, stimulation
and social reasons as compared to men (Berlin 2003).
53
Gender differences in depressive behavior beginning in adolescence have been
suggested, whereby girls are more likely to be depressed than boys (Allison 2001),
attributed to the changes in hormones, societal expectations and experiences that
accompany puberty and menarche in females (Born 2002). Theoretically, cognitive style
characterized by rumination and negative inferential style in childhood and adolescence
(Hankin 2001) and the social role-gender interaction theory (Gove 1977), with its
emphasis on greater stress placed on married women by extensive family role demands,
in adults have been most often employed to account for gender differences in depression.
Finally, gender has also been shown to have an effect on physical activity. Findings from
existing studies have suggested that men and women exercise for different reasons, in
relation to their eating and related attitudes, with men exercising for positive goals (e.g.,
fitness), whereas women exercising to escape perceived negative consequences (e.g.,
weight gain) (Penas-Lledo 2002).
However, these existing studies are limited by their gender focus either on a single
health behavior (cigarette use, distress, physical activity) or on a specific developmental
period. Lack of a gender-specific focus has also been noted in programs designed for the
prevention of obesity for children and youth (Flynn 2006). However, gender differences
in outcomes of obesity prevention programs have been observed whereby two
interventions with a physical activity focus were found to be effective only for boys
(Sallis 2003; Kain 2004), while other interventions have demonstrated positive outcomes
only in girls (Flores 1995; Gortmaker 1999). This suggests that potential gender
differences in prevention efforts merit further investigation in order that future programs
are better tailored to meet gender needs.
54
Our initial study helped to establish support for simultaneous longitudinal
associations of cigarette use in adolescence to continuing cigarette use, psychological
distress, physical activity, subjective rating of health in emerging adulthood and, finally,
to body mass index (BMI) in early adulthood (Jasuja 2008). The present study evaluated
overall gender differences in interrelations of these multiple risk factors over multiple
developmental periods to BMI in early adulthood. Based on previous research, it was
specifically hypothesized that first, for females, cigarette use may be more strongly
associated with psychological distress than for males; and second, that the relation
between smoking and body weight for females would be negative and stronger for
females than males. To evaluate changes in relations between these constructs over time
and also based on when these constructs were measured, four waves of data were
selected for the current study, which represent four distinct developmental periods,
namely early adolescence(mean:12.06 years), beginning of emerging adulthood
(mean:19.47 years), mid-emerging adulthood (23.22 years), and early adulthood (30.06
years).

Methods
Study background
The current study uses data collected as part of a long-term follow-up of a large
drug abuse prevention trial, the Midwestern Prevention Project (MPP), which began in
Kansas City, Missouri in Fall 1984 and a three-year lagged replication in Indianapolis,
Indiana in Fall 1987 (Pentz 1989). The MPP study was reviewed and full active consent
was approved by the Institutional Review Board of the University of Southern California,
55
Los Angeles. Informed consent forms were repeated each grant cycle with refusal rates
less than 4% at each cycle.
The research design of the MPP is a two-group design, with random assignment of
schools to a program or control condition. The measurement design is longitudinal, with
baseline, 6-month follow-up, and annual follow-ups through high school (12
th
grade), and
18
th
month follow-ups thereafter. The MPP study includes 16 waves of data collection
across four theoretically distinct developmental periods (Junior High, High School,
Emerging Adulthood, and Early Adulthood) representing 23 years, from ages 11 through
34 (Pentz 2006). The intervention component of MPP is a comprehensive multi-faceted
program comprising of five components namely, school, parent, community organization,
health policy, and mass media programs for adolescent drug use prevention. Detailed
description of the design and intervention programs are reported elsewhere (Pentz 1993).

Sample and sample quality
Details of this sample are reported previously (Pentz 1989). Beginning with the
original panel of 1606 participants in Kansas City and 3413 in Indianapolis, planned sub-
sampling over multiple grant periods resulted in 425 and 717 subjects in Kansas City and
Indianapolis respectively, in the current 2003-05 wave (total N=1142). Response rates for
Kansas City and Indianapolis for the 2003-2005 wave are 83.4% and 64.2% respectively.
The population was representative of the Kansas City and Indianapolis populations at the
time the study was initiated (i.e. 1984 in Kansas City and 1987 in
Indianapolis). Currently, sixty percent of this sample is female. Based on self-reported
ethnicity, current sample is currently 83% White, between 27-34 years, and fifty four
56
percent are in the low SES strata. For the current analyses, a ‗stacked‘ data at the early
adulthood wave (mean age=30.06 years) for Kansas City and Indianapolis was used. In
the current study, ‗stacking‘ refers to combining waves of same grade/age cohort data
corresponding to each developmental period studied for each site. This ‗stacked‘ sample
was further restricted to only those participants who had complete data on BMI, thereby
comprising of a total of 1139 subjects (459 males and 676 females).

Measures
Measures were administered in schools during the early adolescence period and a
combination of mail, phone and web surveys were used thereafter for the remaining three
developmental periods.
Cigarette use was the only construct used at multiple waves (early adolescence,
beginning of emerging adulthood, and mid-emerging adulthood) in the present study. In
the early adolescence period, cigarette use items measured lifetime (1=one puff-6=more
than 5 packs), monthly (1=none-6=more than 1 pack), weekly (1=none-6=more than 1
pack), and current usage (1=used to smoke but don‘t-7=a pack a day or more)?‖ (Pentz
1989) and (Johnston 2003). Unlike the early adolescence period, cigarette use items in
both the beginning and mid-emerging adulthood waves measured lifetime, monthly,
weekly, and past 24 hours and the range as well as the wording of the questions for all
these items (except for the lifetime use item) was the same (1=none-7=more than 10
packs). The lifetime cigarette use item from beginning of emerging adulthood and mid-
emerging adulthood period was not included in the current analyses as it was a
dichotomous variable in Indianapolis and Kansas City survey, respectively. The other
57
cigarette use variables in all the waves studied were categorical. The Cronbach alpha for
the cigarette use items was .87 in early adolescence, .95 in the beginning of emerging
adulthood I, and .95 in mid-emerging adulthood.
Psychological distress was used at the beginning of emerging adulthood wave in the
current study. Items measuring distress were originally derived from the Hopkins
Symptom Checklist (Derogatis 1973; Hansell 1991) and represent a 13-item scale, which
include distinct constructs of anxiety (4 items; ―I feel nervous or anxious, I feel I have
nothing to look forward to, I think seriously of a way to hurt myself, I am no good for
anything at all (1=almost never-4=almost always)‖; Cronbach alpha: .70), depression (6
items; ―I feel I don‘t want to do anything, I worry about things in my life, I am tired
during the day, I feel lonely, I feel sad, I find it hard to keep my mind (1=almost never-
4=almost always)‖; Cronbach alpha: .77), and somatic symptoms (3 items; ―I feel dizzy
or lighthearted, I get headaches, I have pains in my heart or chest (1=almost never-
4=almost always)‖; Cronbach alpha:.70). The Cronbach alpha for the entire distress scale
was .84.
Physical activity (CDC 2005) was used only at mid-emerging adulthood. Since the
physical activity items were not measured in the same grade/age cohort in Indianapolis
corresponding to Kansas City, physical activity items were taken from the first wave they
were measured before the emerging adulthood II wave in Indianapolis. Further, only 2
physical activity items were measured in Indianapolis compared to the 3 items in Kansas
City. Subsequently, only the following 2 items were used in the current analyses ―How
many days a week do you exercise (1=0 days on most weeks-6=5 days a week)?‖; ―How
58
many times a week do you exercise so hard that your heart breathes fast or you breathe
hard (1=I don‘t exercise at all-8=6 or more times a week)?‖ (Cronbach alpha: 90).
Subjective rating of health (SAMHSA 1997,1998) was used only at mid-emerging
adulthood wave and was measured by 1 item ―Compared to other people your age, how
would you rate your health (1=poor, 2=average/fair, 3=good, 4=very good,
5=excellent)?‖.
Body mass index (BMI), the primary outcome, was measured at only early
adulthood and was used as a continuous variable in the current analyses. Self-reported
weight (in pounds) and height (both in feet and inches) were converted first into
kilograms (weight) and meters (height) and then used for computation of BMI (weight in
kilograms/(height in meters)
2
(CDC 2005).

Data Analyses
Prior to data analysis, recoding of items was done to make the scales consistent
across both the sites. Subsequent analysis was conducted by ‗stacking‘ waves of data. To
evaluate the potential clustering effect of schools, interclass correlations (ICC) were
generated for the cigarette use items at the school level, which was the only construct
measured at the early adolescence period (mean=12.06 years; 6
th
/7
th
grade) in the current
study using SAS/PROC MIXED (SAS 1999-2001). The ICC‘s for the cigarette use items
were small (lifetime: .06; monthly: .03; weekly: .03; current: .05) thereby indicating that
the effect of school was not large for appropriate adjustments to be made. Since the
subjects had transitioned into adults in the beginning of emerging adulthood, mid-
59
emerging adulthood, and early adulthood, ICC‘s were not generated for the remaining
three developmental periods.
The analysis proceeded in three steps. First, using SAS (SAS 1999-2001),
exploratory factor analysis (EFA) was conducted separately on the items measuring
cigarette use, psychological distress, and physical activity for the two sites. An individual
item was retained if it had a factor loading higher than .50. As expected, all items
representing cigarette use in each of the three waves loaded on one factor. Similarly, the
two items measuring physical activity also loaded on one factor. Items measuring
cigarette use and physical activity were treated as single indicators in the analyses. EFA
on the psychological distress items resulted in three separate factors, namely depression,
anxiety, and somatic symptoms. Thus, three composite scores constituting depression,
anxiety, and somatic symptoms were created for the distress construct.
Additionally, a successive reductive strategy was used, which included Pearson‘s
correlation and chi-square tests, to determine which covariates to use for the current
analyses. First, baseline age, baseline race, baseline socio-economic status, current socio-
economic status, and group condition were included in the hypothesized structural model
along with gender to determine whether any of the covariates had significant paths to the
constructs with the inclusion of gender and hence need to be controlled for in the current
analyses. Except for group condition, all the remaining covariates had some significant
paths to the constructs of interest resulting in exclusion of intervention condition as a
covariate. Second, Pearson‘s correlations were generated in SAS (SAS 1999-2001)
between gender and baseline age, baseline race, baseline socio-economic status and
current socio-economic status to establish which of these remaining covariates were
60
significantly correlated to gender. Only baseline age was significantly correlated with
gender (r=-.08; p=.007) and hence, was deleted from current analyses. Finally, to test if
there were any significant differences by gender on the remaining covariates (baseline
race, baseline socio-economic status, and current socio-economic status), chi-square tests
were generated in SAS (SAS 1999-2001). There were no significant differences on any of
these remaining covariates by gender on a p<.05 significance level and hence, none of the
remaining covariates were included in the current analyses.
In the second step of data analysis, the factors and variables obtained from EFA
were subjected to confirmatory factor analysis (CFA) to yield separate measurement
models for males and females. In the two CFA models, correlational paths between all
the constructs were included as free parameters. To capitalize on all available data, EQS
software (Bentler 1995) was used with the missing data option to generate the CFA and
the subsequent multiple group models for testing measurement invariance. To improve
model fit, specifically goodness-of-fit indices which refer to whether the model captures
the relationships adequately, the Lagrange Multiplier (LM) test (Bentler 1995) of the
EQS program were applied. Further, to encapsulate repeated observations within
individuals over time in the structural model, correlated errors were included for similar
cigarette use items (e.g. monthly cigarette use at early adolescence to monthly cigarette
use at beginning of emerging adulthood and subsequently, from monthly cigarette use at
the beginning of emerging adulthood to monthly cigarette use at mid-emerging
adulthood). Similar pattern of correlated errors was repeated for weekly and past 24 hours
cigarette use over multiple developmental periods. These correlated errors were only
61
added for cigarette use since it was the only construct measured more than once i.e. at
early adolescence, beginning of emerging adulthood, and mid-emerging adulthood.

Testing measurement invariance
The steps suggested by Pentz and Chou (Pentz 1994) for testing measurement
invariance between two groups were used in the current study. As a starting point,
separate models, M
M0
for males and M
F0
for females, were generated. These basic models
were modified by adding correlated measurement errors to obtain a better fit before
combining groups for testing of measurement invariance (M
M1
for males; M
F1
for
females). Thereafter, a basic model with multiple groups (M0) was developed which was
followed by fully constrained models that sequentially constrained all factor loadings,
defined as the correlation between the original variables and the factor, (M1),
covariances, which is a measure of how much two random variables vary together, (M2),
and regression weights, referred to as the estimate of the relationship between the
dependent and the explanatory variable, (M3) to be equal across the two groups. Each
new model was constructed by adding extra constraints to the old model. Modified or
partially constrained models (represented by an asterisk in Table 1) for each of the fully
constrained models were constructed if any of the constraints imposed were not
appropriate. The LM test and the 
2
difference test or the likelihood ratio test (Pentz &
Chou, 1994) was used to guide this procedure.

62
Results
Mean differences on constructs for males vs. females
Mean differences on constructs used and tested in each developmental period in the
present study for males (n = 459) and females (n = 676) demonstrated that females had
significantly higher psychological distress scores compared to males (mean (se):
10.17(5.34) vs. 8.55 (5.29)). However, males scored significantly higher on physical
activity (mean (se): 5.93 (3.75) vs. 4.66 (3.66)) and body mass index (mean (se): 27.07
(4.61) vs. 26.37 (6.81)) than females.

Measurement models
Males
The measurement model (Figure 3a) demonstrates the results of CFA for males
( 
2
=134.172, df=96, NFI=.985, CFI=1.000, RMSEA=.029). There were no significant
covariances between cigarette use at both early adolescence and the beginning of
emerging adulthood periods to psychological distress at the beginning of emerging
adulthood. Further, psychological distress at the beginning of emerging adulthood did not
have significant correlations with cigarette use, physical activity and self-rated health at
mid-emerging adulthood. However, both distress and cigarette use at the beginning of
emerging adulthood were significantly correlated with BMI at early adulthood. Cigarette
use at each of three development periods was not significantly correlated with either
physical activity or subjective rating of health at mid-emerging adulthood. Finally, there
63
was no significant correlation between cigarette use and physical activity at the mid-
emerging adulthood wave or between physical activity and BMI measured at early
adulthood.

Figure 3a: Measurement model (Males)

Females
Figure 3b demonstrates the results of the measurement model for females on the
constructs of interest ( 
2
=143.972, df=94, NFI=.986, CFI=1.000, RMSEA=.028). Unlike
the CFA model for males, cigarette use at the beginning of emerging adulthood and mid-
emerging adulthood were significantly correlated with distress at the beginning of
Early Adulthood
(mean:30.06 yrs)
BMI
Mid-Emerging Adulthood
(mean:23.22 yrs)
Subjective rating
of health
Psychological
distress
Physical
activity
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.985
CFI=1.000
2 =134.172
(96)
RMSEA=.029
N=459
Only significant paths are shown in this figure (p<.05)
.362
-.138
-.133
-.266
Figure 3a. Measurement model (Males)
.215
.639
.199
.494
Early Adulthood
(mean:30.06 yrs)
BMI
Mid-Emerging Adulthood
(mean:23.22 yrs)
Subjective rating
of health
Psychological
distress
Physical
activity
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.985
CFI=1.000
2 =134.172
(96)
RMSEA=.029
N=459
Only significant paths are shown in this figure (p<.05)
.362
-.138
-.133
-.266
Figure 3a. Measurement model (Males)
.215
.639
.199
.494
BMI
Mid-Emerging Adulthood
(mean:23.22 yrs)
Subjective rating
of health
Psychological
distress
Physical
activity
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.985
CFI=1.000
2 =134.172
(96)
RMSEA=.029
N=459
Only significant paths are shown in this figure (p<.05)
.362
-.138
-.133
-.266
Figure 3a. Measurement model (Males)
.215
.639
.199
.494
64
emerging adulthood wave. Further, cigarette use at middle school had significant
correlations with physical activity and subjective rating of health at mid-emerging
adulthood. Additionally, compared to the model for males, psychological distress had
significant correlations with all the three constructs at mid-emerging adulthood namely,
cigarette use, physical activity, and self-rated health. However, unlike the model for
males, cigarette use both at the beginning of emerging adulthood and mid-emerging
adulthood were significantly correlated with self-rated health. However, cigarette use at
mid-emerging adulthood did not have significant correlations with physical activity at
mid-emerging adulthood. Finally, compared to the CFA model for males in which
cigarette use and self-rated health at mid-emerging adulthood were significantly
correlated with BMI at early adulthood, in the model for females, physical activity and
self-rated health had significant correlations with BMI.

Figure 3b: Measurement model (Females)

Early Adulthood
(mean:30.06 yrs)
Early Adolescence
(mean:12.06 yrs)
Cigarette use
Model fit:
NFI=.986
CFI=1.000
2 =143.972
(94)
RMSEA=.028
N=676
Only significant paths are shown (p<.05)
BMI
Mid-Emerging Adulthood
(mean:23.22 yrs)
Subjective rating
of health
Psychological
distress
Physical
activity
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use
Cigarette use
.427
.364
-.198
-.168
.359
.707
-.209
.368
-.221
-.395
-.322
.370
-.216
-.271
Early Adulthood
(mean:30.06 yrs)
Early Adolescence
(mean:12.06 yrs)
Cigarette use
Model fit:
NFI=.986
CFI=1.000
2 =143.972
(94)
RMSEA=.028
N=676
Only significant paths are shown (p<.05)
BMI
Mid-Emerging Adulthood
(mean:23.22 yrs)
Subjective rating
of health
Psychological
distress
Physical
activity
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use
Cigarette use
.427
.364
-.198
-.168
.359
.707
-.209
.368
-.221
-.395
-.322
.370
-.216
-.271
65
Summary results of tests of measurement invariance
Results for testing of measurement invariance across the two groups are reported in
Table 4. As is evident from the table, the difference test (M1-M0) with 
2
(10)=11.278,
p=.34, accepted the null hypothesis that the factor loadings are equal across the two
groups. On the other hand, the significant difference in 
2
between M2-M1 ( 
2
(4)=15.567, p=.00) indicated that not all covariance constraints are appropriate.
Specifically, as suggested by the 
2
difference tests, the constraints between cigarette use
and distress at emerging adulthood I ( 
2
(1)=6.289, p=.01) and between physical activity
and self-rated health at mid-emerging adulthood ( 
2
(1)=5.513, p=.02) were not
acceptable and hence, were released in the partially constrained model, M2*. Similar
results were observed for the regression weight constraints as indicated by the significant
difference in 
2
between M3-M2* ( 
2
(11)=20.532, p=.040). Further, 
2
difference tests
suggested that the paths between self-rated health and BMI ( 
2
(1)=12.732, p=<.001) and
physical activity and BMI ( 
2
(1)=6.466, p=.01) were not equal across the two groups.
Thus, on the basis of statistical tests it can be concluded that overall data were invariant
across males and females with only 2 covariances and 2 regression weights different
between the two groups out of a total of 25 parameters tested.

66
Table 4: Summary results of tests of measurement invariance for males and females
Group 
2
df p NFI CFI RMSEA
Males
M
M0
275.016 107 .000 .942 .978 .059
M
M1
146.019 102 .002 .983 1.000 .031
Females
M
F0
409.57 107 .000 .935 .956 .065
M
F1
167.733 102 .000 .983 1.000 .031
Combined
M0 313.749 204 .000 .983 1.000 .031
M1 325.027 214 .000 .982 1.000 .030
M1-M0 11.278 10 .34
M2 340.594 218 .000 .982 1.000 .032
M2-M1 15.567 4 .00
M2* 328.645 216 .00 .982 1.000 .030
M2*-M1 3.618 2 .16
M3 349.177 227 .000 .980 1.000 .031
M3-M2* 20.532 11 .040
M3* 335.646 225 .000 .981 1.000 .029
M3*-M2* 7.001 9 .64
with all factor loadings constrained equal across groups; M2=model with all covariances
constrained equal over MI; M3= model with all regression weights constrained equal over M2*.
An asterisk indicated a model with relevant parameters partially constrained equal across
groups. M
M0
and M
F0
= basic theoretical models for males (N=459) and females (N=676) groups,
respectively;M
M1
and M
F1
=modified models with correlated errors suggested by the LM test on
M
M0
and M
F0,
respectively M0=basic model combining males and females groups, M1=model
Note. p-values are based on a two tailed test.

Final structural model
The final SEM model with unstandardized estimates is shown in Figure 4. As
reported above, two covariances varied between males and females. While males had a
negative correlation between cigarette use and distress at the beginning of emerging
adulthood, this correlation was positive for females. Second, males reported a stronger
positive correlation between physical activity and subjective rating of health at mid-
emerging adulthood as compared to females. Additionally, two regression weights
appeared to be different across males and females. Compared to males who reported a
positive relation between physical activity and BMI, this relationship was negative for
67
females. Second, the negative relation between subjective rating of health and BMI was
stronger for females compared to males.

Figure 4: Final structural model

Discussion
The primary objective of the current study was to examine potential gender
differences in the relations of cigarette use, psychological distress, physical activity, self-
rated health, and BMI longitudinally, from adolescence to adulthood. As hypothesized,
cigarette use was more strongly associated with distress for females as compared to males
based on the tests of measurement invariance. Compared to the negative correlation
between cigarette use and distress in males, females had a positive correlation between
cigarette use and distress thereby suggesting that females by adulthood use smoking more
Early Adulthood
(mean:30.06 yrs)
BMI
Mid-Emerging Adulthood
(mean:23.22 yrs)
Subjective rating
of health
Psychological
distress
Physical
activity
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
A parameter with one value indicates that the parameter was constrained equal across the two groups.
A parameter with two values indicates that the parameter was not equal with the estimate for females shown in parentheses.
8.822
-.020
-.085
-1.460 (-1.914)
.728
-.013
.340
-1.227 (4.616)
-.094
-.094
.320
.133
.397 (-.450)
-.883
.873 (.485)
Early Adulthood
(mean:30.06 yrs)
BMI
Mid-Emerging Adulthood
(mean:23.22 yrs)
Subjective rating
of health
Psychological
distress
Physical
activity
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
A parameter with one value indicates that the parameter was constrained equal across the two groups.
A parameter with two values indicates that the parameter was not equal with the estimate for females shown in parentheses.
8.822
-.020
-.085
-1.460 (-1.914)
.728
-.013
.340
-1.227 (4.616)
-.094
-.094
.320
.133
.397 (-.450)
-.883
.873 (.485)
68
to compensate for distress. This finding supports previous research in which gender
differences in the association between cigarette smoking and elevated depressive
symptoms was reported. In a study conducted on an adolescent population, a significant
positive association was found between smoking and depressive disorder in females only
(Poulin 2005).This strong cigarette use and distress relation for females by adulthood
might suggest that early universal prevention programs for tobacco use be supplemented
with later distress prevention or coping skills training for females. However, the second
hypothesis that the relation between cigarette use and BMI would be negative and
stronger for females than males was not supported as results suggested that this
relationship did not vary across the two groups. Previous research has shown smoking to
be negatively associated with body weight in both males and females (Rasmussen 2003;
Bamia 2004).
Further, tests of measurement invariance demonstrated that two regression weights
differed for males and females, namely the effect of physical activity on BMI and the
effect of self-rated health on BMI. Compared to the positive physical activity-BMI
relation for males, this relation was negative for females. Existing research has shown
physical activity to be inversely related to BMI in adult females (Ball 2001), and to have
a positive association with BMI in adolescent males (Parsons 2006) and no association in
adult males (Ball 2001; Parsons 2006). Interestingly, this positive association between
physical activity and BMI in males could be possibly due to BMI being a less accurate
predictor of body fatness. Recent evidence on BMI as a surrogate measure of adiposity
suggests that it may provide misleading information in muscular (not fat) individuals,
specifically athletes (Prentice 2001). The other regression weight which significantly
69
differed between the two groups was the path between self-rated health and BMI, with
females showing a stronger effect than males. Existing literature has shown poor self-
rated health as a risk factor for obesity mainly among older adults (Kaplan 2003).
However, to our knowledge, gender differences in the self-rated health-BMI relation have
not been studied yet.
Besides the cigarette use-distress association mentioned above, another covariance
parameter that differed between the two genders was the physical activity-self rated
health correlation whereby males had significantly higher positive correlation between
these two variables as compared to females. This positive correlation between physical
activity and subjective rating of health is consistent with the findings of an existing study
conducted on male adults, in which low self-rated health was found to have a significant
association with low levels of physical activity (Norman 2002). However, the current
study extends beyond previous research by demonstrating potential gender differences in
the physical activity-self-rated health relation.
The current results must be considered in light of study limitations. The first
limitation is that measurement of BMI is limited only to the early adulthood period since
drug abuse prevention rather than obesity was the focus of the original parent MPP study.
In the literature, overweight in childhood/adolescence has been shown to be a strong
predictor of overweight/obesity in adulthood (Guo 2000). Hence, adolescent overweight
status cannot be used as a covariate in analyses due to limitations of the MPP dataset.
A second limitation concerning the comprehensiveness of measures both with
regard to the single item used to measure subjective rating of health as well as the two
items representing physical activity. Though the use of a single-item question measuring
70
self-rated health has warranted caution however, self-rated health has also been
demonstrated to be a valid measure for measuring health status in adolescents
(Agyemang 2006) as well as adults (Gregor 2006). The two physical activity items were
taken from the 2005 Youth Behavior Risk Survey developed by the CDC (CDC 2005).
Though the limitations of these YRBS physical activity items have been acknowledged,
however, this was done in a study conducted on middle school population (Troped 2007).
Further, even though diet, specifically protein and fat intakes, has been shown to be a
positive predictor of BMI (Skinner 2004) in existing research, however, one of the
limitations of the MPP dataset is the lack of a dietary measure.
A third limitation, common to many prevention studies, is that the study relied
primarily on self-report survey data. Specifically, with reference to BMI measurement,
which is computed using self-reported height and weight, it has been suggested that these
self-report statistics often yield inaccurate information, as men are more likely to
overestimate their height (Spencer 2002) while women have been shown to
underestimate their weight (Niedhammer 2000). However, as a benchmark diagnostic test
for measurement of body weight, BMI has been used extensively in the obesity field
(Neumark 2003).
A fourth potential limitation concerns the issue of timing of the measurement of the
exposure and the outcome in this study. For instance, items measuring psychological
distress were not included in the survey since baseline and were measured starting from
the beginning of emerging adulthood. However, items such as distress were not the focus
of the original study or intervention, and much like other longitudinal studies, constructs
of interest were added to reflect developmental changes (Newcomb 1997). Additionally,
71
the temporal ordering of constructs in this study supports past research which has
suggested that temporally cigarette use starts as social experimentation and distress
follows much later on.
Regardless of study limitations, findings from the current study illustrate certain
differences in relationships of cigarette use, psychological distress, physical activity, self-
rated health, and BMI between males and females across multiple developmental periods.
This study suggests several issues to focus on while conducting research, designing,
building and implementing prevention programs targeting these health behaviors.
Specifically, the co-morbid relationship between distress and smoking behaviors suggests
that prevention efforts aimed at reducing smoking in adult females should incorporate an
additional component for addressing depression and anxiety in this population. Further,
the positive relation between physical activity and BMI in adult males demonstrates the
need for utilizing more accurate measures for measuring body fat in the area of obesity
research especially for males. On the other hand, the negative relation between these two
health behaviors in females supports the consistent findings of low physical activity as a
major proximal predictor of obesity risk. However, the relations hypothesized in this
study over different developmental periods bring into light the need for obesity
prevention programs to focus on other predisposing risk factors such as cigarette use,
depression and self-rated health in adolescence and emerging adulthood which might
signal later obesity risk in adulthood. These findings, in conjunction with a better
understanding of the causes of such differences, will have important public health
implications for the development and targeting of gender-specific prevention strategies.

72
Chapter 5: Study 3

Introduction

Coping has been fundamentally defined as ―continually changing behavioral and
cognitive efforts to manage external and/or internal demands that are appraised as
exceeding the individual‘s resources‖ (Lazarus 1984). Overall, coping styles have been
distinguished into more direct and adaptive coping modes (i.e., problem-focused, primary
control, or approach coping) and more indirect and generally maladaptive coping efforts
(emotion-focused, secondary control, or avoidance coping) (Compas 2001). Recently, a
multi-dimensional structure of coping has been proposed in the literature whereby it has
been suggested that coping styles can be differentiated into two adaptive coping styles
(emotion- and problem-focused coping) and one maladaptive coping style, which is
represented by nine coping strategies (Compas 2001; Hampel 2006). According to this
three-factor model of coping, emotion-focused coping is comprised of minimization and
distraction; problem-focused coping consists of situation control, positive self-
instructions, and social support; while maladaptive coping includes passive avoidance,
rumination, resignation, and aggression.
In general, previous studies have shown that adaptive coping strategies were linked
with better psychological adjustment, whereas maladaptive coping strategies have been
shown to be associated with increased depressive symptoms (Ebata 1991; Ebata 1994;
Chan 1995; Herman-Stahl 1999; Seiffge-Krenke 2000). To elucidate further, Chan (1995)
found that adolescents diagnosed as depressed and anxious were all characterized by
frequent use of withdrawal or avoidance, independent of gender. Others suggest that
73
depressive symptomatology among adolescents is accompanied by higher levels of
passive and avoidant coping but lower levels of active and approach coping styles (Ebata
1991; Herman-Stahl 1999). In contrast, others propose that less depressed individuals are
more likely to rely on active coping styles (Vickers 2003).
Conversely, existing research has also suggested that avoidant coping styles, such
as aggression and withdrawal, may be considered as a risk factor for developing or
increasing depressive symptoms (Herman-Stahl 1995; Seiffge-Krenke 2000; Murberg
2005). Specifically, in the study conducted by Herman-Stahl, Stemmler, & Petersen
(1995), it was found that approach-oriented copers reported the fewest depressive
symptoms, whereas avoidant copers reported the most. Adolescents who changed from
using approach oriented coping to avoidant coping during the first year showed a
significant increase in depressive symptoms, whereas adolescents who switched from
avoidant to approach-oriented coping during the same time period showed a decrease in
depressive symptoms.
A significant positive relation between maladaptive coping strategies, such as
avoidance and emotion-focused, and eating pathology (especially bulimic pathology) has
also been noted in previous studies (Troop 1994; Yager 1995; Nagata 2000). Further,
these studies have also found a negative relation between bulimic symptoms and task-
oriented coping, a form of adaptive coping (Nagata 2000). Impulsivity, a characteristic
feature of bulimia, has been hypothesized to be related to maladaptive coping strategies
(Nagata 2000). In terms of body weight, in a study conducted on weight gain in a college
population, women in the body mass index (BMI) gain group were more likely to use
maladaptive coping behaviors (Adams 2007). Additionally, a reduction in emotion-
74
focused coping has been found to lead to weight loss and problem-focused coping
strategies have been have been used in designing effective obesity intervention programs
(Ryden 2003).
However, the relation of coping with both psychological distress and body mass
index (BMI) has not been examined simultaneously in existing studies. Based on
existing research since adaptive and maladaptive approach have shown to be more
relevant to obesity risk, the purpose of the present study was to examine the potential
roles of maladaptive and adaptive coping as a mediator or a moderator in the relation
between distress and BMI. It is specifically hypothesized, first, that maladaptive coping
strategies will mediate the distress-BMI relation, whereby distress at the beginning of
emerging adulthood will predict maladaptive coping strategies, comprising of avoidance,
negative affect, and drug use items, at mid-emerging adulthood, finally leading to
increased BMI at the early adulthood period. Second, adaptive coping strategies,
comprising of physical activity, support, and relaxation items, will have a moderating
effect on the distress-BMI relation, whereby there will be stronger associations between
psychological distress and BMI in individuals with low adaptive coping skills as
compared to individuals with high adaptive coping skills. Examination of the relation
between these constructs over time and also based on when these constructs were
measured, four waves of data were selected for the current study, which represent four
distinct developmental periods, namely early adolescence (mean=12.06 years), beginning
of emerging adulthood (mean=19.25 years), mid-emerging adulthood (mean=23.17
years), and early adulthood (mean=30.06 years).

75
Methods
Study background
The current study uses data collected as part of a long-term follow-up of a large
drug abuse prevention trial, the Midwestern Prevention Project (MPP), which began in
Kansas City, Missouri in Fall 1984 and a three-year lagged replication in Indianapolis,
Indiana in Fall 1987 (Pentz 1989). The MPP study was reviewed and full active consent
was approved by the Institutional Review Board of the University of Southern California,
Los Angeles. Informed consent forms were repeated each grant cycle with refusal rates
less than 4% at each cycle.
The research design of MPP is a two-group design, with random assignment of
schools to a program or control condition. The measurement design is longitudinal, with
baseline, 6-month follow-up, and annual follow-ups through high school (12
th
grade), and
18
th
month follow-ups thereafter. The MPP study includes 16 waves of data collection
across four theoretically distinct developmental periods (Junior High, High School,
Emerging Adulthood, and Early Adulthood) representing 23 years, from ages 11 through
34 (Pentz 2006). The intervention component of the MPP is a comprehensive multi-
faceted program comprising of five components namely, school, parent, community
organization, health policy, and mass media programs for adolescent drug use prevention.
Detailed description of the design and intervention programs are reported elsewhere
(Pentz 1993).

76
Sample and sample quality
Details of this sample are reported previously (Pentz 1989). Beginning with the
original panel of 1606 participants in Kansas City and 3413 in Indianapolis, planned sub-
sampling over multiple grant periods resulted in 425 and 717 subjects in Kansas City and
Indianapolis respectively, in the current 2003-05 wave (total N=1142). Response rates for
Kansas City and Indianapolis for the 2003-2005 wave are 83.4% and 64.2% respectively.
The population was representative of the Kansas City and Indianapolis populations at the
time the study was initiated (i.e. 1984 in Kansas City and 1987 in
Indianapolis). Currently, sixty percent of this sample is female. Based on self-reported
ethnicity, current sample is currently 83% White, between 27-34 years, and fifty four
percent are in the low SES strata. For the current study, data analyses will be conducted
by ‗stacking‘ waves of data, referred to in this study as combining waves of same
grade/age cohort data corresponding to each developmental period studied for both sites,
thereby comprising of a total of 5019 subjects.

Measures
Measures were administered in schools during the early adolescent period and a
combination of mail, phone and web surveys were used thereafter for the remaining three
developmental periods.
Cigarette use was the only construct used at multiple waves (early adolescent,
beginning of emerging adulthood, mid-emerging adulthood, and early adulthood) in the
present study. In the early adolescent period, cigarette use items measured lifetime
(1=one puff-6=more than 5 packs), monthly (1=none-6=more than 1 pack), weekly
77
(1=none-6=more than 1 pack), and current usage (1=used to smoke but don‘t-7=a pack a
day or more)?‖ (Pentz 1989; Johnston 2003). Cigarette use items in the beginning, mid-
emerging adulthood and early adulthood waves measured lifetime, monthly, weekly, and
past 24 hours and the range for all these items (except for the lifetime use item) was the
same (1=none-7=more than 10 packs). The lifetime cigarette use item from the beginning
of emerging adulthood, mid-emerging adulthood, and early adulthood period was not
included in the current analyses. The other cigarette use variables in all the waves studied
were categorical. The Cronbach alpha for the cigarette use items was .88 in early
adolescence, .95 in the beginning of emerging adulthood, .95 in mid-emerging adulthood,
and .96 in the early adulthood period
Psychological distress was used at the beginning of emerging adulthood, mid-
emerging adulthood, and early adulthood waves in the current study. Items measuring
distress were originally derived from the Hopkins Symptom Checklist (Derogatis 1973;
Hansell 1991) and represent a 13-item scale, which include distinct constructs of anxiety
(4 items; ―I feel nervous or anxious, I feel I have nothing to look forward to, I think
seriously of a way to hurt myself, I am no good for anything at all (1=almost never-
4=almost always)‖, depression (6 items; ―I feel I don‘t want to do anything, I worry about
things in my life, I am tired during the day, I feel lonely, I feel sad, I find it hard to keep
my mind (1=almost never-4=almost always)‖; and somatic symptoms (3 items; ―I feel
dizzy or lighthearted, I get headaches, I have pains in my heart or chest (1=almost never-
4=almost always)‖. At both the mid-emerging adulthood and early adulthood periods,
the item ‗I am tired during the day‘ was not present and hence the sub-construct of
78
depression included a total of 5 items. The Cronbach alpha for the entire distress scale at
beginning of emerging adulthood was .84, .86 in mid-emerging adulthood, and .86 in
early adulthood.
Coping was used at the mid-emerging adulthood wave and was measured with 24
items from the Wills Coping Inventory (Wills 1996) (Cronbach alpha=.92). Factor
analysis on the entire coping construct resulted in two distinct sub-constructs of adaptive
coping (7 items; ―I work it off with physical exercise, I go out and play sports, I get
information that is needed to solve the problem, I talk with my mother or father, I talk
with one of my friends, I try to calm myself, I try deep breathing (1=almost never-
4=almost always)‖; (Cronbach alpha: .81) and maladaptive coping (10 items; ―I tell
myself that the problem is not worth getting upset about, I wait and hope things will get
better with time, I try not to think about it, I withdraw, I get mad at people, I sleep more, I
get away from people, I say or do unpleasant things, I smoke a cigarette to relax, I drink
alcohol/drugs to feel better (1=almost never-4=almost always)‖ (Cronbach alpha:.86).
Body mass index (BMI), the primary outcome, was measured at only the early
adulthood wave and was used as a continuous variable in the current analyses. Self-
reported weight (in pounds) and height (both in feet and inches) were converted first into
kilograms (weight) and meters (height) and then used for computation of BMI (weight in
kilograms/(height in meters)
2
(CDC 2005).

Data analyses
Prior to data analysis, recoding of items was done to make the scales consistent
across both the sites. Subsequent analysis was conducted by ‗stacking‘ waves of data.
79
The analysis proceeded in three steps. First, using SAS (SAS 1999-2001), exploratory
factor analysis (EFA) was conducted separately on the items measuring cigarette use and
psychological distress for the two sites. An individual item was retained if it had a factor
loading higher than .50. As expected, all items representing cigarette use in each of the
three waves loaded on one factor. Items measuring cigarette use were treated as single
indicators in the analyses. EFA on the psychological distress items resulted in three
separate factors, namely depression, anxiety, and somatic symptoms. Thus, three
composite scores constituting depression, anxiety, and somatic symptoms were created
for the distress construct.
In the second step of data analysis, the factors and variables obtained from EFA
were subjected to confirmatory factor analysis (CFA) to yield separate measurement
models with adaptive and maladaptive coping. In the two CFA models, correlational
paths between all the constructs were included as free parameters. To capitalize on all
available data, EQS software (Bentler 1995) was used with the missing data option to
generate the CFA and the subsequent multiple group models for testing measurement
invariance. To improve model fit, specifically goodness-of-fit indices which refer to
whether the model captures the relations adequately, the Lagrange Multiplier (LM) test
(Bentler 1995) of the EQS program were applied.

Testing measurement invariance
The steps suggested by Pentz and Chou (Pentz 1994) for testing measurement
invariance between the low adaptive and high adaptive coping groups were used in the
80
current study. As a starting point, separate models, M
L0
for low adaptive coping and M
H0
for high adaptive coping, were generated. These basic models were modified by adding
correlated measurement errors to obtain a better fit before combining groups for testing
of measurement invariance (M
L1
for low adaptive coping; M
H1
for high adaptive coping).
Thereafter, a basic model with multiple groups (M0) was developed which was followed
by fully constrained models that sequentially constrained all factor loadings, defined as
the correlation between the original variables and the factor, (M1), covariances, which is
a measure of how much two random variables vary together, (M2), and regression
weights, referred to as the estimate of the relationship between the dependent and the
explanatory variable, (M3) to be equal across the two groups. Each new model was
constructed by adding extra constraints to the old model. Modified or partially
constrained models (represented by an asterisk in Table 3) for each of the fully
constrained models were constructed if any of the constraints imposed were not
appropriate. The LM test and the 
2
difference test or the likelihood ratio test (Pentz &
Chou, 1994) was used to guide this procedure.

Results
Mean differences on outcomes
Mean differences between high and low adaptive coping groups and high and low
maladaptive coping groups on the outcomes at early adulthood period are reported in
Table 5. As is evident from the table, subjects in the low adaptive coping group (N=937)
scored significantly higher on all the outcomes measured at early adulthood as compared
81
to the high adaptive coping group (N=783). Comparison of the low maladaptive coping
(N=880) group to the high maladaptive coping group (N=839) indicated that subjects in
the high maladaptive coping group reported significantly high scores on overall
psychological distress and its sub-constructs used in the current study, namely,
depression, anxiety, and somatic symptoms.

Table 5: Mean differences on outcomes for adaptive and maladaptive coping
Low adaptive coping High adaptive coping Low maladaptive coping High maladaptive coping
(N=937) (N=783) (N=880) (N=839)
Mean (SE) Mean (SE) Mean (SE) Mean (SE)
Early adulthood outcomes
Depression 4.28 (.14) 3.69 (.12)** 3.50 (.12) 4.44 (.13)***
Anxiety 1.29 (.08) .95 (.06)** .93 (.07) 1.31 (.08)***
Somatic symptoms 1.37 (.07) 1.15 (.06)** 1.10 (.06) 1.42 (.07)***
Distress 6.95 (.25) 5.79 (.20)*** 5.53 (.22) 7.16 (.23)***
Lifetime cig. use 13.47 (.40) 12.09 (.41)* 12.33 (.41) 13.20 (.40)
Monthly cig. use 7.00 (.46) 4.96 (.42)** 5.72 (.45) 6.23 (.44)
Weekly cig. use 6.45 (.45) 4.20 (.39)*** 5.10 (.43) 5.56 (.42)
Past 24 hour cig. use 4.40 (.35) 2.68 (.28)*** 3.18 (.31) 3.90 (.33)
BMI 27.12 (.30) 26.29 (.30)
+
26.57 (.30) 26.84 (.30)
Note: +p<.10, *p<.05, **p<.01, ***p<.001;
Proc univariate in SAS was used to find the median to create the four adaptive coping groups at mid-emerging adulthood.

82

Coping item M SD Distress Depression Anxiety
Somatic
symptoms
Lifetime
cig. use
Monthly
cig. use
Weekly
cig. use
Past 24 hrs.
cig use BMI
Maladaptive coping
Smoke a cig. to relax .56 .93 .09* .09* .06 .07* .35* .41* .40* .39* -.02
Drink alcohol or use drugs .42 .67 .08* .09* .08* .01 .18* .16* .16* .13* .04
Get mad at people .85 .76 .17* .16* .14* .12* 0.04 -.03 -.03 .00 .04
I sleep more .79 .81 .13* .13* .11* .07 0.007 -.04 -.05 -.03 .05
Get away from people .85 .80 .09* .12* .03 .03 0.008 -.09* -.09* -.08* -.02
Get mad or say or do unpleasant things .62 .81 .15* .13* .14* .12* 0.04 .02 .01 .04 .05
Tell myself problem not worth getting upset 1.33 .96 -.05 -.04 -.04 -.05 -.04 -.09* -.11* -.10* -.04
Wait and hope things will be better with time 1.07 .87 .08* .09* .04 .06 -.05 -.06 -.07 -.06 -.005
Try not to think about it .94 .80 .05 .08* .03 -.009 -.008 -.11* -.11 -.08* -.010
I withdraw .59 .75 .20* .22* .15* .11* -.004 -.03 -.05 -.01 .02
Adaptive coping
Get info needed to deal with the problem 1.29 1.11 -.07* -.05 -.07 -.07* -.010 -.11* -.13* -.13* -.08*
Work it off with physical exercise .90 .92 -.11* -.11* -.05 -.11* -.06 -.12* -.15* -.14* -.08*
Go out and play sports .76 .92 -.11* -.12* -.06 -.09* -.11* -.13* -.16* -.16* .00
Talk with my mother or father 1.18 .99 -.02 -.04 -.004 .01 -.02 -.09* -.10* -.08* -.07*
Talk with one of my friends 1.73 .95 -.05 -.02 -.09* -.03 0.01 -.03 -.05 -.04 -.05
Try to calm myself 1.16 .93 -.03 -.02 -.05 -.02 -.03 -.10* -.10* -.09* -.05
Try deep breathing .81 .88 .01 .01 .01 .02 .04 -.003 -.009 .01 .01
Note: *p<.05; N=5019.

Measurement models
Adaptive coping
The measurement model (Figure 5a) demonstrates the results of CFA for adaptive
coping ( 
2
=764.957, df=256, NFI=.995, CFI=1.000, RMSEA=.020). There were no
significant covariances between cigarette use at early adolescence to psychological
distress at the beginning of emerging adulthood, which was also not significantly
correlated to adaptive coping at mid-emerging adulthood. Further, there was no
significant association between cigarette use and distress at mid-emerging adulthood,
between cigarette use at mid-emerging adulthood and distress at early adulthood, and
between distress at mid-emerging adulthood and cigarette use at early adulthood.
Summary of descriptive statistics for the complete set of coping items used
in modeling and bivariate correlations between each of these coping measures at
mid-emerging adulthood and the early adulthood outcomes are shown in Table 6.

Table 6: Summary descriptive statistics and correlations for coping items

TA

83
Additionally, adaptive coping at mid-emerging adulthood was not significantly correlated
to either distress at early adulthood. Finally, BMI at early adulthood was only
significantly correlated to adaptive coping at mid-emerging adulthood and distress at
early adulthood.

Figure 5a: Measurement model (Adaptive coping)

Maladaptive coping
The measurement model (Figure 5b) demonstrates the results of CFA for
maladaptive coping ( 
2
=875.599, df=247, NFI=.994, CFI=1.000, RMSEA=.023). Similar
to the adaptive coping model, there were no significant covariances between cigarette use
Early Adulthood
(mean:30.06 yrs)
Mid-Emerging Adulthood
(mean:23.22 yrs)
Psychological
distress
Psychological
distress
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.995
CFI=1.000
2 =764.957
(256)
RMSEA=.020
N=5019
Only significant paths are shown in this figure (p<.05)
.322
.245
Cigarette use
BMI
Psychological
distress
.656
.152
.256
-.164
.450
Adaptive
coping
.296
-.089
.160
-.152
.132
-.085
.654
.615
.174
-.092
.128
.466
.394
-.084
-.088
Early Adulthood
(mean:30.06 yrs)
Mid-Emerging Adulthood
(mean:23.22 yrs)
Psychological
distress
Psychological
distress
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.995
CFI=1.000
2 =764.957
(256)
RMSEA=.020
N=5019
Only significant paths are shown in this figure (p<.05)
.322
.245
Cigarette use
BMI
Psychological
distress
.656
.152
.256
-.164
.450
Adaptive
coping
.296
-.089
.160
-.152
.132
-.085
.654
.615
.174
-.092
.128
.466
.394
-.084
-.088
84
at early adolescence to psychological distress at the beginning of emerging adulthood.
However, compared to the adaptive coping CFA model, distress at the beginning of
emerging adulthood was significantly correlated to maladaptive coping at mid-emerging
adulthood. Further, unlike the adaptive coping model, there was no significant association
between cigarette use and maladaptive coping at mid-emerging adulthood or between
maladaptive coping at mid-emerging adulthood and cigarette use at early adulthood.
Finally, there was no significant correlation between maladaptive coping at mid-
emerging adulthood and BMI at early adulthood.

Figure 5b: Measurement model (Maladaptive coping)

Early Adulthood
(mean:30.06 yrs)
Mid-Emerging Adulthood
(mean:23.22 yrs)
Psychological
distress
Psychological
distress
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.994
CFI=1.000
2 =875.599
(247)
RMSEA=.023
N=5019
Only significant paths are shown in this figure (p<.05)
.325 .637
Cigarette use
BMI
Psychological
distress
.520
.141
.
.488
Maladaptive
coping
.173
.139
.155
.214
.304
.542
.135
.070
.254
.414
.197
.468
.258
.554
Early Adulthood
(mean:30.06 yrs)
Mid-Emerging Adulthood
(mean:23.22 yrs)
Psychological
distress
Psychological
distress
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.994
CFI=1.000
2 =875.599
(247)
RMSEA=.023
N=5019
Only significant paths are shown in this figure (p<.05)
.325 .637
Cigarette use
BMI
Psychological
distress
.520
.141
.
.488
Maladaptive
coping
.173
.139
.155
.214
.304
.542
.135
.070
.254
.414
.197
.468
.258
.554
85
Structural models
Adaptive coping
The structural model (Figure 6a) demonstrates the results for adaptive coping
( 
2
=873.367, df=268, NFI=.994, CFI=1.000, RMSEA=.021). Cigarette use at the
beginning of emerging adulthood significantly predicted adaptive coping at mid-
emerging adulthood which subsequently, predicted all the three outcomes at early
adulthood, namely, cigarette use, distress, and BMI. Further, adaptive coping had
significant correlations with both cigarette use and distress at mid-emerging adulthood.

Figure 6a: Structural model (Adaptive coping)

Early Adulthood
(mean:30.06 yrs)
Mid-Emerging Adulthood
(mean:23.22 yrs)
Psychological
distress
Psychological
distress
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.994
CFI=1.000
2 =873.367
(268)
RMSEA=.021
N=5019
Only significant paths are shown in this figure (p<.05)
Cigarette use
BMI
Psychological
distress
.126
.290
.089
Adaptive
coping
.157
.102
.337 .665 .669
.473
.446
-.116
-.085
-.189
-.099
-.081
-.105
-.115
Early Adulthood
(mean:30.06 yrs)
Mid-Emerging Adulthood
(mean:23.22 yrs)
Psychological
distress
Psychological
distress
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.994
CFI=1.000
2 =873.367
(268)
RMSEA=.021
N=5019
Only significant paths are shown in this figure (p<.05)
Cigarette use
BMI
Psychological
distress
.126
.290
.089
Adaptive
coping
.157
.102
.337 .665 .669
.473
.446
-.116
-.085
-.189
-.099
-.081
-.105
-.115
86
Maladaptive coping
The structural model (Figure 6b) demonstrates the results for maladaptive coping
( 
2
=1087.990, df=259, NFI=.993, CFI=1.000, RMSEA=.025). Compared to the structural
model for adaptive coping, psychological distress (and not cigarette use) at the beginning
of emerging adulthood significantly predicted maladaptive coping at mid-emerging
adulthood. However, unlike the adaptive coping model, maladaptive coping did not have
any significant direct path to any of the three outcomes at early adulthood. Further,
maladaptive coping was significantly correlated to only distress at mid-emerging
adulthood.

Figure 6b: Structural model (Maladaptive coping)

Early Adulthood
(mean:30.06 yrs)
Mid-Emerging Adulthood
(mean:23.22 yrs)
Psychological
distress
Psychological
distress
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.993
CFI=1.000
2 =1087.990
(259)
RMSEA=.025
N=5019
Only significant paths are shown in this figure (p<.05)
Cigarette use
BMI
Psychological
distress
.121
.505
Maladaptive
coping
.158
.156
.347 .603 .596
.487
.079
.472
.260
-.131
Early Adulthood
(mean:30.06 yrs)
Mid-Emerging Adulthood
(mean:23.22 yrs)
Psychological
distress
Psychological
distress
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.993
CFI=1.000
2 =1087.990
(259)
RMSEA=.025
N=5019
Only significant paths are shown in this figure (p<.05)
Cigarette use
BMI
Psychological
distress
.121
.505
Maladaptive
coping
.158
.156
.347 .603 .596
.487
.079
.472
.260
-.131
87

Summary results of tests of measurement invariance
Results for testing of measurement invariance for the low adaptive coping and high
adaptive coping groups are reported in Table 7. As is evident from the table, the
difference test (M1-M0) with 
2
(15)=137.565, p<.0001, rejected the null hypothesis that
the factor loadings are equal across the high adaptive coping and low adaptive coping
groups. The LM test demonstrated that ten out of a total of 15 factor loading constraints
were not appropriate and hence, needed to be released. Specifically, at the early
adolescence period the loadings of monthly, weekly and current cigarette use on the early
adolescent cigarette use construct, at the beginning of emerging adulthood the loadings of
weekly and past 24 hours cigarette use, at the mid-emerging adulthood the loadings of
past 24 hours cigarette use and the loadings of anxiety and somatic symptoms on distress,
and finally, at the early adulthood period the loadings of weekly and past 24 hours
cigarette use differed between the low adaptive and high adaptive coping groups. Once
these 10 constraints were released in the partially constrained factor loadings model
(M1*), the difference test (M1*-M0) with 
2
(5)=11.067, was non-significant thereby
indicating that the constraints imposed were adequate. Since a considerable number of
factor loadings (10 out of a total of 15) were significantly different between the two
groups, further tests of measurement invariance, specifically, testing of covariance and
regression estimates constraints were not preformed.

88

Table 7: Summary results of tests of measurement invariance for low vs. high adaptive
coping

Group 
2
df p NFI CFI RMSEA
Low adaptive coping
M
L0
660.241 214 .000 .971 .989 .047
M
L1
509.379 209 .000 .980 .999 .039
High adaptive coping
M
H0
698.126 214 .000 .956 .979 .054
M
H1
463.072 209 .000 .976 .999 .039
Combined
M0 972.459 418 .000 .978 .999 .039
M1 1110.024 433 .000 .973 .994 .043
M1-M0 137.565 15 <.0001
M1* 983.526 423 .000 .978 .999 .039
M1*-M0 11.067 5 .05
An asterisk indicated a model with relevant parameters partially constrained equal across
groups. M
M0
and M
F0
= basic theoretical models for low adaptive coping (N=937) and high
adaptive coping (N=783) groups respectively;M
M1
and M
F1
=modified models with correlated
errors suggested by the LM test on M
M0
and M
F0
, respectively M0=basic model combining low
and high adaptive coping groups, M1=model with all factor loadings constrained equal across
groups.
Note. p-values are based on a two tailed test.

Discussion
The basic objective of this study was to examine the potential role of
maladaptive coping as a mediator and adaptive coping as a moderator in the relation
between psychological distress and BMI. As hypothesized, first, maladaptive coping did
not operate as a mediator in the distress-BMI relationship as suggested by the structural
model (Figure 6b). Though psychological distress at the beginning of emerging adulthood
had a direct path to maladaptive coping at mid-emerging adulthood, however, there was
no significant association between maladaptive coping and BMI at early adulthood. This
significant relation between distress and maladaptive coping is consistent with findings
89
from previous studies in which an association between maladaptive coping and
depression has been noted (Seiffge-Krenke 2000). However, the hypothesized path
between maladaptive coping and BMI was not significant in the current study. Though
previous research has suggested a significant relation between maladaptive coping
strategies, such as emotion-focused and avoidance, this has been shown primarily in
bulimic subjects (Nagata 2000). In a recent study, Adams and Rini (Adams 2007) found
that women in the BMI gain group were more likely to use maladaptive coping strategies,
however, this study was cross-sectional in design.
The SEM (Figure 6a) generated for adaptive coping suggested a mediational role
for adaptive coping in the cigarette use-BMI relation. Specifically, cigarette use at the
beginning of emerging adulthood was associated with reduced adaptive coping skills at
mid-emerging adulthood which subsequently, predicted increased BMI at early
adulthood. Existing research has shown smoking to be positively related to negative or
maladaptive coping methods and negatively associated with positive or adaptive coping
skills such as parental support (Siqueira 2000) and physical activity (Klesges 1992), both
of which constituted two sub-constructs of the adaptive coping construct in the current
study. The negative path between adaptive coping and BMI found in the current study is
consistent with findings from previous studies which have shown both physical inactivity
(Ball 2001) and lack of parental support (Rohrer 2004) to be associated with increased
risk of obesity.
The second hypothesis that adaptive coping would moderate the distress-BMI
relation was supported. As evident from the multiple group analyses conducted for the
90
low adaptive and high adaptive coping groups, 10 out of a total of 15 factor loadings
were significantly different between the two groups thereby suggesting a difference in the
perception of measures used in the current study by these two groups. On the basis of the
statistical tests conducted, it can be concluded that the data were not invariant across the
low and high adaptive coping groups. This finding suggests the development and design
of different survey measures and items to be administered to the two adaptive coping
groups. To our knowledge, the potential role of adaptive coping as a moderator in the
distress-body weight relation has not been examined before.
However, the current results must be considered in light of study limitations. The
first limitation is that body weight measured as BMI is limited only to the early adulthood
period since drug abuse prevention rather than obesity was the focus of the original
parent MPP study. Overweight in childhood and adolescence has been shown to be a
strong predictor of both overweight and obesity later in adulthood (Guo 2000). Due to the
limitations of the MPP dataset, adolescent overweight status cannot be used as a covariate
in the current analyses.
A second limitation, common to many prevention studies, is that the study relied
primarily on self-report survey data. Specifically, with reference to BMI measurement,
which is computed using self-reported height and weight, it has been suggested that these
self-report statistics often yield inaccurate information, as men are more likely to
overestimate their height (Spencer 2002) while women have been shown to
underestimate their weight (Niedhammer 2000). However, as a benchmark diagnostic test
91
for measurement of body weight, BMI has been used extensively in the obesity field
(Neumark-Sztainer 2003).
A final limitation concerns the issue of timing of the measurement of the exposure
and the outcome in this study. For instance, items measuring psychological distress and
coping were not included in the survey since baseline and were measured starting from
the beginning of emerging adulthood. However, items such as distress were not the focus
of the original study or intervention, and much like other longitudinal studies, constructs
of interest were added to reflect developmental changes (Newcomb 1997). Additionally,
the temporal ordering of constructs in this study supports past research which has
suggested that temporally cigarette use starts as social experimentation and distress and
coping follow much later on.
Regardless of study limitations, findings from the current study suggest the need to
incorporate development of strong adaptive coping skills such as increase in physical
activity levels, strong support from both parents and peers, and relaxation techniques as
part of smoking prevention programs to subsequently prevent the development of
continued cigarette use, later distress and obesity risk. Further, the developmentally
staggered results demonstrate the potential need to focus primarily on cigarette use and
psychological distress in the early adolescent and adulthood years as risk factors to other
health behaviors such as obesity risk in later adulthood. Finally, the tests of measurement
invariance bring into light the development and design of different items measuring
cigarette use, distress and coping for the low and high adaptive coping groups studied.

92

Chapter 6: Discussion

Brief summary of findings with respect to each of the proposed five hypotheses

This dissertation attempts to potentially identify adolescence and adulthood
correlates for later obesity risk. Five hypotheses were proposed to be tested in the current
dissertation.

H1. Cigarette use, Psychological Distress, and BMI: Longitudinally, cigarette use
and psychological distress will show significant relations to early adulthood BMI:
Partially supported
As hypothesized, analysis on the MPP dataset suggests that cigarette use at mid-
emerging adulthood had a significant negative relation with BMI at early adulthood. This
inverse relation between smoking and BMI has been suggested in existing studies, mostly
cross-sectional in design, in which current smokers were found to have lower BMI than
non-smokers in adults (Lissner 1992; Klesges 1998; Bamia 2004; Akbartabartoori 2005;
Xu 2007). The current study extends previous research by demonstrating the significant
negative effect of continued or long-term cigarette use, starting from early adolescence to
mid-emerging adulthood, on early adult BMI.
However, the relation between psychological distress and later BMI was not
supported, though there was a significant positive correlation between distress and BMI
at early adulthood. Recent evidence from existing studies with regard to the depression-
BMI relation has been mixed. While six studies, to date, found that depression in
childhood, early and late adolescence significantly predicted the subsequent development
93
of obesity at 1-year follow-up (Goodman 2002), at 4-year follow-up (Stice 2005), in
emerging adulthood (Pine 2001; Richardson 2003), in young adulthood (Franko 2005)
and in mid-adulthood (Hasler 2005), two studies found no significant depression-obesity
relation (Pine 1997; Bardone 1998). One study showed a significant association with
depression and anxiety and later weight gain (Anderson 2006). Additionally, four studies
reported significant depression-obesity relations out of which two included only females
as their sample (Franko 2005; Stice 2005). Though the remaining two studies included
both males and females, however, a significant longitudinal relation between depression
and obesity was found only in females (Richardson 2003; Hasler 2005).
Several possible explanations have been suggested for this finding of gender
differences in the relation between depression and obesity. First, women with depression
report more symptoms of increased appetite and weight gain as compared to men (Carter
2000). Second, engagement in physical activity may differ by sex and may account for
the differences in the relation between depression and obesity. Research has shown
physical activity to decrease with age for

males and females, but the decrease has been
suggested to be more profound

in females (Dovey 1998). Finding of a non-significant
relation between psychological distress and BMI in the current study could be possibly
due to analysis of this relation on a population-based sample in which a small proportion
of subjects is ―distressed‖ as compared to an at-risk population.

94
H2: Gender as a moderator: Gender will moderate the cigarette use-distress-BMI
relation such that there will be stronger associations of distress, cigarette use, and BMI
for females compared to males: Partially supported.

Based on the findings from the multiple group model generated in Study 2, females
did not show significantly stronger relations between cigarette use and BMI associations
as compared to males. Previous research has shown smoking to have a similar significant
negative association with body weight in both genders (Rasmussen 2003; Bamia 2004).
However, in this study it appears that females did have stronger association between
cigarette use and psychological distress than males. This finding supports previous
research in which gender differences in the association between cigarette smoking and
elevated depressive symptoms was reported (Poulin 2005). Further, findings from the
current study show significant gender differences in the relation between physical activity
and BMI with males reporting a positive relation between physical activity and BMI
while females reported a negative relation. There was also a difference in the negative
relation between self-rated health and BMI was also found in the study with females
reporting a stronger negative estimate as compared to males. Finally, the positive
correlation between physical activity and self-rated health was stronger for males as
compared to females.
To the best of my knowledge, gender differences have previously been studied in a
longitudinal study specifically with regard to both the relation between self-rated health
and BMI and the association of physical activity and self-rated health. Further, the
95
finding of significant gender difference in the direction of relation between physical
activity and BMI is interesting and several possible explanations can be suggested to
support it. First, the positive association between physical activity and BMI in males
could be possibly due to BMI being a less accurate predictor of body fatness. Recent
evidence on BMI as a surrogate measure of adiposity suggests that it may provide
misleading information in muscular (not fat) individuals, specifically athletes (Prentice
2001). Other inexpensive measures of body fat such as waist circumference, waist-hip
ratio (WHR) and even bioimpedence could be coupled together and used as proxies for
body fatness. Second, previous research has shown physical activity to decrease with age
for

males and females, but the decrease has been suggested to be more profound

in
females (Dovey 1998). Thus, this could imply that females are possibly not exercising as
much as men and thus, increased BMI might be an indicator of overweight rather than
muscle fat as reported in men.

H3: Subjective rating of health: Poor subjective rating of health at mid-emerging
adulthood will have a significant relation to BMI at early adulthood: Supported
Based on analyses on the dataset, it does appear that self-rated health at mid-
emerging adulthood had a direct path to later BMI in early adulthood. Further, the
multiple group model generated for Study 2 demonstrated significant gender differences
in the relation between self-rated health and BMI, with females reporting stronger
negative association between self-rated health and BMI than males. Evidence from a
recent cross-sectional study reported significant inverse associations between BMI and
96
self-rated health in adults aged 18 years or older (Lim 2007). Additionally, in the older
adult population, poor self-rated health has shown to be a risk factor for obesity (Kaplan
2003). Though the relation between self-rated health and obesity has been examined
primarily in the older adult population, this finding of a significant negative relation
between self-rated health and BMI in the current study suggests the possibility of a
temporal relation between these two variables, whereby poor self-rated could be studied
as a precursor to later obesity risk in a much younger population of adults aged 30 years.

H4a: Coping as a mediator: Maladaptive or non-productive coping strategies
(composed of drug use, avoidance, and negative affect coping items) at mid-emerging
adulthood will mediate the beginning of emerging adulthood distress-early adulthood
BMI relation: Not Supported.
As demonstrated in the structural model (Figure 6b) generated for Study 3,
maladaptive coping did not mediate the relationship between distress at the beginning of
emerging adulthood and BMI at early adulthood. The structural model demonstrated that
though psychological distress at the beginning of emerging adulthood had a direct path to
maladaptive coping at mid-emerging adulthood, however, there was no significant
association between maladaptive coping and BMI at early adulthood. This significant
relation between distress and maladaptive coping is consistent with findings from
previous studies in which an association between maladaptive coping and depression has
been noted (Seiffge-Krenke 2000). However, the hypothesized path between maladaptive
coping and BMI was not significant in the current study. Though previous research has
97
suggested a significant relation between maladaptive coping strategies, such as emotion-
focused and avoidance, this has been shown primarily in bulimic subjects (Nagata 2000).
Thus, evidence from longitudinal studies conducted on non-clinical population (i.e.
without any eating disorders) on the relation between maladaptive coping and body
weight seems to be limited. A possible explanation for a non-significant finding between
maladaptive coping and BMI in the current study could be due to the fact that
maladaptive coping did not include direct measures of unhealthy eating to cope with a
stressful situation (e.g. ―When I am stressed, I tend to overeat/eat unhealthy‖ or When I
am stressed, I don‘t think what I am eating‖, etc.).
On the other hand, findings from Study 3 demonstrated that adaptive coping,
comprised of physical activity, relaxation, and support items, mediated the relation
between cigarette use at the beginning of emerging adulthood and BMI at early
adulthood. Existing research has shown smoking to be positively related to negative or
maladaptive coping methods and negatively associated with positive or adaptive coping
skills such as parental support (Siqueira 2000) and physical activity (Klesges 1992), both
of which constituted two of the three sub-constructs of the adaptive coping construct in
the current study. The negative path between adaptive coping and BMI found in the
current study is consistent with findings from previous studies which have shown both
physical inactivity (Ball 2001) and lack of parental support (Rohrer 2004) to be
associated with increased risk of obesity.
Thus, the current study adds to existing research by exploring the potential role of
coping (both adaptive and maladaptive) in the relations between the constructs of
98
interests and infact demonstrated a significant mediating effect of adaptive coping in the
cigarette use-BMI relation and in the cigarette use-distress relation.

H4b: Coping as a moderator: Adaptive or productive coping strategies (composed
of physical activity, support, and relaxation items) will moderate the distress-BMI
relation: Supported.
Based on the multiple group analyses conducted in Study 3, 10 out of a total of 15
factor loadings tested were different between the low adaptive coping and high adaptive
coping groups, thereby demonstrating the moderating role of adaptive coping in the
relation between the modeled constructs. To our knowledge, the potential role of adaptive
coping as a moderator in the cigarette use-distress-BMI relations has not been examined
before. What would be interesting would be to further explore the moderating effect of
the two groups-high adaptive coping and high maladaptive coping on the relations
between these constructs.

H5: Psychological distress and Physical activity: Distress at beginning of
emerging adulthood will have a negative path to physical activity at mid-emerging
adulthood, which will then have a significant negative association with distress at early
adulthood: Supported.
As hypothesized, psychological distress at the beginning of emerging adulthood had
a negative path to physical activity levels at mid-emerging adulthood, which
subsequently had a negative relation to distress at early adulthood (Refer to Appendix A).
99
Previous research has consistently shown that participation in physical activity is
associated with positive mood (Williamson 2001) as well as reduced physical activity
levels to be cross-sectionally (Goodwin 2003; Brosnahan 2004) and prospectively
associated with an increased risk of later depression (Farmer 1988; Camacho 1991;
Lampinen 2000; Beard 2007). Additionally, exercise has been suggested as an effective
intervention for reducing symptoms of depression and anxiety (Zoeller 2007). Thus, the
current study adds to existing research by simultaneously evaluating the direct paths from
distress to physical activity and physical activity to distress. Finding of significant
relation between distress and physical activity suggests important implications for current
prevention programs for e.g. efforts to reduce or prevent depression might include an
additional component to increase physical activity levels thereby preventing the
subsequent recurrence of later depression.

Overall implications of current findings to prevention programs
Based on the findings from each of the three studies conducted and the five
hypotheses tested, the following recommendations can be made for designing effective
prevention programs for cigarette use, psychological distress, and obesity.
1.Study 1: In this study continued cigarette use beginning from early adolescence to
mid-emerging adulthood had an inverse relation with BMI in early adulthood. This
protective effect of smoking on BMI poses a conundrum in research because even though
smoking may help weight control, however, the health risks associated with smoking are
well-documented in the literature. As part of smoking prevention programs itself, this
100
weight control ―benefit‖ of cigarette use on body weight will have to be counteracted and
de-emphasized by focusing on other healthy ways to lose or maintain weight such as
healthy diet and exercise.
Findings from this study also demonstrated the well-established relation in
literature between early cigarette use and later psychological distress, primarily
depression. In light of this finding, cigarette use prevention programs should incorporate
components that focus on reducing later psychological distress. Though distress did not
have a significant relation to later BMI in either Study 1 or in the Appendix A, however,
it did consistently show direct paths to later physical activity and self-rated health. In
terms of prevention of later obesity risk, these developmentally staggered results suggest
the need to focus on reducing distress thereby resulting in increased physical activity
levels and perceptions of self-rated health and subsequently reducing obesity risk. The
finding of a significant negative relation between physical activity to both psychological
distress and BMI reiterates the strong emphasis on exercise as an effective intervention
strategy for prevention of both distress and obesity. The positive association of poor self-
rated health with later BMI from Appendix Figure 1 suggests the potential use of these
health perceptions as indicators of later obesity risk. Subjective rating of health as an
indicator would be both easy and inexpensive to measure and could be possibly used at
an early developmental period as a predictor of later obesity risk. Individuals who
perceive their health to be poor could be administered effective intervention strategies
such as increased physical activity levels to prevent the occurrence of later obesity.
101
2.Study 2: Findings from Study 2 illustrated certain differences in relations of
cigarette use, psychological distress, physical activity, and BMI between males and
females. The positive association of cigarette use and distress in females suggest that
prevention programs for tobacco use be supplemented with distress prevention training
for females. Further, as suggested in Study 2, the positive relation between physical
activity and BMI in adult males could be possibly due to BMI being a less accurate
predictor of body fatness, especially in muscular (not fat) males such as athletes. This
finding demonstrates the need for utilizing as well as exploring more accurate measures
for measuring body fat in the area of obesity research especially for males. On the other
hand, the negative relation between physical activity and BMI in females supports the
consistent findings of low physical activity as a major proximal predictor of obesity risk.
As noted earlier, research has shown physical activity to decrease with age for

males and
females, but the decrease has been suggested to be more profound

in females (Dovey
1998). Based on this evidence, females are possibly not exercising as much as men and
thus, increased BMI might be an indicator of overweight rather than muscle fat as
reported in men.
3.Study 3: The significant finding of adaptive coping as a mediator in the cigarette
use-BMI relation in Study 3 suggests the need to incorporate development of strong
adaptive coping skills such as increase in physical activity levels, strong support from
both parents and peers, and relaxation techniques as part of smoking prevention programs
to subsequently prevent the development of later obesity risk. Since the tests of
measurement invariance between the low adaptive coping and high adaptive coping
102
groups demonstrated that the overall data were not invariant across the two groups, the
development and design of different items measuring cigarette use, distress and coping
for the two groups studied needs to be emphasized. Additionally, another possible
explanation for the lack of invariance could be perhaps that there is still something
unmeasured that is driving the differences between the low adaptive and high adaptive
coping groups. Future analyses evaluating the moderating effect of adaptive coping (high
adaptive coping vs. low adaptive coping) could control for the potential effect of
maladaptive coping to see if the results still hold.

Strengths and limitations of the current dissertation
The major strength of this study is that it potentially explores multiple correlates to
obesity risk (such as cigarette use, distress, self-rated health, and coping) that heretofore
have been studied in a limited fashion in prevention research. Hence, findings from this
study in general will inform prevention program developers in the design of programs
that target multiple health behaviors or constructs. Findings yield several concrete
possibilities for adding content to existing prevention programs, creating boosters to
those programs or even developing complementary prevention programs for instance
efforts aimed at preventing or reducing depression should include a component that
emphasizes increased physical activity which will subsequently assist in the prevention of
later obesity risk. A further strength is the evaluation of the relations between these
constructs over multiple developmental periods, beginning in early adolescence to early
103
adulthood, thereby informing researchers about the dynamics of these relationships in
different populations.
However, findings from this study should be considered in light of the following
major limitations:

1. First, in the MPP dataset the measurement of BMI is restricted only to the early
adulthood period thereby not allowing the possibility of controlling for BMI as a
covariate in the earlier waves. In research, overweight/obesity at an earlier developmental
period, specifically childhood and adolescence, has been shown to be a strong predictor
of overweight/obesity in adulthood (Guo 2000).This severely limits the possibility of
drawing any causality or temporality between the constructs studied and BMI due to the
non-existence of BMI measure on any wave studied including baseline. However, the
data are drawn from a randomized trial which implies that any biases affecting the
relations between these constructs studied are distributed randomly across the population-
based sample. What can be claimed from current findings is the existence of significant
relations between BMI and cigarette use, and self-rated health, thereby drawing attention
of researchers to these relatively unexplored constructs and their relation to obesity risk.

2. Another significant limitation concerns the issue of temporal ordering of constructs as
restricted by the nature of the MPP dataset. For instance, items measuring psychological
distress and coping were measured starting from the beginning of emerging adulthood,
while self-rated health and physical activity items were measured from mid-emerging
104
adulthood. Since these items were not present since baseline, these measures could not be
modeled in the current study at each wave used in analyses, thereby restricting the scope
of current findings in terms of inferring causal and temporal relations. Thus, in the
current study, inter-relations between these constructs can be simply evaluated as
significant or non-significant and no claims to causality can be made.
Since the original scope and design of the trial focused on drug use, and not on
psychological distress or obesity, this could be considered a special case of planned
missingness (Graham 2006). Planned missingness occurs when an investigator chooses
to collect partial data using a carefully controlled missing data design. According to
Graham and colleagues, planned missingness is usually conducted to mange costs,
manage data quality, and lighten the burden on respondents. Planned missingness results
in collection of smaller and more manageable amount of data from which reasonable
conclusions can be drawn. Additionally, since the MPP was primarily developed and
designed as a drug abuse prevention trial, items measuring depression, coping, physical
activity, and BMI were added as they became nationally important issues and were added
to reflect developmental changes (Newcomb 1997) .

3. Third limitation concerning the comprehensiveness of measures both with regard to
the single item used to measure subjective rating of health as well as the two items
representing physical activity. Though the use of a single-item question measuring self-
rated health has warranted caution however, self-rated health has also been demonstrated
to be a valid measure for measuring health status in adolescents (Agyemang 2006) as
105
well as adults (Gregor 2006). The two physical activity items were taken from the 2005
Youth Behavior Risk Survey developed by the CDC (CDC 2005) and moderate reliability
of these items has been reported in measuring moderate to vigorous physical activity
(Troped 2007).

4. A fourth limitation, common to many prevention studies, is that the study relied
primarily on self-report survey data. Specifically, with reference to BMI measurement,
which was computed using self-reported height and weight in this study, it has been
suggested that these self-report statistics often yield inaccurate information, as men are
more likely to overestimate their height (Spencer et al., 2002) while women have been
shown to underestimate their weight (Niedhammer et al., 2000). However, as a
benchmark diagnostic test for measurement of body weight, BMI has been used
extensively in the obesity field (Neumark-Sztainer et al., 2003).
Despite these limitations, findings from this study have important implications for
designing future prevention programs that go beyond cigarette use, psychological
distress, or obesity and target multiple constructs. Additionally, developmentally
staggered results from this study will provide an opportunity to identify targets for
intervention at each developmental period studied.

Overall conclusions
I believe that the overarching issue driving these inter-relations between cigarette
use, psychological distress, physical activity, coping, self-rated health, and BMI is the
106
concept of body image dissatisfaction. Body dissatisfaction can be associated with each
of these constructs developmentally. At early adolescence, cigarette use is usually
thought of being undertaken for social experimentation as well as a weight control
strategy by female adolescents who are dissatisfied by the way their bodies look. In terms
of depression and body image dissatisfaction, I think a bi-directional relation exists
(which can be supported by literature), whereby depression can lead to dissatisfaction
with one‘s body and body dissatisfaction can lead to depression, specifically for females.
Body dissatisfaction certainly propels individuals to get involved in physical activity for
improvement of body image. There seems to be a logical relation between maladaptive
coping and body dissatisfaction, whereby dissatisfaction with the way one‘s body looks
can make them indulge in maladaptive coping strategies such as overeating to provide
comfort. Finally, obesity can result in increased body dissatisfaction in an individual.
Keeping this overarching issue of body dissatisfaction in perspective while
understanding these relations, adolescence would probably be the period when its effect
would be most pronounced, specifically among females, thereby making the inter-
relations between these constructs stronger during that period. What would be interesting
to understand is whether there is an increase or decrease in levels of body dissatisfaction
in other developmental periods after early adolescence.
From a theoretical perspective, primarily intrapersonal or P-level theories (Pentz
1999) seem to be most instructive in explaining the dynamics in these relations between
cigarette use, depression, physical activity, maladaptive coping, self-rated health, and
BMI. Adaptive coping is the only construct operating at the interpersonal or S-level as it
107
involves getting support from parents, peers etc. thereby involving the interaction of the
individual with the environment. Some of the P-level constructs that are operating these
models are affect (depression) and maladaptive coping.
An interpersonal model, which can be drawn upon in explaining these relations
though only in adolescents is Jessor‘s problem-behavior theory (Jessor 1991). According
to Jessor, there is a clustering of risk factors during adolescence such as smoking, and
dieting as they serve similar functions for adolescents such as independence from adults,
rebelliousness, etc. (French 1996). From a problem-behavior perspective, adolescent
cigarette smoking in the current model can viewed as a normative transgression
motivated by goals such as rejecting the norms of conventional society, affirming
membership in a peer group, or asserting independence from parents.
Other plausible theoretical models, which can be put forward in understanding the
dynamics operating between these constructs is the transactional model of stress and
coping. In research, the Transactional Model of Stress and Coping has been used as a
framework for evaluating the processes of coping with stressful events (Lazarus 1977;
Cohen 1984). According to this model, stressful experiences are construed as person-
environment transactions. These transactions are mediated by firstly the person‘s
appraisal of the stressor as stressful, controllable, positive etc. (‗primary appraisal) and
secondly on the social and cultural resources at his or her disposal i.e. ‗secondary
appraisal‘. Actual coping efforts aimed at regulation of the problem result in outcomes of
the coping process such as emotional well-being and development of health behaviors.
Applied this theory to the current model, stressful negative life events experienced by an
108
individual such as divorce, death, etc. can result in adoption of either maladaptive or
adaptive coping strategies to deal with that stressor depending both on the personality of
the individual as well as the resources available to the individual to deal with the stressor.
Adoption of maladaptive coping strategies can result in negative health outcomes such as
cigarette use, psychological distress and can even lead to obesity risk. On the other hand,
engagement in adaptive coping strategies can reduce the likelihood of each of these
adverse health outcomes.
From a developmental perspective, identification of important developmental tasks,
challenges and milestones, and the competency to achieve these at each stage of life of an
individual is imperative. One such developmental model is the identity developmental
theory as proposed by Eric Erickson (Erickson 1963). The core of Erickson‘s theory is
that each developmental crisis associated with each stage must be resolved at each
successive stage of development. Further, if these tasks are not mastered at the
appropriate stage of development, the individual will be at a disadvantage in making
subsequent adjustments. For instance, in the adolescent period, the major crisis
confronting an adolescent is establishing his or her own identity and avoiding identity
diffusion. In this developmental period, an adolescent may view smoking, or the use of
alcohol and other drugs, as a way of expressing a sense of independence. Identity
developmental theory as applied to the current model would suggest that adolescents who
exhibit negative identities and deviant behavior such as being sexually active, smoking,
and drinking alcohol continue with these negative behaviors which translate into other
109
maladjustments such as depression, continued cigarette use, etc. as the adolescent
progresses through other stages of life.

Future research
Future research should attempt to replicate similar longitudinal models in children
and older adults to get a comparative perspective of the relations between these
constructs across different populations. Apart from physical inactivity, another traditional
risk factor for obesity risk cited frequently in research is unhealthy dietary choices or
practices (Nicklas 2003; St-Onge 2003). Future research should incorporate poor food
choice items, possibly as part of maladaptive coping, and examine its relation to other
constructs modeled in this study. Further, innovative approaches to obesity prevention
should focus on improving poor decision making (Davis 2004) as well as low impulse
control (Holtkamp 2004; Hubel 2006) specifically in terms of food choices. For instance,
interaction of both poor-decision making and low impulse control can result in possibly
repetitive sequences of overeating, eating non-healthy foods, choosing immediate
sedentary activities (e.g. TV watching) instead of healthier alternatives and even
maintenance of these unhealthy patterns, thereby increasing the risk of obesity.
110

References

Adams, T., Rini, A. (2007). "Predicting 1-year change in body mass index among college
students." Journal of American College Health 55(6): 361-5.

Agyemang, C., Denktas, S, Bruijnzeels, M, Foets, M. (2006). "Validity of the single-item
question on self-rated health status in first generation Turkish and Moroccans
versus native Dutch in the Netherlands." Public Health 120(6): 543-50.

Ahmad, K., Jafar, T.H., Chaturvedi, N. (2005). "Self-rated health in Pakistan: results of a
national health survey." BMC Public Health 5(51).

Akbartabartoori, M., Lean, M.E., Hankey, C.R.. (2005). "Relationships between cigarette
smoking, body size and body shape." International Journal of Obesity 29(2): 236-
43.

Allison, S., Roeger, L., Martin, G., Keeves, J. (2001). " Gender differences in the
relationship between depression and suicidal ideation in young adolescents." Aust
N Z J Psychiatry 35(4): 498–503.

Anderson, S. E., Cohen, P., Naumova, E.N., Must, A (2006). "Association of depression
and anxiety disorders with weight change in a prospective community-based
study of children followed up into adulthood." Archives of Pediatrics &
Adolescent Medicine 160(3): 285-91.

Aneshensel, C. S., Huba, G.J. (1983). "Depression, alcohol use, and smoking over one
year: a four-wave longitudinal causal model. ." J Abnormal Psychol 92: 134-150.

Appels, A., Bosma, H., Grabauskas, V. (1996). "Self-rated health and mortality in a
Lituanian and a Dutch population." Soc Sci Med 42: 681-689.

Arbuckle, J. L. (2003). Amos 5.0 Update to the Amos User's Guide, Smallwaters
Corporation, Chicago IL.

Aung, A. T., Hickman, N.J., & Moolchan, E.T. (2003). " Health and performance related
reasons for wanting to quit: gender differences among teen smokers." Substance
Use & Misuse 38(8): 1095-107.

Austin, S. B., Gortmaker, S.L. (2001). "Dieting and smoking initiation in early adolescent
girls and boys: a prospective study." Am J Public Health. 91(3): 446-50.

111
Ball, K., Owen, N., Salmon, J., Bauman, A., Gore, C.J. (2001). "Associations of physical
activity with body weight and fat in men and women." International Journal of
Obesity Related Metabolic Disorders 25(6): 914-19.
Bamia, C., Trichopoulou, A., Lenas, D., Trichopoulos, D. (2004). "Tobacco smoking in
relation to body fat mass and distribution in a general population sample." Int J
Obes Relat Metab Disord. 28(8): 1091-6.

Bardone, A. M., Moffitt, T.E., Caspi, A., Dickson, N., Stanton, W.R., Silva, P.A. (1998).
"Adult physical health outcomes of adolescent girls with conduct disorder,
depression, and anxiety." Journal of American Academy of Child Adolescence
Psychiatry 37(6): 594-601.

Barefoot, J. C., Heitmann, B.L., Helms, M.J., Williams, R.B., Surwit, R.S., Siegler, I.C.
(1998). "Symptoms of depression and changes in body weight from adolescence
to mid-life." Int J Obes Relat Metab Disord. 22: 688-694.

Bartal, M. (2001). "Health effects of tobacco use and exposure. ." Monaldi Arch Chest
Dis 56(6): 545-554.

Baskin, M. L., Ard, J., Franklin, F., Allison, D.B. (2005). "Prevalence of obesity in the
United States." Obesity Reviews 6(1): 5-7.

Basterra-Gortari, F. J., Bes-Rastrollo, M., Forga, L., Martínez, J.A., Martínez-González,
M.A. (2007). "Validity of self-reported body mass index in the National Health
Survey." An Sist Sanit Navar. 30(3): 373-81.

Beard, J. R., Heathcote, K., Brooks, R., Earnest, A., Kelly, B. (2007). "Predictors of
mental disorders and their outcome in a community based cohort." Soc Psychiatry
Psychiatr Epidemiol. 42(8): 623-630.

Bentler, P. M. (1995). EQS Structural Equations Program Manual. Encino, CA,
Multivariate Software, Inc.

Berkey, C. S., Rockett, H. R., Field, A. E., Gillman, M. W., Frazier, A. L., Camargo, C.
A., & Colditz, G. A. (2000). "Activity, dietary intake, and weight changes in a
longitudinal study of preadolescent and adolescent boys and girls." Pediatrics 105:
E56.

Berlin, I., Singleton, E.G., Pedarriosse, A.M., Lancrenon, S., Rames, A., Aubin, H.J.,
Niaura, R. (2003). "The Modified Reasons for Smoking Scale: factorial structure,
gender effects and relationship with nicotine dependence and smoking cessation
in French smokers." Addiction 98(11): 1575-83.

112
Biro, F. M., McMahon, R.P., Striegel-Moore, R., Crawford, P.B., Obarzanek, E.,
Morrison, J.A., Barton, B.A., Falkner, F. (2001). "Impact of timing of pubertal
maturation on growth in black and white female adolescents: The National Heart,
Lung, and Blood Institute Growth and Health Study." Journal of Pediatrics
138(5): 617-618.

Boon, C. S., Clydesdale, F.M. (2005). "A review of childhood and adolescent obesity
interventions." Crit Rev Food Sci Nutr. 45(7-8): 511-25.

Born, L., Shea, A., Steiner, M. (2002). "The roots of depression in adolescent girls: Is
menarche the key?" Curr Psychiatry Rep 4(6): 449-60.

Borrelli, B., Hogan, J.W., Bock, B., Pinto, B., Roberts, M., Marcus, B. (2002).
"Predictors of Quitting and Dropout Among Women in a Clinic-Based Smoking
Cessation Program " Psychology of Addictive Behaviors 16(1): 22-27.

Boyle, P. (1997). "Cancer, cigarette smoking and premature death in Europe: a review
including the Recommendations of European Cancer Experts Consensus Meeting,
Helsinki, October 1996." Lung Cancer 17(1): 1-60.

Brener, N. D., McManus, T., Galuska, D.A. (2003). "Reliability and validity of self-
reported height and weight among high school students." Journal of Adolescent
Health. 32: 281-287.

Brent, D. A., Holder, D., Kolko, D., Birmaher, B., Baugher, M., Roth, C., Iyengar, S.,
Johnson, B.A. (1997). "A clinical psychotherapy trial for adolescent depression
comparing cognitive, family, and supportive therapy." Archives of General
Psychiatry 54(9): 877-85.

Breslau, N. (1995). "Psychiatric comorbidity of smoking and nicotine dependence."
Behavior Genetics 25(2): 95-101.

Breslau, N., Johnson, E.O. (2001). "Predicting smoking cessation and major depression
in nicotine-dependent smokers." Am J Public Health. 90(7): 1122-7.

Breslau, N., Kilbey, M.M., Andreski, P. (1993). "Nicotine dependence and major
depression. New evidence from a prospective investigation." Arch Gen
Psychiatry. 50(1): 31-5.

Breslau, N., Peterson, E.L., Schultz, L.R., Chilcoat, H.D., Andreski, P. (1998). "Major
depression and stages of smoking. A longitudinal investigation.
." Arch Gen Psychiatry. 55(2): 161-66.

113
Broder, B. I. (1995). Adolescent stress and styles of coping: Predictors and moderators of
psychological distress. Doctoral dissertation, University of Southern California.

Brook, D. W., Brook, J.S., Zhang, C., Cohen, P., Whiteman, M. (2002). "Drug use and
the risk of major depressive disorder, alcohol dependence, and substance use
disorders." Arch Gen Psychiatry. 59(11): 1039-44.

Brook, J. S., Cohen, P., Brook, D.W. (1998). "Longitudinal study of co-occurring
psychiatric disorders and substance use." J Am Acad Child Adolesc Psychiatry.
37(3): 322-30.

Brook, J. S., Schuster, E., Zhang, C. (2004). "Cigarette smoking and depressive
symptoms: a longitudinal study of adolescents and young adults." Psychological
Rep. 95(1): 159-66.

Brosnahan, J., Steffen, L.M., Lytle, L., Patterson, J., Boostrom, A. (2004). "The relation
between physical activity and mental health among Hispanic and non-Hispanic
white adolescents." Arch Pediatr Adolesc Med. 158(8): 818-23.

Brown, R. A., Lewinsohn, P.M., Seeley, J.R., Wagner, E.F. (1996). "Cigarette smoking,
major depression, and other psychiatric disorders among adolescents." J Am Acad
Child Adolesc Psychiatry. 35(12): 1602-10.

Burke, V., Milligan, R.A., Beilin, L.j., Dunbar, D., Spencer, M., Balde, E., & Gracey,
M.P. (1997). "Clustering of health-related behaviors among 18-year-old
Australians." Prev Med 26(724-733).

Cairney, J., Wade, T.J. (1998). "Correlates of body weight in the 1994 National
Population Health Survey." International Journal of Obesity Related Metabolic
Disorders, 22(6): 584-591.

Calfas, K., Taylor, W.C. (1994). "Effects of physical activity on psychological variables
in adolescence." Pediatr Exerc Sci. 6: 406-423.

Camacho, T. C., Roberts, R.E., Lazarus, N.B., Kaplan, G.A., Cohen, R.D. (1991).
"Physical activity and depression: evidence from the Alameda County Study."
American Journal of Epidemiology 134(2): 220-31.

Camp, D. E., Klesges, R.C., Relyea, G. (1993). "The relationship between body weight
concerns and adolescent smoking." Health Psychology 12(1): 24-32.

Carmody, T. P. (1989). "Affect regulation, nicotine addiction, and smoking cessation, ."
Journal of Psychoactive Drugs, 21: 331-342.

114
Carter, J. D., Joyce, P.R., Mulder, R.T., Luty, S.E., and McKenzie, J. (2000). "Gender
differences in the presentation of depressed outpatients: a comparison of
descriptive variables." J. Affect. Disord. 61: 59-67.

Catenacci, V. A., Wyatt, H.R. (2007). "The role of physical activity in producing and
maintaining weight loss." Nat Clin Pract Endocrinol Metab. 3(7): 518-29.

Cawley, J., Markowitz, S., Tauras, J. (2004). "Lighting up and slimming down: the
effects of body weight and cigarette prices on adolescent smoking initiation." J
Health Econ. 23(2): 293-311.

CDC (2005) "Youth Risk Behavior Survey." Volume, DOI:

Chan, D. W. (1995). "Depressive symptoms and coping strategies among Chinese
adolescents in Hong Kong." Journal of Youth and Adolescence 24: 267-279.

Chandola, T., & Jenkinson, C. (2000). "Validating self-rated health in different ethnic
groups." Ethnic Health 5(2): 151-9.

Chen, K., & Kandel, D.B. (1995). "The natural history of drug use from adolescence to
the mid-thirties in a general population sample. ." American Journal of Public
Health 85: 41-47.

Chi, I., Yip, P.S., Chiu, H.F., Chou, K.L., Chan, K.S., Kwan, C.W., Conwell, Y., Caine,
E. (2005). "Prevalence of Depression and Its Correlates in Hong Kong's Chinese
Older Adults." American Journal of Geriatric Psychiatry 13(5): 409-16.

Choi, W. S., Patten, C.A., Gillin, J.C., Kaplan, R.M., Pierce, J.P. (1997). "Cigarette
smoking predicts development of depressive symptoms among U.S. adolescents."
Annals of Behavioral Medicine 19(1): 42-50.

Chung, H., Park, Y., Lanza, S.T. (2005). " Latent transition analysis with covariates:
pubertal timing and substance use behaviours in adolescent females." Statistics in
Medicine 24(18): 2895-910.

Clarke, G. N., Hawkins, W., Murphy, M., Sheeber, L.B., Lewinsohn, P.M., Seeley, J.R.
(1995). "Targeted prevention of unipolar depressive disorder in an at-risk sample
of high school adolescents: a randomized trial of a group cognitive intervention."
Journal of American Academy of Child Adolescence Psychiatry 34(312-21).

Cohen, F. (1984). Coping. Behavioral Health: A Handbook of Health Enhancement and
Disease Prevention. . J. A. H. S.M. Weiss, N.E. Miller & S.M. Weiss. New York,
Wiley.

115
Cohen, S., Schwartz, J.E., Bromet, E.J., Parkinson, D.K. (1991). "Mental health, stress,
and poor health behaviors in two community samples." Prev Med. 20(2): 306-15.

Compas, B. E., Connor-Smith, J.K., Saltzman, H., Harding, A., Wadsworth, M.E. (2001).
"Coping with stress during childhood and adolescence problems, progress, and
potential in theory and research." Psychological Bulletin 127(1): 87-127.

Cooper, T. V., Klesges, R.C., Robinson, L.A., Zbikowski, S.M. (2003). "A prospective
evaluation of the relationships between smoking dosage and body mass index in
an adolescent, biracial cohort." Addict Behav. 28(3): 501-12.

Crawford, A. M., Pentz, M.A., Chou, C.P., Li, C., Dwyer, J.H. (2003). "Parallel
developmental trajectories of sensation seeking and regular substance use in
adolescents." Psychol Addict Behav 17(3): 179-92.

Crawley, H. F., While, D. (1995). "The diet and body weight of British teenage smokers
at 16-17 years." Eur J Clin Nutr. 49(12): 904-14.

Crisp, A. H., Halek, C., Sedgewick, P., Stravraki, C., Williams, E., & Kiossis, I. (1998).
"Smoking and pursuit of thinness in schoolgirls in London and Ottawa."
Postgraduate Medical Journal 74: 473-479.

Crocker, P., Kowalski, N., Kowalski, K., Chad, K., Humbert, L., Forrester, S. (2001).
"Smoking behaviour and dietary restraint in young adolescent women: the role of
physical self-perceptions." Can J Public Health. 92(6): 428-432.

Davis, D., Levitan, R.D., Muglia, P., Bewell, C., Kennedy, J.L. (2004). "Decision-
making deficits and overeating: A risk model for obesity. ." Obesity Research 12:
929-935.

Deckelbaum, R. J., Williams, C.L. (2001). "Childhood obesity: the health issue." Obesity
Research 9(Suppl 4): 239S-243S.

Derogatis, L. R., Lipman, R.S., & Covi, L. (1973). "SCL-90: an outpatient psychiatric
rating scale-preliminary report." Psychopharmacology Bulletin 9(13-27).

Dietz, W. H. (1998). "Health consequences of obesity in youth: Childhood predictors of
adult disease." Pediatrics 101: 518-525.

Difranza, J. R., Riggs, N. R., Pentz, M. A. (in press). "Time to re-examine old definitions
of nicotine dependence. ." Nicotine and Tobacco Research.

116
DiPietro, L., Anda, R.F., Williamson, D.F., Stunkard, A.J. (1992). "Depressive symptoms
and weight change in a national cohort of adults." Int J Obes Relat Metab Disord.
16: 745-753.

Dixon, J. B., Dixon, M.E., O'Brien, P.E. (2003). "Depression in association with severe
obesity." Archives of Internal Medicine 163: 2058-2065.

Dovey, S. M., Reeder, A.I., Chalmers, D.J. (1998). "Continuity and change in sporting
and leisure time physical activities during adolescence." Br J Sports Med. 32: 53-
57.

Drøyvold, W. B., Holmen, J., Midthjell, K., Lydersen, S. (2004). "BMI change and
leisure time physical activity (LTPA): an 11-y follow-up study in apparently
healthy men aged 20-69 y with normal weight at baseline." International Journal
of Obesity Related Metabolic Disorders 28(3): 410-7.

Duncan, B., Rees, D.I. (2005). "Effect of smoking on depressive symptomatology: a
reexamination of data from the National Longitudinal Study of Adolescent
Health." Am J Epidemiol. 62(5): 461-70.

DuRant, R. H., Smith, J.A., Kreiter, S.R., & Krowchuk, D.P. (1999). "The relationship
between early age of onset of initial substance use and engaging in multiple health
risk behaviors among young adolescents." Archives of Pediatric Adolescent
Medicine 153: 286-291.

Ebata, A. T., Moos, R. H. (1994). "Personal, situational, and contextual correlates of
coping in adolescence." Journal of Research on Adolescence 4: 99-125.

Ebata, A. T., Moos, R. H. (1991). "Coping and adjustment in distressed and healthy
adolescents." Journal of Applied Psychology 17: 33-54.

Ebeling, H., Moilanen, I., Linna, S.L., Tirkkonen, T., Ebeling, T., Piha, J., Kumpulainen,
K., Rasanen, E., Tamminen, T., Almqvist, F. (1999). "Smoking and drinking
habits in adolescence--links with psychiatric disturbance at the age of 8 years."
Eur Child Adolesc Psychiatry. 8(Suppl. 4): 68-76.

Erickson, E. H. (1963). Childhood and Society. New York, Norton.

Escobedo, L. G., Anda, R. F., Smith, P. F., Remington, P.L., Mast, E.E. (1990).
"Sociodemographic characteristics of cigarette smoking initiation in the United
States. Implications for smoking prevention policy. ." The Journal of the
American Medical Association 264(12): 1550-1555.

117
Escobedo, L. G., Reddy, M., & DuRant, R.H. (1997). "Relationship between cigarette
smoking and health risk and problem behaviors among US adolescents." Archives
of Pediatric Adolescent Medicine 151: 66-71.

Essau, C. A., Conradt, J., Ederer, E.M. (2005). "Anxiety prevention among
schoolchildren." Versicherungsmedizin 56(3): 123-30.

Faith, M. S., Matz, P.E., & Jorge, M.A. (2002). "Obesity-depression associations in the
population." Journal of Psychosomatic Research 53: 935-942.

Fan, A. Z., Pentz, M.A., Dwyer, J., Bernstein, K., Li, C. (2002). " Attrition in a
longitudinal drug abuse prevention study." International Journal of Mental Health
Promotion 4(2): 15-23.
Farmer, M. E., Locke, B.Z., Moscicki, E.K., Dannenberg, A.L., Larson, D.B., Radloff,
L.S. (1988). "Physical activity and depressive symptoms: the NHANES I
Epidemiologic Follow-up Study." American Journal of Epidemiology 128(6).

Fava, M. (2000). "Weight gain and antidepressants." Journal of Clinical Psychiatry
61(suppl 11): 37-41.

Fergusson, D. M., Goodwin, R.D., Horwood, L.J. (2003). "Major depression and
cigarette smoking: results of a 21-year longitudinal study." Psychological
Medicine 33(8): 1357-67.

Ferraro, K. F., Yu, Y. (1995). "Body weight and self-ratings of health." Preventive
Medicine 32(5): 429-36.

Field, A. E., Camargo, C.A., Jr., Taylor, C.B., Berkey, C.S., Roberts, S.B., & Colditz,
G.A. (2001). "Peer, parent and media influences on the development of weight
concerns and frequent dieting among preadolescent and adolescent girls and
boys." Pediatrics 107: 54-60.

Flores, R. (1995). "Dance for health: improving fitness in African American and Hispanic
adolescents." Public Health Rep 110: 189-193.

Flynn, M. A., McNeil, D.A., Maloff, B., Mutasingwa, D., Wu, M., Ford, C., Tough, S.C.
(2006). "Reducing obesity and related chronic disease risk in children and youth:
a synthesis of evidence with 'best practice' recommendations." Obesity Reviews
7(Supplement 1): 7-66.

Forshee, R. A., Anderson, P.A., Storey, M.L. (2004). "The role of beverage consumption,
physical activity, sedentary behavior, and demographics on body mass index of
adolescents. ." International Journal of Food Sciences and Nutrition 55(6): 463-
478.
118

Franklin, M. E., Kozak, M.J., Cashman, L.A., Coles, M.E., Rheingold, A.A., Foa, E.B.
(1998). "Cognitive-behavioral treatment of pediatric obsessive-compulsive
disorder: an open clinical trial." Journal of American Academy of Child
Adolescence Psychiatry 37(4): 412-9.

Franko, D. L., Striegel-Moore, R. H., Bean, J., Tamer, R., Kraemer, H.C., Dohm, F.A.,
Crawford, P.B., Schreiber, G., Daniels, S.R. (2005). "Psychosocial and Health
Consequences of Adolescent Depression in Black and White Young Adult
Women " Health Psychology 24(6): 586-593.

Franko, D. L., Striegel-Moore, R.H., Thompson, D., Schreiber, G.B., Daniels, S.R.
(2005). "Does adolescent depression predict obesity in black and white young
adult women?" Psychol Med. 35(10): 1505-13.
Freedman, D. S., Serdula, M.K., Percy, C.A., Ballew, C., White, L. (1997). "Obesity,
levels of lipids and glucose, and smoking among Navajo adolescents." Journal of
Nutrition 127(10 Suppl.): 2120S-2127S.

French, S. A., & Jeffery, R.W. (1995). "Weight concerns and smoking: A literature
review." Annals of Behavioral Medicine 17: 234-244.

French, S. A., Perry, C.L. (1996). "Smoking among adolescent girls: Prevalence &
etiology." Journal of the American Medical Women's Association 51: 25-28.

French, S. A., Story, M., Downes, B., Resnick, M.D., Blum, R.W. (1995). "Frequent
dieting among adolescents: Psychosocial and health behavior correlates." Am J
Public Health. 85: 695-701.

Friedman, M. A., & Brownell, K.D. (1995). , 117 (1), 3-20. (1995). "Psychological
correlates of obesity: Moving to the next research generation." Psychological
Bulletin 117(1): 3-20.

Galambos, N. L., Leadbeater, B.J., Barker, E.T. (2004). "Gender differences in and risk
factors for depression in adolescence: A 4-year longitudinal study." International
Journal of Behavioral Development 28(1): 16-25.

Gazmararian, J., Baker, D., Parker, R., Blazer, D.G. (2000). "A multivariate analysis of
factors associated with depression: evaluating the role of health literacy as a
potential contributor." Archives of Internal Medicine 160(21): 3307-14.

Ge, X., Conger, R.D., Elder, G.H. Jr (2001). " Pubertal transition, stressful life events,
and the emergence of gender differences in adolescent depressive symptoms."
Dev Psychol 37(3): 404-417.

119
Geckova, A., Tuinstra, J., Pudelsky, M., Kovarova, M., van Dijk, M.P., Groothoff, J.W.,
and Post, D. (2001). "Self reported health problems of Slovak adolescents."
Journal of Adolescence 24(5): 635-645.

Gilpin, E. A., Choi, W.S, Berry, C., Pierce, J.P. (1999). "How many adolescents start
smoking each day in the United States?" Journal of Adolescent Health 25: 248-
255.

Gofin, J., Kark, J.D., Halfon, S.T., Friedlander, Y., Stein, Y. (1982). "Cigarette smoking
and its relation to anthropometric characteristics and biochemical variables in
Jerusalem 17-year-olds and adults." Isr J Med Sci. 18(12): 1233-1241.

Goodman, E., Capitman, J. (2000). "Depressive symptoms and cigarette smoking among
teens." Pediatrics 106(4): 748-55.
Goodman, E., Whitaker, R.C. (2002). "A prospective study of the role of depression in
the development and persistence of adolescent obesity." Pediatrics 109(3): 97-
504.

Goodwin, R. D. (2003). " Association between physical activity and mental disorders
among adults in the United States." Preventive Medicine 36(6): 698-703.

Gortmaker, S. L., Peterson, K., Wiecha, J., Sobol, A.M., Dixit, S., Fox, M.K., Laird, N.
(1999). "Reducing obesity via a school-based interdisciplinary intervention
among youth: Planet Health." Archives of Pediatric Adolescence Medicine
153(4): 409-18.

Gove, W. R., & Geerken, M.R. (1977). "The effect of children and employment on the
mental health of married men and women." Social Forces 56(1): 66-76.

Graham, J. W., Flay, B.R., Johnson, C.A. (1984). "Reliability of self-report measures of
drug use in prevention research: Multiform-multimethod evaluation of Project
SMART questionnaire items." Journal of Drug Education 14: 175-193.

Graham, J. W., Taylor, B.J., Olchowski, A.E., Cumsille, P.E. (2006). "Planned missing
data designs in psychological research." Psychological Methods 11(4): 323-343.

Gregor, K., Zvolensky, M.J, Leen-Feldner, E.W; Yartz, A.R., Feldner, M.T. (2006).
"Perceived health: A test of incremental validity in relation to smoking outcome
expectancies, motivation to smoke and desire to quit smoking." Cognitive
Behaviour Therapy 35(1): 28-42.

Grigg, M., Bowman, J., Redman, S. (1996). "Disordered eating and unhealthy weight
reduction practices among adolescent females." Preventive Medicine 25: 748-756.

120
Guallar-Castillon, P., Lopez Garcia, E., Lozano Palacios, L., Gutierrez-Fisac, J.L.,
Banegas Banegas, J.R., Lafuente Urdinguio, P.J., Rodriguez Artalejo, F. (2002).
"The relationship of overweight and obesity with subjective health and use of
health-care services among Spanish women." International Journal of Obesity
Related Metabolic Disorders 26(2): 247-52.

Gunnel, l. D., Berney, L., Holland, P., Maynard, M., Blane, D., Frankel, S., Smith, G.D.
(2000). " How accurately are height, weight and leg length reported by the
elderly, and how closely are they related to measurements recorded in
childhood?" International Journal of Epidemiology 29: 456-464.

Guo, S. S., Huang, C., Maynard, L.M., et al. (2000). "BMI during childhood,
adolescence, and young adulthood in relation to adult overweight and adiposity:
the Fels Longitudinal Study." International Journal of Obesity and Related
Metabolic Disorders 24: 1628-35.
Haapanen, N., Miilunpalo, S., Pasanen, M., Oja, P., Vuori, I. (1997). "Association
between leisure time physical activity and 10-year body mass change among
working-aged men and women." International Journal of Obesity and Related
Metabolic Disorders 21(4): 288-96.

Halek, C., Kerry, S., Humphrey, H., Crisp, A.H., & Hughes, J.M. (1993). "Relationship
between smoking, weight, and attitudes to weight in adolescent schoolgirls."
Postgraduate Medical Journal 69: 100-106.

Hampel, P., Petermann, F. (2006). "Perceived stress, coping, and adjustment in
adolescents." Journal of Adolescent Health 38(4): 409-415.

Han, B. (2001). "The impact of age, gender, and race on the relationship between
depression and self-rated health in community-dwelling older adults: a
longitudinal study." Home Health Care Services Quarterly 20(3): 27-43.

Han, B. (2002). "Depressive symptoms and self-rated health in community-dwelling
older adults: a longitudinal study." Journal of American Geriatric Society 50(9):
1549-56.

Hankin, B. L., Abramson, L.Y. (2001). "Development of gender differences in
depression: an elaborated cognitive vulnerability-transactional stress theory."
Psychol. Bull. 127: 773-796.

Hansell, S., White, H. R. (1991). "Adolescent drug use, psychological distress, and
physical symptoms." Journal of Health and Social Behavior 32(3): 288-301.

121
Hasler, G., Lissek, S., Ajdacic, V., Milos, G., Gamma, A., Eich, D., Rossler, W., Angst,
J. (2005). "Major depression predicts an increase in long-term body weight
variability in young adults." Obes Res. 13(11): 1991-8.

Hasler, G., Pine, D.S., Kleinbaum, D.G., Gamma, A., Luckenbaugh, D., Ajdacic, V.,
Eich, D., Rossler, W., Angst, J. (2005). "Depressive symptoms during childhood
and adult obesity: the Zurich Cohort Study." Molecular Psychiatry 10(9): 842-50.

Herman-Stahl, M., Petersen, A. C. (1999). "Depressive symptoms during adolescence:
Direct and stress-buffering effects of coping, control beliefs, and family
relationships. ." Journal of Applied Developmental Psychology 20: 45-62.

Herman-Stahl, M., Stemmler, M., Petersen, A.C. (1995). "Approach and avoidant coping:
Implications for adolescent health." Journal of Youth and Adolescence 24: 649-
665.

Holtkamp, K., Konrad, K., Muller, B., Heussen, N., Herpertz, S., Herpertz-Dahlmann, B.,
Hebebrand, J. (2004). "Overweight and obesity in children with Attention-
Deficit/Hyperactivity Disorder." International Journal of Obesity and Related
Metabolic Disordorders 28(5): 685-9.

Hoppa, H., & Hallstrom, T. (1981). "Weight gain in adulthood in relation to
socioeconomic factors, mental illness, and personality traits: A prospective study
of middle-aged women." Journal of Psychosomatic Research 25: 83-89.

Hubel, R., Jass, J., Marcus, A., Laessle, R.G. (2006). "Overweight and basal metabolic
rate in boys with attention-deficit/hyperactivity disorder." Eat Weight Disord.
11(3): 139-46.

Hughes, J. R., Hatsukami, D. K, Mitchell, J. E. & Dahlgren, L. A. (1986). "Prevalence of
smoking among psychiatric outpatients,." American Journal of Psychiatry 143:
993-997.

Idler, E. L., Benyamini, Y. (1997). "Self-rated health and mortality: a review of twenty-
seven community studies." Journal of Health Social Behavior 38: 21-37.

Ismail, K., Sloggett, A., De Stavola, B. (2000). "Do common mental disorders increase
cigarette smoking? Results from five waves of a population-based panel cohort
study." Am J Epidemiol. 152(7): 651-7.

Istvan, J., Zavela, K., Weidner, G. (1992). "Body weight and psychological distress in
NHANES I." Int J Obes Relat Metab Disord. 16(12): 999-1003.

122
Jago, R., Baranowski, T., Baranowski, J.C., Thompson, D., Greaves, K.A. (2005). "BMI
from 3-6 y of age is predicted by TV viewing and physical activity, not diet."
International Journal of Obesity and Related Metabolic Disorders 29(6): 557-64.

Jakicic, J. M. O., A.D. (2005). "Physical activity considerations for the treatment and
prevention of obesity." American Journal of Clinical Nutrition 82: 226S-9S.

Jasuja, G. K., Chou, C.P., Riggs, N.R., Pentz, M.A. (2008). "Early Cigarette Use and
Psychological Distress as Predictors of Obesity Risk in Adulthood." Nicotine and
Tobacco Research (in-press). 10(2): 325-335.

Jeffery, R. W., French, S.A., Forster, J.L., Spry, V.M. (1991). "Socioeconomic status
differences in health behaviors related to obesity: the Healthy Worker Project."
International Journal of Obesity 15(10): 689-96.

Jessor, R. (1991). "Risk behavior in adolescence: A psychosocial framework for
understanding and action." Journal of Adolescent Health 12: 597-605.
John, U., Meyer, C, Rumpf, HJ, Schumann, A, Dilling, H, Hapke, U (2005). "Self-rated
general health and psychiatric disorders in a general population sample." Eur
Psychiatry. 20(3): 223-28.

John, U., Meyer, C., Rumpf, H.J., Hapke, U. (2004). "Smoking, nicotine dependence and
psychiatric comorbidity--a population-based study including smoking cessation
after three years." Drug Alcohol Depend. 76(3): 287-95.

John, U., Meyer, C., Rumpf, H.J., Hapke, U. (2005). "Relationships of psychiatric
disorders with overweight and obesity in an adult general population." Obes Res.
13(1): 101-9.

Johnson, E. O., Rhee, S.H., Chase, G.A., Breslau, N. (2004). "Comorbidity of depression
with levels of smoking: an exploration of the shared familial risk hypothesis."
Nicotine and Tobacco Research 6(6): 1029-38.

Johnston, L. D., O'Malley, P.M., Bachman, J.G. (2003). Monitoring the Future national
survey results on drug use, 1975-2000. Volume 1: Secondary School Students
(NIH Publication No. 02-5106). Bethesda, MD: National Institute on Drug Abuse.

Kain, J., Uauy, R., Albala, Vio, F., Cerda, R., Leyton, B. (2004). "School-based obesity
prevention in Chilean primary school children: methodology and evaluation of a
controlled study." International Journal of Obesity and Related Metabolic
Disorders, 28(4): 483-93.

123
Kandel, D., Davies, M, Karus, D, Yamaguchi, K (1986). "The consequences in young
adulthood of adolescent drug involvement. An overview." Arch Gen Psychiatry.
43(8).

Kandel, D. B., Davies, M. (1986). "Adult sequelae of adolescent depressive symptoms."
Arch Gen Psychiatry. 43(3).

Kandel, D. B., Johnson, J.G., Bird, H.R., Canino, G., Goodman, S.H., Lahey, B.B.,
Regier, D.A., Schwab-Stone, M. (1997). "Psychiatric disorders associated with
substance use among children and adolescents: findings from the Methods for the
Epidemiology of Child and Adolescent Mental Disorders (MECA) Study." J
Abnorm Child Psychol. 25(2): 121-32.

Kandel, D. B., Johnson, J.G., Bird, H.R., Weissman, M.M., Goodman, S.H., Lahey, B.B.,
Regier, D.A., Schwab-Stone, M.E. (1999). "Psychiatric comorbidity among
adolescents with substance use disorders: findings from the MECA Study."
Journal of the American Academy of Child and Adolescent Psychiatry 38(6):
693-9.

Kandel, D. B., Yamaguchi, K., Chen, K. (1992). "Stages of progression in drug
involvement from adolescence to adulthood: further evidence for the gateway
theory." Journal of Studies on Alcohol 53(5): 447-57.

Kaplan, M. S., Huguet, N., Newsom, J.T., McFarland, B.H., Lindsay, J. (2003).
"Prevalence and correlates of overweight and obesity among older adults:
findings from the Canadian National Population Health Survey." Journal of
Gerontology Series A: Biological Sciences and Medical Sciences 58(11): 1018-
30.

Kaur, H., Hyder, M.L., Poston, W.S. (2003). " Childhood overweight: an expanding
problem. ." Treat Endocrinology 2(6): 375-88.

Kendler, K. S., Neale, M. C., MacLean, C. J., Heath, A. C., Eaves, L. J., & Kessler, R. C.
(1993). "Smoking and major depression: A causal analysis." Archives of General
Psychiatry 50: 36-43.

Killen, J. D., Fortmann, S.P., Kraemer, H. C., Varady, A.N., Davis, L., Newman, B.
(1996). "Interactive Effects of Depression Symptoms, Nicotine Dependence, and
Weight Change on Late Smoking Relapse " Journal of Consulting and Clinical
Psychology 64(5): 1060-1067.

Killen, J. D., Robinson, T.N., Haydel, K.F., Hayward, C., Wilson, D.M., Hammer, L.D.,
Litt, I.F., Taylor, C.B. (1997). "Prospective study of risk factors for the initiation
of cigarette smoking." J Consult Clin Psychol. 65(6): 1011-6.
124

Kimm, S. Y., Barton, B.A., Obarzanek, E., McMahon, R.P., Kronsberg, S.S., Waclawiw,
M.A., Morrison, J.A. (2002). " Obesity development during adolescence in a
biracial cohort: the NHLBI Growth and Health Study." Pediatrics 110(5): E54.

Kirkcaldy, B. D., Shephard, R.J., Siefen, R.G. (2002). "The relationship between physical
activity and self-image and problem behavior among adolescents." Soc Psychiatry
Psychiatr Epidemiol 37: 544-550.

Klesges, R. C., Eck, L.H., Isbell, T.R., Fulliton, W., Hanson, C.L. (1990). "Smoking
status: effects on the dietary intake, physical activity, and body fat of adult men."
American Journal of Clinical Nutrition 51(784-9).

Klesges, R. C., Elliott, V.E., Robinson, L.A. (1997). "Chronic dieting and the belief that
smoking controls body weight in a biracial, population-based adolescent sample."
Tobacco Control 6(2): 89-94.

Klesges, R. C., Klesges, L.M., Haddock, C.K., Eck, L.H. (1992). "A longitudinal analysis
of the impact of dietary intake and physical activity on weight change in adults."
American Journal of Clinical Nutrition 55: 818-22.
Klesges, R. C., Robinson, L.A., Zbikowski, S.M. (1998). "Is smoking associated with
lower body mass in adolescents? A large-scale biracial investigation." Addict
Behav. 23(1): 109-113.

Klesges, R. C., Ward, K.D., Ray, J.W., Cutter, G., Jacobs, D.R. Jr, Wagenknecht, L.E.
(1998). "The prospective relationships between smoking and weight in a young,
biracial cohort: the Coronary Artery Risk Development in Young Adults Study."
Journal of Consulting and Clinical Psychology 66(6): 987-93.

Klesges, R. C., Zbikowski, S.M., Lando, H.A., Haddock, C.K., Talcott, G.W., Robinson,
L.A. (1998). "The relationship between smoking and body weight in a population
of young military personnel." Health Psychol. 17(5): 454-8.

Lamb, J. M., Puskar, K.R., Sereika, S.M., Corcoran, M. (1998). " School-based
intervention to promote coping in rural teens." American Journal of Maternal
Child Nursing 23(4): 187-94.

Lampinen, P., Heikkinen, R.L., Ruoppila, I. (2000). "Changes in intensity of physical
exercise as predictors of depressive symptoms among older adults: an eight-year
follow-up." Preventive Medicine 30(5): 371-80.

Lanza, S. T., Collins, L.M. (2002). "Pubertal timing and the onset of substance use in
females during early adolescence. ." Prevention Science 3(1): 69-82.

125
Larsen, J. K., Ouwens, M., Engels, R.C., Eisinga, R., van Strien, T. (2008). "Validity of
self-reported weight and height and predictors of weight bias in female college
students." Appetite 50(2-3): 386-389.

Lazarus, R. S., & Folkman, S. (1984). Stress appraisal and coping. New York, Springer.

Lazarus, R. S., Cohen, J.B. (1977). Environmental Stress. Human Behavior and
Environment. I. A. a. J. F. Wohlwill. New York, Plenum. 2.

Lewinsohn, P. M., Roberts, R. E., Seeley, J. R., Rohde, P., Gotlib, I. H. and Hops, H.
(1994). "Adolescent psychopathology: II. Psychosocial risk factors for
depression." Journal of Abnormal Psychology 103: 302-315.

Lim, W. Y., Ma, S., Heng, D., Bhalla, V., Chew, S.K. (2007). "Gender, ethnicity, health
behaviour & self-rated health in Singapore." BMC Public Health 27(7): 184.

Lissner, L., Bengtsson, C., Lapidus, L., Bjorkelund, C. (1992). "Smoking initiation and
cessation in relation to body fat distribution based on data from a study of
Swedish women." Am J Public Health. 82(2): 273-5.

Luepker, R. V., Perry, C.L., McKinlay, S.M., et al. (1996). "Outcomes of a field trial to
improve children‘s dietary patterns and physical activity: The Child and
Adolescent Trial for CardiovascularHealth." Journal of the American Medical
Association 275: 768–776.

Maffeis, C., Tatò, L. (2001). "Long-term effects of childhood obesity on morbidity and
mortality." Hormone Research 55(Supplement 1): 42-45.

Manora, O., Matthewsb, S., Powerb, C. (2000). "Dichotomous or categorical response?
Analysing self-rated health and lifetime social class." International Journal of
Epidemiology 29: 149-157.

McElroy, S., Kotwal, R., Malhotra, S., Nelson, E., Keck, P. & Nemeroff, C. (2004). "Are
mood disorders and obesity related? A review for the mental health professional."
Journal of Clinical Psychiatry 65: 634-651.

McGee, D. L., Liao, Y., Cao, G. (1999). "Self-reported health status and mortality in a
multiethnic US cohort. ." Am J Epidemiol 149: 41-46.

McGee, R., Williams, S., Stanton, W. (1998). "Is mental health in childhood a major
predictor of smoking in adolescence?" Addiction 93(12): 1869-74.

126
Merry, S., McDowell, H., Wild, C.J., Bir, J., Cunliffe, R. (2004). "A randomized placebo-
controlled trial of a school-based depression prevention program." Journal of
American Academy of Child Adolescence Psychiatry, 43(5): 538-47.

Mulsant, B. H., Ganguli, M., Seaberg, E.C. (1997). "The relationship between self-rated
health and depressive symptoms in an epidemiological sample of community-
dwelling older adults." Journal of American Geriatric Society 45(8): 954-8.

Murberg, T. A., Bru, E. (2005). "The role of coping styles as predictors of depressive
symptoms among adolescents: a prospective study." Scandinavian journal of
psychology 46(4): 385-93.

Nagata, T., Toshihiko, M.D., Matsuyama, M. Kiriike, N., Iketani, T., Toshiya, M.D.,
Oshima, J. (2000). " Stress coping strategy in Japanese patients with eating
disorders: relationship with bulimic and impulsive behaviors." The Journal of
Nervous and Mental Disease 188(5): 280-286.

Nerin, I., Crucelaegui, A., Novella, P., Ramon y, P, Sobradiel, N., Gerico, R. (2004). "A
survey on the relationship between tobacco use and physical exercise among
university students." Arch Bronconeumol. 40(1): 5-9.

Neumark-Sztainer, D., Paxton, S.J., Hannan, P.J., Haines, J., Story, M. (2006). "Does
body satisfaction matter? Five-year longitudinal associations between body
satisfaction and health behaviors in adolescent females and males." Journal of
Adolescent Health 39(2): 244-51.

Neumark-Sztainer, D., Story, M., Hannan, P.J., Rex, J. (2003). "New Moves: A school-
based obesity prevention program for adolescent girls." Preventive Medicine 37:
41-51.

Neumark, S. D., Story, M., Hannan, P.J., Rex, J. (2003). "New Moves: A school-based
obesity prevention program for adolescent girls." Preventive Medicine 37: 41-51.

Newcomb, M. D. (1997). "Psychosocial predictors and consequences of drug use: a
developmental perspective within a prospective study." Journal of Addictive
Diseases 16(1): 51-89.

Nicklas, T. A., Baranowski, T., Cullen, K.W., Berenson, G. (2001). "Eating patterns,
dietary quality and obesity." Journal of the American College of Nutrition 20:
599-608.

127
Nicklas, T. A., Johnson, C.C., Myers, L., et al. (1998). "Outcomes of a high school
program to increase fruit and vegetable consumption: Gimme 5—a fresh nutrition
concept for students." Journal of School Health 68: 248-53.

Nicklas, T. A., Yang, S.J., Baranowski, T., Zakeri, I., Berenson, G. (2003). "Eating
patterns and obesity in children. The Bogalusa Heart Study." American Journal of
Preventive Medicine 25: 9-16.

Niedhammer, I., Bugel, I., Bonenfant, S., Goldberg, M., Leclerc, A. (2000). "Validity of
self-reported weight and height in the French GAZEL cohort." Int J Obes Relat
Metab Disord. 24(9): 1111-8.

Noppa, H., Hallstrom, T. (1981). "Weight gain in adulthood in relation to socioeconomic
factors, mental illness, and personality traits: a prospective study of middle-aged
women." J Psychosom Res 25: 83-89.

Norman, A., Bellocco, R., Vaida, F., Wolk, A (2002). "Total physical activity in relation
to age, body mass, health and other factors in a cohort of Swedish men."
International Journal of Obesity and Related Metabolic Disorders 26(5): 670-5.

Ogden, C. L., Carroll, M.D., Curtin, L.R., McDowell, M.A., Tabak, C.J., Flegal, K.M.
(2006). "Prevalence of overweight and obesity in the United States, 1999-2004."
JAMA 295(13): 1549-55.

Ogden, C. L., Yanovski, S.Z., Carroll, M.D., Flegal, K.M. (2007). "The epidemiology of
obesity." Gastroenterology 132(6): 2087-102.
Okosun, I. S., Choi, S., Matamoros, T., Dever, G.E. (2001). "Obesity is associated with
reduced self-rated general health status: evidence from a representative sample of
white, black, and Hispanic Americans." Preventive Medicine 32(5): 429-36.

Ostbye, T., Pomerleau, J., Speechley, M., Pederson, L.L., Speechley, K.N. (1995).
"Correlates of body mass index in the 1990 Ontario Health Survey." CMAJ
152(11): 1811-1817.

Page, R. M., Allen, O., Moore, L., & Hewitt, C. (1993). "Weight-related concerns and
practices of male and female adolescent cigarette smokers and nonsmokers."
Journal of Health Education 24: 339-346.

Parsons, T. J., Manor, O., Power, C. (2006). "Physical activity and change in body mass
index from adolescence to mid-adulthood in the 1958 British cohort."
International Journal of Epidemiology 35(1): 197-204.

Patten, C. A., Choi, W.S., Vickers, K.S., Pierce, J.P. (2001). "Persistence of depressive
symptoms in adolescents." Neuropsychopharmacology. 25(Suppl. 5): S89-91.
128

Patton, G. C., Carlin, J.B., Coffey, C., Wolfe, R., Hibbert, M., Bowes, G. (1998).
"Depression, anxiety, and smoking initiation: a prospective study over 3 years."
American Journal of Public Health 88(10): 1518-22.

Penas-Lledo, E., Sancho, L., Waller, G. (2002). "Eating attitudes and the use of alcohol,
tobacco, and exercise among male and female adolescents." Eat Behav. 3(2): 101-
11.

Pentz, M. A. (1999). Prevention aimed at individuals: An integrative transactional
perspective. . New York, Oxford University Press.

Pentz, M. A., Chou, C.P. (1994). "Measurement invariance in longitudinal clinical
research assuming change from development and intervention." Journal of
Consulting and Clinical Psychology 62: 450-462.

Pentz, M. A., Dwyer, J.H., MacKinnon, D.P., Flay, B.R., Hansen, W.B., Wang, E.Y.I.,
Johnson, C.A. (1989). "A multicommuntiy trial for primary prevention of
adolescent drug abuse." JAMA 261(22): 3259-3266.

Pentz, M. A., Mihalic, S.F., Grotpeter, J.K. (1997). The Midwestern Prevention Project.
Boulder, CO, Center for the Study and Prevention of Violence, Institute of
Behavioral Science, University of Colorado.

Pentz, M. A., Riggs, N. R., Johnson, C. A., Chou, C-P., Duan, L. (2006). "Prevention of
growth in drug use over four developmental periods: Long-term effects of the
Midwestern Prevention Project (STAR)." Manuscript submitted for publication.
Pentz, M. A., Valente, T. W. (1993). Project STAR: A substance abuse prevention
campaign in Kansas City. Thousand Oaks, CA, Sage Publications.

Pine, D. S., Cohen, P., Brook, J., Coplan, J.D. (1997). "Psychiatric symptoms in
adolescence as predictors of obesity in early adulthood: a longitudinal study."
American Journal of Public Health 87(8): 1303-10.

Pine, D. S., Goldstein, R.B., Wolk, S., & Weissman, M.M. (2001). "The association
between childhood depression and adulthood body mass index." Pediatrics
107(5): 1049-1056.

Pomerleau, O. F., Pomerleau, C.S. (1984). "Neuroregulators and the reinforcement of
smoking: Towards a biobehavioral explanation." Neuroscience and Biobehavioral
Review 8: 503-513.

129
Possel, P., Horn, A.B., Groen, G., Hautzinger, M. (2004). "School-based prevention of
depressive symptoms in adolescents: a 6-month follow-up." Journal of American
Academy of Child Adolescence Psychiatry 43(8): 1003-10.

Potter, B. K., Pederson, L.L., Chan, S.S.H., Aubut, J.L., Koval, J.J. (2004). "Does a
relationship exist between body weight, concerns about weight, and smoking
among adolescents? An integration of the literature with an emphasis on gender."
Nicotine and Tobacco Research 6(3): 397-425.

Poulin, C., Hand D, Boudreau B, Santor D. (2005). "Gender differences in the association
between substance use and elevated depressive symptoms in a general adolescent
population." Addiction 100(4): 525-35.

Power, C., Lake, J.K., Cole, T.J. (1997). "Body mass index and height from childhood to
adulthood in the 1958 British born cohort." American Journal of Clinical
Nutrition 66(5): 1094-101.

Prentice, A. M. (2001). "Overeating: The Health Risks." Obesity Research 9: S234-238.

Prentice, A. M., Jebb, S.A. (2001). "Beyond Body Mass Index." Obesity Reviews 2(3):
141-7.

Proctor, M., Moore, L., Gao, D., Cupples, L., Bradlee, M., Hood, M., & and R. Ellison
(2003). "TV viewing and change in body fat from preschool to early adolescence:
The Framingham Children‘s Study." International Journal of Obesity and Related
Metabolic Disorders 27: 827-833.

Rasmussen, F., Tynelius, P., Kark, M. (2003). "Importance of smoking habits for
longitudinal and age-matched changes in body mass index: a cohort study of
Swedish men and women." Prev Med. 37(1): 1-9.
Repetto, P. B., Caldwell, C.H., Zimmerman, M.A. (2005). "A longitudinal study of the
relationship between depressive symptoms and cigarette use among African
American adolescents." Health Psychology 24(2): 209-19.

Richardson, L. P., Davis, R., Poulton, R., McCauley, E., Moffitt, T.E., Caspi, A.,
Connell, F. (2003). "A longitudinal evaluation of adolescent depression and adult
obesity." Archives of Pediatric Adolescents 157: 739-745.

Roberts, R. E., Deleger, S., Strawbridge, W.J., Kaplan, G.A. (2003). "Prospective
association between obesity and depression: evidence from the Alameda County
Study." Int J Obes Relat Metab Disord. 27(4): 514-21.

130
Robinson, L. A., Klesges, R.C., Zbikowski, S.M., Glaser, R. (1997). "Predictors of risk
for different stages of adolescent smoking in a biracial sample." J Consult Clin
Psychol. 65(4): 653-62.

Robinson, T. N. (1999). "Behavioural treatment of childhood and adolescent obesity." Int
J Obes 23(Supplement 2): S52-7.

Rodriguez, D., Moss, H.B., Audrain-McGovern, J. (2005). "Developmental heterogeneity
in adolescent depressive symptoms: associations with smoking behavior."
Psychosom Med. 67(2): 200-10.

Rohrer, J. E., Rohland, B.M. (2004). "Psychosocial risk factors for obesity among women
in a family planning clinic." BMC Family Practice 5: 20.

Rosenbaum, M., Leibel, R., & Hirsch, J. (1997). "Obesity." New England Journal of
Medicine 337: 396-407.

Ryden, A., Karlsson, J., Persson, L.O., Sjostrom, L., Taft, C., Sullivan, M. (2001).
"Obesity-related coping and distress and relationship to treatment preference."
British Journal of Clinical Psychology 40: 177-88.

Ryden, A., Karlsson, J., Sullivan, M., Torgerson, J.S., Taft, C. (2003). "Coping and
distress: what happens after intervention? A 2-year follow-up from the Swedish
Obese Subjects (SOS) study." Psychosomatic Medicine 65(3): 35-42.

Sallis, J. F., McKenzie, T.L., Conway, T.L., Elder, J.P., Prochaska, J.J., Brown, M., Zive,
M.M., Marshall, S.J., Alcaraz, J.E. (2003). "Environmental interventions for
eating and physical activity: a randomized controlled trial in middle schools."
American Journal of Preventive Medicine 24(3): 209-17.

SAMHSA (1997,1998). National Household Survey on Drug Abuse, SAMHSA, Office
of Applied Studies.

SAS (1999-2001). Version 8.1 of the SAS system for Windows. Cary, NC, SAS Institute
Inc.

Segovia, J., Bartlett, R.F., Edwards, A.C. (1989). "The association between self-assessed
health status and individual health practices." Can J Pub Health 80: 32-37.

Seiffge-Krenke, I. (1993). "Coping behavior in normal and clinical samples: more
similarities than differences?" Journal of Adolescence 16: 285-304.

Seiffge-Krenke, I. (2000). "Causal links between stressful events, coping style, and
adolescent symptomatology." Journal of Adolescence 23: 675-691.
131

Seo, D. C., Torabi, M.R. (2006). "Racial/ethnic differences in body mass index,
morbidity and attitudes toward obesity among U.S. adults." Journal of the
National Medical Association 98(8): 1300-8.

Simons-Morton, B., Crump, A.D., Haynie, D.L., Saylor, K.E., Eitel, P., Yu, K. (1999).
"Psychosocial, School, and Parent Factors Associated with Recent Smoking
among Early-Adolescent Boys and Girls." Preventive Medicine 28: 138-148.

Siqueira, L., Diab, M., Bodian, C., & Rolnitzky, L. (2000). "Adolescents becoming
smokers: the roles of stress and coping methods." Journal of Adolescent Health
27(6): 399-408.

Skinner, J. D., Bounds, W., Carruth, B.R., Morris, M., Ziegler, P. (2004). "Predictors of
children's body mass index: a longitudinal study of diet and growth in children
aged 2-8 y." International Journal of Obesity and Related Metabolic Disorders
28(4): 476-82.

Sleskova, M., Salonna, F., Geckova, A., Madarasova, van D., Jitse, P., Groothoff, J. W.
(2005). "Health status among young people in Slovakia: Comparisons on the basis
of age, gender and education " Social Science & Medicine 61(12): 2521-2527.

Spencer, E. A., Appleby, P.N., Davey, G.K., Key, T.J. (2002). "Validity of self-reported
height and weight in 4808 EPIC-Oxford participants." Public Health Nutrition 5:
561-565.

Spruijt-Metz, D. (1995). "Personal incentives as determinants of adolescent health
behavior: The meaning of behavior. ." Health Education Research 10(3): 355-64.

Spruijt-Metz, D. (1999). Adolescence, affect, and health. Studies in adolescent
development. Hove, East Sussex, UK, Psychology Press.

St-Onge, M. P., Keller, K.L., Heymsfield, S.B. (2003). "Changes in childhood food
consumption patterns: a cause for concern in light of increasing body weights."
American Journal of Clinical Nutrition 78: 1068-73.

Steuber, T. L., Danner, F. (2006). "Adolescent smoking and depression: Which comes
first?" Addict Behav. 31(1): 133-6.

Stewart, A. L., Hays, R.D., Wells, K.B., Rogers, W.H., Spritzer, K.L., & Greenfield, S.
(1994). "Long-term functioning and well-being outcomes associated with physical
activity and exercise in patients with chronic conditions in the Medical Outcomes
Study." Journal of Clinical Epidemiology 47: 719-730.

132
Stice, E., Presnell, K., Shaw, H., Rohde, P. (2005). "Psychological and behavioral risk
factors for obesity onset in adolescent girls: A prospective study." Journal of
Consulting and Clinical Psychology 73(2): 195-202.

Stice, E., Shaw, H. (2003). "Prospective Relations of Body Image, Eating, and Affective
Disturbances to Smoking Onset in Adolescent Girls: How Virginia Slims "
Journal of Consulting and Clinical Psychology 71(1): 129-135.

Storey, M. L., Forshee, R.A., Weaver, A.R., Sansalone, W.R. (2003). "Demographic and
lifestyle factors associated with body mass index among children and adolescents.
." International Journal of Food Sciences and Nutrition 54(6): 491-503.

Strauss, R. S., Mir, H.M. (2001). "Smoking and weight loss attempts in overweight and
normal-weight adolescents." International Journal of Obesity Related Metabolic
Disorders 25(9): 1381-5.

Studer, H. K., & Weicker, H. (2001). "Physiology and pathophysiology of the
serotonergic system and its implications on mental and physical performance." Int
J Sports Med 22: 467-481.

Thomas, H. (2006). "Obesity prevention programs for children and youth: why are their
results so modest?" Health Education Research 21(6): 783-95.

Tomeo, C. A., Field, A.E., Berkey, C.S., Colditz, G.A., Frazier, A.L. (1999). "Weight
concerns, weight control behaviors, and smoking initiation." Pediatrics. 104(4 Pt.
1): 918-24.

Townsend, J., Wilkes, H., Haines, A., Jarvis, M. (1991). "Adolescent smokers seen in
general practice: health, lifestyle, physical measurements, and response to
antismoking advice." BMJ. 303(6808): 947-50.

Troop, N. A., Holbrey, A., Trowler, R., Treasure, J.L. (1994). "Ways of coping in women
with eating disorders." The Journal of Nervous and Mental Disease 182(10): 535-
540.

Troped, P. J., Wiecha, J.L., Fragala, M.S., Matthews, C.E., Finkelstein, D.M., Kim, J.,
Peterson, K.E. (2007). "Reliability and validity of YRBS physical activity items
among middle school students." Medicine and Science in Sports and Exercise
39(3): 416-25.

Turbin, M. S., Jessor, R., Costa, F.M. (2000). "Adolescent cigarette smoking: Health-
related behavior or normative transgression?" Prevention Science 1: 115-124.

133
USDHHS (2004). The health consequences of smoking: a report of the Surgeon General.
Atlanta, Georgia, United States Department of Health and Human Services, CDC,
National Center for Chronic Disease Prevention and Health Promotion, Office on
Smoking and Health.

Van Lenthe, F. J., Kemper, C.G., Van Mechelen, W. (1996). "Rapid maturation in
adolescence results in greater obesity in adulthood: the Amsterdam Growth and
Health Study." American Journal of Clinical Nutrition, 64(1): 180-24.

Vickers, K. S., Patten, C. A., Lane, K., Clark, M. M., Crogan, I. T., Schroeder, D. R.,
Hurt, R. D. (2003). "Depressed versus nondepressed young adult tobacco users:
Differences in coping style, weight concerns and exercise level." Health
Psychology 22: 498-503.

Vingilis, E., Wade, T.J., Adlaf, E. (1998). "What factors predict student self-rated
physical health?" Journal of Adolescent Health 21(1): 83-97.

Vingilis, E., Wade, T.J., Seeley, J.S. (2002). "Predictors of adolescent self-rated health.
Analysis of the National Population Health Survey." Can J Pub Health 93(3): 193-
197.

Wang, M. Q., Fitzhugh, E.C., Green, B.L., Turner, L.W., Eddy, J.M., Westerfield, R.C.
(1999). "Prospective social-psychological factors of adolescent smoking
progression." J Adolesc Health. 24(1): 2-9.

Warren, C. W., Kann, L., Small, M., Santelle, J., Collins, J., Kolbe, L. (1997). "Age of
initiating selected health-risk behaviors among high school students in the United
States. ." Journal of Adolescent Health 21: 225-231.

Wellman, N. S., Friedberg, B. (2002). "Causes and consequences of adult obesity: health,
social and economic impacts in the United States." Asia Pacific Journal of
Clinical Nutrition 11(Supplement 8): S705-9.

Williamson, D., Dewey, A., Steinberg, H. (2001). "Mood change through physical
exercise in nine- to ten-year-old children." Percept Mot Skills. 93: 311-316.

Wills, T. A., Hirky, A.E. (1996). Coping and substance abuse: a theoretical model and
review of the evidence. New York, Wiley.

Wilson, D. B., Smith, B.N., Speizer, I.S., Bean, M.K., Mitchell, K.S., Uguy, L.S., Fries,
E.A. (2005). "Differences in food intake and exercise by smoking status in
adolescents." Preventive Medicine 40(6): 272-79.

134
Windle, M., Windle, R.C. (2001). "Depressive symptoms and cigarette smoking among
middle adolescents: prospective associations and intrapersonal and interpersonal
influences." Journal of Consulting and Clinical Psychology 69(2): 215-26.

Wing, R. R. (1999). "Physical activity in the treatment of the adulthood overweight and
obesity: current evidence and research issues." Med. Sci. Sports. Exerc 31: S547-
S552.

Wiseman, C. V., Turco, R.M., Sunday, S.R., Halmi, K.A. (1998). "Smoking and body
image concerns in adolescent girls." International Journal of Eating Disorders
24(4): 429-33.

Wood, A., Harrington, R., Moore, A. (1996). "Controlled trial of a brief cognitive-
behavioural intervention in adolescent patients with depressive disorders." Journal
of Child Psychology and Psychiatry 37(6): 737-46.

Wothke, W. (1998). Longitudinal and multi-group modeling with missing data. Modeling
longitudinal and multiple group data: Practical issues, applied approaches and
specific examples. K. U. S. T. D. Little, and J. Baumert. Mahwah, New Jersey,
Lawrence Erlbaum Publishers.

Wu, L. T., Anthony, J.C. (1999). "Tobacco smoking and depressed mood in late
childhood and early adolescence." Am J Public Health. 89(12): 1837-40.

Xu, F., Yin, X.M., Wang, Y. (2007). "The association between amount of cigarettes
smoked and overweight, central obesity among Chinese adults in Nanjing,
China." Asia Pac J Clin Nutr. 16(2): 240-7.

Yager, J., Rorty, M., Rossotto, E. (1995). "Coping styles differ between recovered and
nonrecovered women with bulimia nervosa, but not between recovered women
and non-eating-disordered control subjects." The Journal of Nervous and Mental
Disease 183(2): 86-94.

Young, M. A., Scheftner, W.A., Fawcett, J., Klerman, G.L. (1990). "Gender differences
in the clinical features of unipolar major depressive disorder." The Journal Of
Nervous And Mental Disease 178(3): 200-203.

Zoeller, R. F. (2007). "Physical Activity: Depression, Anxiety, Physical Activity, and
Cardiovascular Disease: What's the Connection?" American Journal of Lifestyle
Medicine 1(3): 175-180.

135
Appendix A: Complete structural model

Early Adulthood
(mean:30.06 yrs)
BMI
Mid-Emerging Adulthood
(mean:23.22 yrs)
Psychological
distress
Psychological
distress
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.908; CFI=.917
c2 (df)=3312.877 (393)
RMSEA=.038
N=5019
.403***
.661***
.128**
.143**
.681***
-.230***
Cigarette use
Psychological
distress
-.103**
.675***
-.155**
.011*
.521***
.138**
.097*
Physical
activity
Subjective rating
of health
-.074*
-.189**
-.211***
-.102**
.350*** -.242***
*p<.05, **p<.01, ***p<.001 (two-tailed test)
Model adjusted for baseline age, ethnicity, gender, intervention condition, baseline SES, and current SES.
Early Adulthood
(mean:30.06 yrs)
BMI
Mid-Emerging Adulthood
(mean:23.22 yrs)
Psychological
distress
Psychological
distress
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.908; CFI=.917
c2 (df)=3312.877 (393)
RMSEA=.038
N=5019
.403***
.661***
.128**
.143**
.681***
-.230***
Cigarette use
Psychological
distress
-.103**
.675***
-.155**
.011*
.521***
.138**
.097*
Physical
activity
Subjective rating
of health
-.074*
-.189**
-.211***
-.102**
.350*** -.242***
*p<.05, **p<.01, ***p<.001 (two-tailed test)
Model adjusted for baseline age, ethnicity, gender, intervention condition, baseline SES, and current SES.
136
Appendix B: Model with both maladaptive and adaptive coping

Early Adulthood
(mean:30.06 yrs)
Mid-Emerging Adulthood
(mean:23.22 yrs)
Psychological
distress
Maladaptive
coping
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.988
CFI=.997
2 =1889.830
(345)
RMSEA=.030
N=5019
Only significant paths are shown in this figure (p<.05)
. 296
.647
Cigarette use
BMI
Psychological
distress
.157
.665
.407
.480
Psychological
distress
Adaptive
coping
-.088
.229
-.084
-.124 .140
.510
.265
.854
.112
-.268
-.127
Early Adulthood
(mean:30.06 yrs)
Mid-Emerging Adulthood
(mean:23.22 yrs)
Psychological
distress
Maladaptive
coping
Early Adolescence
(mean:12.06 yrs)
Beginning of Emerging Adulthood
(mean:19.47 yrs)
Cigarette use Cigarette use
Cigarette use
Model fit:
NFI=.988
CFI=.997
2 =1889.830
(345)
RMSEA=.030
N=5019
Only significant paths are shown in this figure (p<.05)
. 296
.647
Cigarette use
BMI
Psychological
distress
.157
.665
.407
.480
Psychological
distress
Adaptive
coping
-.088
.229
-.084
-.124 .140
.510
.265
.854
.112
-.268
-.127

Abstract (if available)

Abstract Cigarette use, depression, and obesity are major health threats to society. The three studies of this dissertation examined the inter-relations of cigarette use, psychological distress, physical activity, self-rated health, coping, and body mass index (BMI) among subjects aged 11-34 years in a drug abuse prevention trial, the Midwestern Prevention Project (MPP), in Kansas City and Indianapolis.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Effects of parent stress on weight-related parenting practices and child obesity risk

PDF

Aggression and victimization as predictors of drug use in early adolescence

PDF

Contextualizing social network influences on substance use among high risk adolescents

PDF

Adolescent stress and styles of coping: Predictors and moderators of psychological distress

PDF

Sociocultural stress, coping and substance use among Hispanic/Latino adolescents

PDF

Anxiety symptoms and nicotine use among adolescents and young adults

PDF

The vicious cycle of inactivity, obesity, and metabolic health consequences in at-risk pediatric populations

PDF

Effects of near-roadway air pollution exposure on obesity, obesity-related behavior, and neurobehavioral deficits during peripuberty

PDF

Emotional intelligence and smoking risk factors in early adolescents

PDF

Investigating factors that influence peer relationships and obesity during middle childhood

PDF

Effects of flavorings in electronic cigarettes on the use and appeal of e-cigarettes among youth and adults

PDF

A network analysis of online and offline social influence processes in relation to adolescent smoking and alcohol use

PDF

Distress tolerance and mindfulness disposition: associations with substance use during adolescence and emerging adulthood

PDF

The effects of mindfulness on adolescent cigarette smoking: Measurement, mechanisms, and theory

PDF

Influences of specific environmental domains on childhood obesity and related behaviors

PDF

Tobacco and marijuana surveillance using Twitter data

PDF

Prospective associations of stress, compulsive internet use, and posttraumatic growth among emerging adults

PDF

Factors and correlates of sexual behaviors among young adults from continuation high schools

PDF

The Internet activities, gratifications, and health consequences related to compulsive Internet use

PDF

Exploring the role of peer influence, linguistic acculturation, and social networks in substance use

Asset Metadata

Creator Kaur, Guneet (author)

Core Title Adolescent cigarette use and psychological distress: relation to adult obesity risk

School Keck School of Medicine

Degree Doctor of Philosophy

Publication Date 05/02/2010

Defense Date 03/03/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag cigarette use,OAI-PMH Harvest,obesity,Preventive Medicine (Health Behavior),psychological distress

Language English

Advisor Pentz, Mary Ann (committee chair), Chou, Chih-Ping (committee member), Prescott, Carol A. (committee member), Sussman, Steven (committee member)

Creator Email guneetka@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1218

Unique identifier UC171555

Identifier etd-Kaur-20080502 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-71899 (legacy record id),usctheses-m1218 (legacy record id)

Legacy Identifier etd-Kaur-20080502.pdf

Dmrecord 71899

Document Type Dissertation

Rights Kaur, Guneet

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu