.3
an».
1...

.163":

East ‘ g

s , .. a. 3.223.. ............2 .
bhufihfig‘. k.
I‘..l..¥bfis.;

. .Vbxifitfii i

2!..wa 4;; end“.
g‘ mat...

it

. Lu!!!

axial.
:{I . 5".
.LhJIHSJD

.: ii?!
- J

.r

. .92.
all .1! .5}?!

1:... 7.5.3)... n if
unswmr.‘

M} ’It; Al

.

Iv fril¥l 1 ,
inf-Ln}.

K).
h”?! An. 2].?
{1.3!

150?. 11.1.
15.53! .. .5!
g‘fiifl'f“

Egg: ‘ 5%.... {n agfumg é gagig V .,. V . , , _, ‘ ,

1 131.5... 1.

 

 

 

.5

 

WESIS

1001/

This is to certify that the

dissertation entitled
RISK FOR EARLY SMOKING ONSET IN A HIGH—RISK POPULATION:
CONTRIBUTIONS OF EARLY CHILD BEHAVIORAL RISK AND
HETEROGENEITY OF PARENTAL SMOKING AND ALCOHOLISM

presented by

Eun Young Mun

has been accepted towards fulfillment
of the requirements for

Ph.D . degree in Psychology

 

 

Major pr ssor

Date 4 /16!2002

MSU i: an Affirmative Action/Equal Opportunity Institution 0- 12771

 

 

LIBRARY

Michigan State

University

 

 

PLACE IN RETURN BOX to remove this c
TO AVOID FINES return on or
MAY BE RECALLED with earlier

heckout from your record.

before date due.

due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRCIDaIeDue.p65-p. 15

RISK FOR EARLY SMOKING ONSET IN A HIGH-RISK POPULATION:
CONTRIBUTIONS OF EARLY CHILD BEHAVIORAL RISK AND
HETEROGENEITY OF PARENTAL SMOKING AND ALCOHOLISM
By

Eun Young Mun

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Psychology

2002

'1:

1L

"b“wgn-I
b3¥§>5usli

1 «+7.. ~
Lniytnérnn

Cirghin

h

5 ‘ .‘
tu\dh‘
o

ABSTRACT
RISK FOR EARLY SMOKING ONSET IN A HIGH-RISK POPULATION:
CONTRIBUTIONS OF EARLY CHILD BEHAVIORAL RISK AND
HETEROGENEITY OF PARENTAL SMOKING AND ALCOHOLISM
By

Eun Young Mun

This study examined early smoking onset in a high-risk population of adolescents.
Participants were 281 families with biological parents and their children (281 sons and 88
daughters) residing in the Mid-Michigan area who completed at least two of the first four
assessments of the ongoing University of Michigan — Michigan State University
Longitudinal Study. Parental cigarette smoking and alcoholism were analyzed using a
group-based semi-parametric modeling approach, resulting in three types of smokers
(Heavy smokers, Light smokers, and Heavy-to-light smokers) and two types of alcoholics
(Alcoholism I and Alcoholism 11). Long-term paternal Heavy smoking, in combination
with long-term maternal smoking, sufficiently elevated the chance of early smoking onset
in offspring. Similarly, children in two-parent families where both parents were
chronically alcoholic (Alcoholism H) had an increased likelihood of early smoking onset.
However, smoking or alcoholism by just one parent in two-parent families did not pose
much risk for early smoking onset in offspring.

Adolescents who started smoking by age 14 were different on domains that traced
back to as early as prenatal development. Adolescents with early smoking onset showed
a higher exposure to maternal daily cigarette smoking. Moreover, they were different on
temperament and behavioral characteristics as early as ages three to five. Adolescents

with early smoking onset were more reactive and approaching. In addition, their mothers

Eun Young Mun
perceived them as having higher levels of negative affect (anxious/depressed), attention
problems, delinquent behavior, and aggressive behavior. Structural equation modeling
analysis on these parental and individual risk factors for early smoking onset revealed
that maternal cigarette smoking during pregnancy led to early smoking onset in offspring

via early child negative affect (i.e., anxious/depressed).

Cepyright by
Eun Young Mun
2002

To Feng

DMD
mmnyuhohli
Mdnwmn
finandxoni
timer Indtbied
mmmd
Darn: bean; b
Tfiflmg
Mama
heads and lam

This res
iomdanon, an

WWW“

i

ACKNOWLEDGEMENTS

It took a while to this end but I am jubilant that it is time to express my gratitude
to many who have helped me along the way. I have greatly enjoyed my time at Michigan
State University both academically and personally. Dr. Hiram E. Fitzgerald and Dr.
Alexander von Eye have nurtured and cared for my growth as a psychologist and I am
forever indebted to them for their unwavering support and encouragement. I would also
like to thank Dr. Robert A. Zucker for his guidance in the literature of alcoholism.
Despite being busy all the time, he has always responded to my needs and provided
insightful suggestions and comments. I would like to express my gratitude to Dr. Joel
Ni g for his valuable observations and suggestions. Finally, I would like to thank my
friends and family. Without them the long joumey would not have been as enjoyable and

fim.

This research was supported by a Student Award Program grant (grant #
276.SAP) to Eun Young Mun from the Blue Cross and Blue Shield of Michigan
Foundation, and, in part, by grants to R. A. Zucker and HE. Fitzgerald from the National

Institute on Alcohol Abuse and Alcoholism (NIAAA grant #5 R01 AA 07065).

vi

L'SI OF TAB
LIST OF FIG!

NROD‘LCI
LIIergerer
Heterogcr
. imi}
Prim: E

Riiii l
lndire

Th5 (LIFT:

METHOD
Parking:

5"”? -1 ‘
PLCAAO‘il .

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................. ix
LIST OF FIGURES ........................................................................................................... xi
INTRODUCTION ............................................................................................................... 1
Intergenerational Transmission of Smoking .................................................................. 5
Heterogeneity of Parental Smoking ............................................................................... 6

A Family History of Alcoholism ................................................................................. 11
Prenatal Exposure to Maternal Cigarette Smoking ...................................................... 13
Risk factor for cigarette smoking in offspring ....................................................... 13
Indirect link to early smoking onset via behavioral characteristics ....................... 14

The Current Study ........................................................................................................ 17
METHOD .......................................................................................................................... 22
Participants ................................................................................................................... 22
Parental Measures ........................................................................................................ 25
Parental smoking at waves 1 through 4 ................................................................. 25
Parental alcohol use disorder at waves 1 through 4 ............................................... 25
Rearrangement of data on parental smoking and alcoholism ................................ 26
Parental antisocial behavior at wave 1 ................................................................... 28
Parental depression at wave 1 ................................................................................ 28
Parental education and occupation at wave 1 ........................................................ 29

Child Measures ............................................................................................................ 29
Prenatal exposure to maternal smoking and drinking ............................................ 29

Child temperature at wave 1 .................................................................................. 30

Child behavioral characteristics at waves 1 and 4 ................................................. 30
Smoking onset ........................................................................................................ 31
Missing Data Estimation .............................................................................................. 32
RESULTS ......................................................................................................................... 36
INTERGENERATIONAL TRANSMISSION OF SMOKING ........................................ 39
Developmental Patterns of Smoking in Adulthood ..................................................... 39
Characteristics of Subtypes of Smokers ...................................................................... 45
Parental Smoking Subtype and Early Smoking Onset in Adolescent Children ........... 52

A FAMILY HISTORY OF ALCOHOLISM ..................................................................... 6O
Developmental Patterns of Alcoholism ....................................................................... 6O
Characteristics of Subtypes of Alcoholism .................................................................. 66
Commonality Between Subtypes of Smokers and Alcoholics .................................... 73
Parental Alcoholism Subtype and Early Smoking Onset in Adolescent Children ...... 78

vii

grill AND C
EARLY SMOK
Prenatal Ex]
Precursors 2
vii". Ea:

.9.le 5 l0 EA

DISCQSSION.

Pair-I a): II

Alcono
th‘.‘

limitation:

A3?E.\DL\ A
.IDPEXDIX 8
APPENDIX (
APPEXDIX [
APPENDIX E
WENDIX I
.tDPENDlX (
APPENDEX

REPEREXCI

EARLY AND CONCURRENT CHARACTERISTICS OF ADOLESCENTS WITH

EARLY SMOKING ONSET ............................................................................................. 84
Prenatal Exposure to Maternal Smoking and Drinking ............................................... 84
Precursors and Concurrent Characteristics of Adolescents

with Early Smoking Onset .................................................................................... 86

PATHS TO EARLY SMOKING ONSET ......................................................................... 94

DISCUSSION .................................................................................................................. 101
Pathways to Early Smoking Onset ............................................................................. 102

Parental smoking .................................................................................................. 102
Parental alcoholism .............................................................................................. 103
Early characteristics of adolescents with early smoking onset ............................ 105
Paths to early smoking onset ................................................................................ 107
Heterogeneous Developmental Patterns of Smoking and Alcoholism ...................... 110
Smoking types ...................................................................................................... 110
Alcoholism types ................................................................................................. 113
Quantitative differences versus qualitative typology ........................................... 115
Limitations and Future Directions ............................................................................. 115

APPENDIX A. Number of Participants by Birth Cohort ............................................... 121

APPENDD( B. Distribution of Birth Cohort Across Year of Assessment .................... 123

APPENDIX C. Matrix of Missingness Pattern .............................................................. 143

APPENDD< D. Number of Cases in the TRAJ Procedure ............................................. 150

APPENDD( E. Descriptive Statistics of Subtypes of Smoking and Alcoholism .......... 157

APPENDIX F. Descriptive Statistics of Adolescent Sons ............................................. 161

APPENDD( G. Daughters .............................................................................................. 163

APPENDEX H. Questionnaires ....................................................................................... 171

REFERENCES ................................................................................................................ 236

viii

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

Table 10.

Table 11.
Table 12.
Table A1.
Table A2.
Table Bl.
Table BZ.
Table BB.
Table B4.
Table B5.

Table B6.

LIST OF TABLES

Descriptive Statistics of All Participants ......................................................... 24
Model Comparisons of Parental Smoking ....................................................... 41
Growth Parameter Estimates for Each of Three Smoking Groups .................. 41

MANOVA Results on Demographic Characteristics and

Psychopathologies by Smoking Subtype and Gender ..................................... 48
Configurations of Parental Smoking Patterns and Early Smoking

Onset Among Adolescent Sons ................................................................. 57-58
Model Comparisons of Parental Alcoholism ................................................... 61
Growth Parameter Estimates for Each of Two Alcoholisms ........................... 64

MANOVA Results on Demographic Characteristics and

Psychopathologies of Two Alcoholisms .......................................................... 68
Associations Among Subtypes of Smokers and Alcoholics ............................ 77
Configurations of Parental Alcoholism Patterns and Early Smoking

Onset Among Adolescent Sons ................................................................. 81 -82
Precursors and Concurrent Characteristics of Smoking Onset Status ............. 89
Polychoric Correlation Matrix ......................................................................... 99
Number of Adolescents by Birth Cohort ....................................................... 121
Number of Parents by Birth Cohort ............................................................... 122
Birth Cohort by Year of Assessment at Wave 1: Sons .................................. 123
Birth Cohort by Year of Assessment at Wave 4: Sons .................................. 124
Birth Cohort by Year of Assessment at Wave 1: Daughters ......................... 125
Birth Cohort by Year of Assessment at Wave 4: Daughters ......................... 126
Birth Cohort by Year of Assessment at Wave 1: Fathers .............................. 127
Birth Cohort by Year of Assessment at Wave 2: Fathers .............................. 129

ix

Table B7. Birth Cohort by Year of Assessment at Wave 3: Fathers .............................. 131

Table B8. Birth Cohort by Year of Assessment at Wave 4: Fathers .............................. 133
Table B9. Birth Cohort by Year of Assessment at Wave 1: Mothers ............................. 135
Table B10. Birth Cohort by Year of Assessment at Wave 2: Mothers .......................... 137
Table B11. Birth Cohort by Year of Assessment at Wave 3: Mothers .......................... 139
Table BIZ. Birth Cohort by Year of Assessment at Wave 4: Mothers .......................... 141
Table C1. Matrix of Missingness Pattern for Parental Measures ................................ 143
Table C2. Matrix of Missingness Pattern for Adolescent Sons’ Measures ................. 144
Table C3. Matrix of Missingness Pattern for Adolescent Daughters’ Measures ........ 147
Table D1. Number of Cases in the TRAJ Analysis of Smoking ................................. 150
Table D2. Number of Cases in the TRAJ Analysis of Alcoholism: Men .................... 152
Table D3. Number of Cases in the TRAJ Analysis of Alcoholism: Women .............. 155
Table E1. Descriptive Statistics of Smoking Subtypes: Men ...................................... 157
Table E2. Descriptive Statistics of Smoking Subtypes: Women ................................ 158
Table E3. Descriptive Statistics of Alcoholism Subtypes: Men .................................. 159
Table E4. Descriptive Statistics of Alcoholism Subtypes: Women ............................ 160
Table F 1. Factors of Early Smoking Onset: Maternal Ratings of Sons ...................... 161
Table F2. Factors of Early Smoking Onset: Paternal Ratings of Sons ....................... 162
Table G1. Parental Smoking Patterns and Early Smoking Onset of Daughters .......... 163
Table GZ. Parental Alcoholism Patterns and Early Smoking Onset of Daughters ...... 165
Table G3. MANOVA Results on Characteristics of Daughters ................................. 167
Table G4. Factors of Early Smoking Onset: Maternal Ratings of Daughters ............. 169
Table GS. Factors of Early Smoking Onset: Paternal Ratings of Daughters ............... 170

Figure 1.
Figure 2.
Figure 3.
Figure 4.
Figure 5.
Figure 6.
Figure 7.
Figure 8.
Figure 9.
Figure 10.
Figure 11.
Figure 12.
Figure 13.
Figure 14.
Figure 15.
Figure 16.
Figure 17.
Figure 18.
Figure 19.
Figure 20.

Figure 21.

LIST OF FIGURES

A heuristic model of early smoking onset in children of smoking parents ..... 20

A simplified model of pathways to early smoking onset ................................. 21
Three distinctive developmental patterns of smoking in adulthood ................ 44
Years of education and smoking subtype ........................................................ 49
Occupational status and smoking subtype ....................................................... 49
Childhood conduct problems and smoking subtype ........................................ 50
Adulthood antisocial behavior and smoking subtype ...................................... 50
Current depressive symptoms and smoking subtype ....................................... 51
Worst-ever depressive symptoms and smoking subtype ................................. 51
Parental smoking subtypes and adolescent children’s smoking onset ............. 56
Developmental patterns of alcoholism among men ...................................... 62
Developmental patterns of alcoholism among women ................................. 63
Years of education and alcoholism subtype .................................................. 69
Occupational status and alcoholism subtype ................................................ 69
Childhood conduct problems and alcoholism subtype ................................. 7O
Adulthood antisocial behavior and alcoholism subtype ............................... 70
Current depressive symptoms and alcoholism subtype ................................ 71
Worst-ever depressive symptoms and alcoholism subtype .......................... 71
Self-reported depressive symptoms and alcoholism subtype ....................... 72
Associations among subtypes of smoking and alcoholism ........................... 76
Parental alcoholism and smoking onset among adolescent sons .................. 80

xi

57.911

t thi“ "'
u

M 1‘
Eli-IE n.)-

'I

,.
b" wrp

5.;nflu - .
5

pm, a;
A AEQ‘I 5 .U -
u

7 1
h;:e-6

Pm 1‘?

a.’.ub.u,

5

I}... 1g
“5‘- us“

..
““m: ‘

In... _ ~
V

.g. ‘

n.re;0,

Egreil.

ran ,
PlLilu

Figure 22.
Figure 23.
Figure 24.
Figure 25.
Figure 26.
Figure 27.
Figure 28.
Figure 29.
Figure 30.

Figure 31.

Prenatal exposure to maternal smoking and children’s smoking .................. 85

Maternal ratings of child temperament and smoking onset .......................... 9O
Paternal ratings of child temperament and smoking onset ........................... 90
Maternal ratings of Anxious/depressed and smoking onset ......................... 91
Maternal ratings of Attention problems and smoking onset ......................... 91
Maternal ratings of Delinquent behavior and smoking onset ....................... 92
Maternal ratings of Aggressive behavior and smoking onset ....................... 92
Paternal ratings of Delinquent behavior and smoking onset ........................ 93
Paternal ratings of Aggressive behavior and smoking onset ........................ 93
A model of intergenerational transmission of smoking: ............................ 100

xii

Ihel'aate‘ 314‘
grit“; If] Eh:
32:11 annual ai
Kitsch-er I:
M3111". but
.. ICDC. l
mature mo
$21105. respec
making is a r

Epider
flit the last I}
m" adults.
Cowlimces

Smarts;

INTRODUCTION

Cigarette smoking is a major cause of preventable premature death and disease in
the United States. During 1990 and 1994, 351 out of 100,000 deaths were attributed to
smoking in the US. (Centers for Disease Control and Prevention (CDC), 2001), and the
total annual average mortality due to smoking was approximately 430,000 (CDC, 1999;
Malarcher et al., 2000). In addition, the economic burden of tobacco use, in parallel to
the health burden, has been estimated annually at $53.3 billion in direct medical costs
alone (CDC, 1999; Malarcher et al., 2000). Indirect costs associated with morbidity and
premature mortality from cigarette smoking have been estimated at $6.9 billion and $40.3
billion, respectively (CDC, 1999; Malarcher et al., 2000). These statistics indicate that

smoking is a major health problem in the United States.

Epidemiological studies have reported an overall decrease in smoking prevalence
over the last three decades in the United States, due to successful smoking cessation
among adults, since the first report by the Surgeon General on detrimental health
consequences of smoking appeared in 1964 (U .S. Department of Health and Human
Services (U SDHHS), 1994). However, there is an indication that smoking prevalence,
especially among girls and women, may be increasing in recent years (U SDHHS, 2001).

Furthermore, there are growing concerns that rates of smoking initiation have not
decreased over the past decade (U SDI-IHS, 1994) and data for the 1990s suggest that
rates of smoking initiation among adolescents are on the rise (Mendez, Warner, &
Courant, 1998). Increasing rates of smoking initiation are a major concern especially
since risk for initiation into cigarette smoking is mostly over by age 20 (Chassin, Presson,

Rose, & Sherman, 1996b; Chassin, Presson, Sherman, & Edwards, 1991; Chen &

Patel. 19%;[1
reflect a series oi
311%; Cher.
PG? '11: dei'elopn
airtidiri and m
Keller ski. & H

as 13 in part};

19951 Patten. C
rte iParton et
termite disc
tug 156d by a
1994i.

IIius. e
511010125 needs
Warm agaii
“03% adolesc
mere856d restr
3930; L'SDH}
Pio‘i‘lded by {P1
Fling initia

Kandel, 1995; USDHHS, 1994), and that changes in smoking status after age 20 mostly
reflect a series of cessation attempts and relapses rather than new initiation (Chassin et
al., 1996b; Chen & Kandel, 1995). Adolescents with early smoking initiation are at risk
for the development of tobacco dependence, and are subject to higher rates of the
morbidity and mortality associated with cigarette smoking in adulthood (Heishman,
Kozlowski, & Henningfield, 1997; Jackson, 1998). Onset of cigarette smoking prior to
age 13, in particular, is related to increased risks for adult daily smoking (Chassin et al.,
1996b; Chassin, Presson, Sherman, & Edwards, 1990; Hanna & Grant, 1999; Jackson,
1998; Patton, Carlin, Wolfe, Hibbert, & Bowes, 1998), lower cessation and higher relapse
rates (Patton et al., 1998), and DSM-IV diagnoses of drug dependence/abuse and lifetime
depressive disorder (Hanna & Grant, 1999). In addition, tobacco is often the gateway
drug used by adolescents who later use alcohol, marijuana, and other drugs (U SDHHS,
1994)

Thus, early onset of cigarette smoking is a behavior of significance whose
etiology needs to be studied in order to set up effective prevention as well as intervention
programs against smoking. So far, public health efforts to control smoking initiation
among adolescents have been geared toward school-based programs and reinforcement of
increased restrictions on the advertisement and sale of tobacco products (CDC, 1994,
2000; USDHHS, 1994). The underlying rationale for the school-based programs is
provided by the findings from mostly cross-sectional or retrospective studies that show
smoking initiation is predicted by an epidemic or exposure model that requires contacts

with smoking peers, parents, and/or siblings (Bobo & Husten, 2000; Rowe, Chassin,

V
w ‘
II. LAC-I

.
«i w K.

\
UL >.A.“.'

Presson, & Sherman, 1996; Rowe & Rodgers, 1991), with the presence of smoking peers
as the single best predictor of adolescent smoking (USDHHS, 1994).

There is little doubt that socialization with smoking peers plays an important role
in smoking behavior among adolescents, but some caution is warranted. First, predictors
of smoking from cross-sectional studies, including socialization factors, are not fully
supported in longitudinal studies of smoking (Chassin, Presson, Montello, Sherman, &
McGrew, 1986; Engels, Knibbe, & Drop, 1999). On the contrary, there is evidence
suggesting that influences of smoking peers may not be as potent as previously
understood when studied longitudinally. A study on non-smoking children initially in the
third or fifth grade found that it was not smoking peers but smokers at home that
predicted higher rates of smoking initiation three years later (Patton et al., 1998).

The issue of comparability of evidence from cross-sectional and longitudinal
research studies is not limited to smoking research (e.g., Ge, Lorenz, Conger, Elder, J r.,
& Sirnons, 1994). It is difficult to establish direction of the causal relationship between
variables from a snap shot approach because it is not clear what occurred first out of two
related events. Although temporal priority does not always establish the causal direction
between events, it is the “single most effective means” of doing so, and that fulfils one of
the three conditions of causation (i.e., isolation, association, and direction of causation;
Bollen, 1989, pp. 40-79). The influences of peer socialization on substance abuse among
adolescents in cross-sectional studies, for example, may reflect both peer-selection (i.e.,
smoking-prone adolescents choose to socialize with peers of the same kind), and peer
socialization (i.e., adolescents start smoking because of peer pressure and modeling of

smoking behavior) processes (Curran, Stice, & Chassin, 1997).

Second. re
Itinnel‘. bctore .
rt psychosociat
aidsspread exper
otters instep a
P01361163U.C0111
Iitotine has rid:
rent that initiai
Sport t a kc
Panerieau et at.
smokers may be
nitiai smokin

N.
V

New spit

2L1 - '
ittnntical rnri

   

tiring adults in
Or 1999), SI
@fprrssion alcc
mortal perso
1999', Picciotto.
tnorttdityor
We Stress. er

P“ ”Winingit

19961

Second, recent studies suggest that the risk structure of smoking initiation may
exist well before adolescents try a first cigarette, possibly involving genetic, biological,
and psychosocial factors. It has long been wondered ‘thy, with extensive exposure and
widespread experimentation with tobacco, some people become nicotine dependent,
others develop a pattern of occasional use, and still others avoid it entirely” (O. F.
Pomerleau, Collins, Shiffrnan, & C. S. Pomerleau, 1993). Although initial sensitivity to
nicotine has widely been recognized as a factor for smoking initiation, it is relatively
recent that initial sensitivity to nicotine and its inverse relation to tolerance have been
proposed as a key to nicotine dependence (i.e., “sensitivity” model of tolerance; see
Pomerleau et al., 1993). According to this model of nicotine dependence, regular
smokers may be those who are constitutionally sensitive to nicotine, react aversely to
initial smoking, and quickly develop tolerance to nicotine.

New epidemiological evidence sheds further light on a constitutional or
pathological model of etiology of smoking. As the prevalence of cigarette smoking
among adults in the US. declined from 42% in 1965 to approximately 25% in 19903
(CDC, 1999), smoking has increasingly become linked to people with conditions such as
depression, alcoholism, attention deficit-hyperactivity disorder, conduct disorder,
antisocial personality disorder, schizophrenia, and Alzheimer’s disease (Hanna & Grant,
1999; Picciotto, 1998; C. S. Pomerleau, 1997). In line with the observation of the
comorbidity of nicotine with other conditions, nicotine is known to help relaxation,
reduce stress, enhance attention, improve cognitive function, and regulate mood, and it
has increasingly been studied as a medication for many medical conditions (Benowitz,

1 996).

Thus. it is
“3h at'ttlnera. tll
line. It Callas. I
canotiditt of n
accushil amonj
people Willi low
ages. that ther
time and betel
identified earlie

The maj
Inchopatholttg
making. drink
retried age 01
1103,2031). 1
Pier to smoki
concurrently.

Pit‘fnls.

r
iii-”e

In the
off. - -
3?an 13 It

in i

PtSDHHS l

Thus, it is plausible that nicotine is more reinforcing as a stimulant for individuals
with a vulnerability factor, including behavioral and affective conditions (e.g., Hughes,
Rose, & Callas, 2000; Picciotto, 1998; O. F. Pomerleau et al., 1993). In addition to the
comorbidity of nicotine, over the past three decades quit attempts have been more
successful among people with higher education and socioeconomic status compared to
people with lower socioeconomic background (C. S. Pomerleau, 1997). These findings
suggest that there are some individuals who are more susceptible to cigarette smoking
above and beyond the influences of peer socialization, and that their susceptibility may be
identified earlier.

The majority of studies that revealed the associations of smoking with other
psychopathologies, however, relied on retrospective, self-reported age of onset of
smoking, drinking, and use of other illicit drugs. The reliability and accuracy of self-
reported age of onset can be inconsistent even within a one-year time period (Johnson &
Mott, 2001). Therefore, it is important that the individual risk attributes of adolescents
prior to smoking onset, and smoking onset itself should be identified prospectively and
concurrently. The current study aims to investigate predictors of early smoking onset in a
high-risk population fiom a prospective longitudinal study of adolescents and their
parents.

Intergenerational Transmission of Smoking

In the literature, the relationship between parental smoking and smoking of the
offspring is relatively understudied. The limited existing studies show inconsistent
findings whether parental smoking is a risk factor for adolescent offspring’s smoking

(U SDI-IHS, 1994). However, there is a growing body of new evidence that parental

racing is related
3111.111 ehildre
Crash. Presson.
flit}; Gnesler. K;

199: Rowe et a]-

This phena
inter olexplana

tlcopmans et al., l
ninerahilit)’ repres
rental etposure t
Patel et al., 1994
Phil. and 5t pare
tether exhaustit-I
rattling nnset. '
minimum?
The current stt
rattan m.
M expc

Plozttlems in C

I

.-
filf
1": C.“

W
[11033)]!

smoking is related to children’s smoking as indicated by a greater prevalence rate of
smoking in children of smokers than in children of nonsmokers (Chassin et al., 1996b;
Chassin, Presson, Todd, Rose, & Sherman, 1998; Cornelius, Leech, Goldschmidt, & Day,
2000; Griesler, Kandel, & Davies, 1998; Kandel, Wu, & Davies, 1994; Patton et al.,
1998; Rowe et al., 1996).

This phenomenon of intergenerational transmission of smoking is open to a
number of explanations, including 1) genetic transmission of susceptibility to nicotine
(Koopmans etal., 1999; Madden et al., 1999; True et al., 1997), 2) a common familial
vulnerability represented by family history of alcoholism (Hanna & Grant, 1999), 3)
prenatal exposure to maternal smoking (Cornelius et al., 2000; Griesler et al., 1998;
Kandel et al., 1994), 4) attentional and behavioral problems in childhood (Griesler et al.,
1998), and 5) parenting behavior (Chassin et al., 1998). These five explanations are
neither exhaustive nor competing with one another as the mechanism of adolescent
smoking onset. It is more likely that early smoking onset in adolescence reflects a
complex interplay among distal and proximal factors via direct and indirect pathways.
The current study focuses on the interplay among a familial vulnerability and an
individual vulnerability. Parental smoking and alcoholism represent the former, while an
prenatal exposure to maternal smoking, child early temperament, and behavioral
problems in childhood embody the latter in the present study.

Heterogeneity of Parental Smoking

In the literature on parental smoking as a risk factor for adolescent children’s

smoking initiation, the possibility of differential influences of parental smoking has

relatively been overlooked. Studies on clinical populations of smokers have noted

1,3... -

I
LL“; 511.0!

a _.
L‘ICIO

(a.

at:
atdztien. tl

I531. A: P;

[hi
adeiescem ‘
mmbldnc:
Wailing Q
Where the hi
Ha‘ilt'lns, &
I996; ZUCke
PPPOlhesiZec

mllldilgn hm

1% 1ban in

differences between heavy smokers and light smokers (O. F. Pomerleau et al., 1993), or
regular smokers and “chippers” (Kassel, Shiffman, Gnys, Paty, & Zettler-Segal, 1994;
Shiffinan, Kassel, Paty, Gnys, & Zettler-Segal, 1994a, Shiffrnan, Paty, Kassel, Gnys, &
Zettler-Segal, 1994b) on various smoking-related measures, including sensitivity to
nicotine, attitude toward smoking, and familial history of smoking. In particular,
Shiffrnan and his colleagues have reported that there are a small number of long-term
light smokers or “chippers” who smoke no more than one to five cigarettes a day but do
not develop nicotine dependence (Kassel et al., 1994; Shiffrnan et al., 1994a, 1994b). In
addition, the age of smoking onset appears to be related to nicotine dependence (Breslau,
Fenn, & Peterson, 1993). Of those who have ever smoked, only a quarter to a third
develop nicotine dependence, and those who initiated smoking between 14 and16 were
more likely to become dependent than those who initiated smoking at an older age
(Breslau et al., 1993).

Unfortunately, parental smoking has rarely been studied as a risk factor for
adolescent children’s smoking initiation, with the exception of studies of familial
resemblance (e.g., Eysenck, 1980; cf. O.F. Pomerleau et al., 1993), due in part to the
prevailing epidemic view of smoking initiation. In contrast to the study of alcoholism
where the heterogeneous nature of alcoholism (Cloninger, 1987; Hill, White, Chung,
Hawkins, & Catalano, 2000; Schulenberg, O’Malley, Bachman, Wadsworth, & Johnston,
1996; Zucker, 1987, 1994) and the differential risks to their offspring have been
hypothesized and tested (Zucker, Ellis, Bingham, & Fitzgerald, 1996), studies of smoking
initiation have focused more on differential factors involved in early stages of smoking

rather than individual differences in smoking pathways. For example, social contact with

rioting peers and
no by adolcst
Sign Kill); it he:
brpersrstent smol
at. 1993 or ht;
Variable-c:
rat-individual ch
lites. 3013); Bert
the ot'the critical
eturn'oersome t
hsorne studies. a
mMmumm
tooling). or to q
stroking onset. is
Earth not only
litterindividna]
films age char
mild“ over 1:

ul

”Wiented ;
Wide the Prfise
P”“elttmientat p

thmmsms (I

smoking peers and parents has been suggested as a major mechanism for the initiation of
smoking by adolescents (Rowe et al., 1996; Rowe & Rodgers, 1991; Rowe, Rodgers, &
Gilson, 2000) whereas genetic contributions have been suggested to be more prevalent
for persistent smoking compared to the initiation of smoking (Madden et al., 1999; True
et al., 1997) or high quantity of smoking (Koopmans et al., 1999).

Variable-centered approaches, however, do not capture individual differences in
intra-individual change across the life span as well as the person—oriented approaches do
(Bates, 2000; Bergman & Magnusson, 1997; Magnusson, 1998, 1999, 2000).

One of the critical issues related to smoking research is that variable-centered approaches
are cumbersome when it comes to handling the timing of smoking transitions across time.
In some studies, age differences are studied as a covariate for the changes from non-
srnoker to experimenter (i.e., initiation of smoking), to regular smoker (i.e., persistence of
smoking), or to quitter. However, there is great variability in timing of smoking —
smoking onset, regular smoking, and quit attempts. Furthermore, in developmental
research not only ag differences but also age changes are of interest. Age differences
(inter-individual variability) refer to differences found among different age groups,
whereas age changes (intra-individual variability) refer to changes within the same
individual over time (Wohlwill, 1973). Studies of distinctive growth patterns from
person-oriented approaches, in particular, can be characterized by many parameters that
include the presence, direction, rate, level, and timing of change, general shape of a
developmental pattern, and age corresponding to specified values of any of those

characteristics (Wohlwill, 1973; cf. Bates, 2000).

., l
\P&.uih11 -
IV 'uv 1 ‘
Tl \ CLELI L

:- .4”...

SECRETS tt 6
33.250 mod»:

Earl
Sable and h
Siblished
l~ as the 1
PM reach:
Who Stopper

“Cline rei

A recent study by Chassin and her colleagues (Chassin, Presson, Pitts, &
Sherman, 2000) is one of the few studies that investigated individual differences in intra-
individual changes in smoking behavior from a population-based cohort sample over a
lengthy period. Using a semiparametic mixture group-based modeling (N agin, 1999;
Nagin & Tremblay, 1999, 2001) designed to investigate heterogeneous developmental
patterns (B. Muthén & L. K. Muthén, 2000; B. Muthén, 2001; Nagin, 1999), four
developmental patterns of cigarette smoking from ages 11 to 31 were identified: Early
Stable (12%), Late Stable (16%), Experimenter (6%), and Quitter (5%) groups. The
abstainers or non-smokers (approximately 60%) and relatively a small number of erratic
smokers were a priori identified and excluded from the semiparametic mixture group-
based modeling analysis.

Early Stable smokers had an early onset of smoking with a steep escalation to a
stable and high consumption level over time. Late Stable smokers were those who
established their regular smoking relatively late, and whose levels of smoking were not as
high as the Early Stable group. Experimenters were those who started smoking early but
never reached levels of smoking by either the Early Stable or the Late Stable smokers and
who stopped smoking before age 20. And finally, Quitters were those who started
smoking relatively late, who reached high levels of chronic smoking, and who quit
smoking before age 25.

These four groups of smokers and quitters were different in various psychosocial
measures, with Early Stable smokers reporting high tolerance for deviance, high external
locus of control, low levels of parental support, and least likelihood of obtaining college

education. In addition, they were also more likely to have parents and friends who

- I
art-uteri 3PM

ot'srnc-kzng
stoked hel
ofcortrol a:
5.: La
ot'srnolsing
Trot:
blotting It
also be {our
iii 3) Are
Ire first iss
tie. elopmey

ellPertinent;

smoked, and held most positive beliefs about the health and psychological consequences
of smoking. On the other hand, Late Stable smokers had few friends and parents who
smoked, held relatively negative beliefs about smoking, and reported high internal locus
of control and parental support. Thus, even though smokers of the Early Stable and Late
Stable trajectories were habitual smokers, they were distinctively different on measures
of smoking consumption and psychosocial measures.

From the findings of Chassin et al.’s study (2000), we can naturally ask the
following two key questions: 1) Can heterogeneous developmental courses of smoking
also be found during the much wider age span in adulthood, ranging from the 205 to 508?
and, 2) Are adolescent children of smokers at different risks for early smoking onset?
The first issue addresses the fact that Chassin et al.’s study (2000) covered a
developmental period from adolescence to young adulthood where smoking initiation and
experimentation most often occur. Therefore, changes in the patterns of cigarette
smoking are more likely. Once smoking is established, however, the possibility exists
that there may not be much variation in smoking. Alternatively, adult smokers may
consist of distinctively different subpopulations that differ on psychosocial characteristics
as well as their smoking characteristics such as quantity and duration of cigarette
smoking. The second question is a natural extension to the first issue: Are there different
levels of risk tied to parental smoking subtypes for early smoking onset? Are there
specific patterns of parental smoking linked to adolescent children’s early smoking
onset? For example, are habitual heavy smokers, occasional smokers, and current
abstainers with a past history of heavy smoking the same in levels of risk for early

smoking in offspring?

10

 

adolescent
III}. Maid:
hfin&1
families tBie

llfailt It};

aeo‘nolism i

One
associated It
SIC 6.139 g

A L amilv History of Alcoholism

There is evidence that alcohol and tobacco go hand in hand for many individuals
in adolescent as well as adult populations (Grant, 1998; Hughes, 1995; Hughes et al.,
2000; Madden, Heath, Starmer, Whitfield, & Martin, 1995; O. F. Pomerleau, 1995;
Shiffman & Balabanis, 1995) and that both alcoholism and habitual smoking run within
families (Bierut, Schuckit, Hesselbrock, & Reich, 2000; Madden, Bucholz, Martin, &
Heath, 2000). On the basis of this comorbidity, some researchers suggest that there may
be a common familial vulnerability for the use of alcohol and tobacco among individuals
with a positive family history of alcoholism (Grant, 1998; Hanna & Grant, 1999; Sher,
Gotham, Erickson, & Wood, 1996). The mechanisms of why a family history of
alcoholism is related to use of alcohol and tobacco in offspring are not yet understood.

One possibility is that the common genetic mechanism may exist that is
associated with but not limited to both alcohol and tobacco use (Lennan et al., 1999).
SLC6A3-9 genotypesl that are known to be associated with late initiation of smoking and
smoking cessation, for example, may also account for reduced need for novelty and
reward by external stimuli such as alcohol and tobacco (Lennan et al., 1999; Sabol et al.,
1999). In addition, there is speculation that family resemblance in the manifestation of
alcoholism is, in part, accounted for by a genetic liability of a general state of CNS
disinhibition/hyperexcitability which can also be found in high risk individuals for
alcoholism, substance abuse, antisocial personality, and attention deficit hyperactive

disorder (Begleiter & Projesz, 1999).

 

l SLC6A3 is the dopamine transporter gene that regulates synaptic dopamine by coding for a

reuptakc protein. The SLC6A3-9 genotype is a variant of the SLC6A3 gene.

11

k

. .
.- ‘
.2?» '

-b-‘...

..~

. -A‘ “‘

D'

7“ ‘
.5,..L..

V

a”? ""4-

“.“g -o

,4.—

IA

“Q

divsx
5.

1“" J
“A.“‘1

'

it“

C-
05

PL
K.

A positive family history of alcoholism also reflects environmental as well as
genetic risks that canalize developmental pathways of children toward maladaptive
behavior, including smoking and drinking (Fitzgerald et al., 1993; Zucker et al., 1996).
Paternal antisocial alcoholism, in particular, is associated with a number of risk factors,
including low family socioeconomic status (Fitzgerald & Zucker, 1995), relationship
disturbances between spouses (Ichiyama, Zucker, Fitzgerald, & Bingham, 1996),
neuroticism (Piejak, Twitchell, Loukas, Fitzgerald, & Zucker, 1996), and depression and
family violence (Ellis, Zucker, & Fitzgerald, 1997). In relation to smoking of adolescent
children, parents who habitually smoke and drink may hold more favorable attitudes and
beliefs toward smoking and drinking, which indirectly influence their children’s drinking
and smoking. Furthermore, alcoholism of parents may interfere with their parental roles
as a major supervisor of their children’s activities.

Thus, it appears that a family history of alcoholism can be considered as a general
risk factor for developmental outcomes in children as well as a more specific risk factor
for substance use, including early onset of smoking. However, a positive family history
of alcoholism has many dimensions. Alcoholism subtype is one aspect that is
underexplored. Although the heterogeneous nature of alcoholism has often been
hypothesized and studied in the literature (Babor et al., 1992; Cloninger, 1987; Zucker,
1987; Zucker, Chermack, & Curran, 2000; Zucker, Ellis, Fitzgerald, Bingham, &
Sanford, 1996; Zucker, Fitzgerald, & Moses, 1995), alcoholism subtypes have never been
studied before in relation to early smoking onset among adolescent children of alcoholics.

Furthermore, a positive family history of alcoholism has many variables, including the

12

 

number of alcoholic parents in family and the specificity of alcoholic parent in relation to
developmental outcomes in children.
Prenzgrl Exposure to Mateml Cigarette Smoking

Risk factor for cigarette smoking in offspring. It is well known that prenatal
exposure to maternal cigarette smoking influences the developing fetus by altering
maternal physiology that limits the amount of oxygen and nutrients to the fetus
(Weitzrnan, Gortrnaker, & Sobol, 1992). A new wave of studies suggests that maternal
smoking during pregnancy may directly be related to smoking initiation by the adolescent
offspring (Cornelius et al., 2000; Griesler et al., 1998; Kandel et al., 1994). A study by
Kandel et al. (1994) provides the first glimpse of a link between prenatal exposure to
smoking and smoking of offspring. A subset of 192 mothers and their first-bom children
aged 9-17 years was drawn from a larger representative sample of adolescents in grades
10 and 11 in New York State public high schools in 1971/72. A significant association
between maternal smoking during pregnancy and the child’s smoking 13 years later was
found with stronger associations for the child’s smoking during the last year than for
ever-smoking, and for daughters than for sons.

Kandel et a1. speculated that maternal smoking during a critical prenatal period of
brain development may modify the dopaminergic system structurally and functionally,
predisposing the child to smoke and to persist in smoking later in life. The speculation of
altered dopaminergic system in the brain due to prenatal exposure to nicotine has also
been echoed by many researchers as a possible mechanism for the association of maternal
smoking with behavior problems (Weitzrnan et al., 1992), conduct disorder (Wakschlag

etal., 1997; Weissman, Warner, Wickramaratne, & Kandel, 1999), attention deficit

l3

hyperactivity disorder (ADHD) (Milberger, Biederrnan, & Faraone, 1997; Millberger,
Biedennan, Faraone, Chen, & Jones, 1996, 1997), and attention problems and impulsive
behaviors (Fried, Watkinson, & Gray, 1992).

Taken together, evidence so far in the literature suggests that prenatal exposure to
nicotine is one of early risk factors for later smoking behavior among adolescents via
alterations in the structure and function of the brain. However, it is also suggested in the
literature that there may be mediating pathways to smoking behavior in adolescence
(Chassin et al., 1998; Griesler et al., 1998). Mediating pathways may include many
psychosocial and ecological dimensions such as positive attitudes toward smoking in
family, availability of cigarettes, and a lower level of awareness of the negative health
consequences of smoking. It is also plausible that prenatal exposure to nicotine may also
play a role in smoking onset among adolescents via behavioral characteristics, including
attention deficit problems and conduct problems in childhood, known antecedents to a
cluster of problem behaviors including smoking in adolescence.

Indirect link to early smoking onset via behavioral Mtefistics. Just as
prenatal exposure to cigarette smoking presents risks for adolescents’ smoking initiation,
so too do behavioral problems in childhood and adolescence. A study by Weitzrnan et a1.
(1992) was the one of the early studies that observed higher behavior problems among
children who were exposed to smoking prenatally. The observed association remained
significant even when variables known for their association with behavior problems were
controlled for, including birth weight, prenatal alcohol consumption, family income, and
parental education and intelligence. Similar relationships were reported in children as

early as three years of age (Day, Richardson, Goldschmidt, & Cornelius, 2000;

14

Io‘nsfz'ld

1113:3111 Cfl

‘ I
,. .- .
3.1.0.0115
.

i3
:3 .
1
'1)
C)‘
C.

(In

In

egarette

[II

t 3...,3 ’
kadllOId
among ad
mhefia

this theor

Townsend, 1998) and adults as late as 34 years old in terms of arrests for nonviolent and
violent crimes in a dose-response relationship (Brennan, Grekin, & Mednick, 1999). In
addition, the associations between prenatal smoking exposure and a host of behavioral
problems including delinquency (Bagley, 1992) were reported not only from maternal
ratings but also independent observer ratings (Fergusson, Horwood, & Lynskey, 1993).

From the findings, we can speculate that the risks of prenatal exposure to maternal
cigarette smoking for early smoking onset in children may indirectly be transmitted via
behavioral characteristics. The association between behavior problems and substance use
among adolescents, including smoking and drinking, has long been modeled and reported
in the frame of problem behgvior theory (R. Jessor & S. L. J essor, 1977). According to
this theory, there is an underlying tendency toward deviancy that is manifested in various
forms: smoking, drinking, early sexual behavior, poor school performance, and
association with deviant peers in adolescence. While these behaviors may be found more
often together than in isolation, recent studies suggest that there may be an alternative
explanation that addresses the more specific nature of the associations of cigarette
smoking with attention, behavior, and affect, and cigarette smoking.

Alternatively, cigarette smoking is suggested to influence selective populations
with negative affect and stress, poor attention, and problems with inhibition, possibly due
to stimulant effects of nicotine (Downey, Pomerleau, & Pomerleau, 1996; Lambert &
Hartsough, 1998; Levin et al., 1996; Milberger et al., 1997; Patton et al., 1998; O. F.
Pomerleau, Downey, Stelson, & C. S. Pomerleau, 1995; Riggs, Mikulich, Whitrnore, &
Crowley, 1999; Tizabi et al., 1999). This self-medication hypothesis is supported in

studies of animals (Tizabi et al., 1999) and adolescents. Smoking initiation is observed

15

more ofien in at
W“ .\‘l. \l'intil
It'll; hltlberge
undue: disoro':
P196 lliiberge

Taken ti
5811511313 or \‘t

I”;
’-
.1

(IE:
:3
._‘.
r4
.‘.
La
r—o

suites Partic
population suc
problems leg-
measured conc
tithe majority
shelter sympt

if behavioral

In sum

more often in adolescents with depression and anxiety (Patton et al., 1998; Riggs et al.,
1999; M. Windle & R. C. Windle, 2001) and with ADHD (Burke, Loeber, & Lahey,
2001; Milberger et al., 1997; Whalen, Jarnner, Henker, Delfino, & Lozano, 2002),
conduct disorder, and behavior problems (Lambert & Hartsough, 1998; Levin et al.,
1996; Milberger et al., 1997).

Taken together, the existing studies appear to converge on an individual
sensitivity or vulnerability to cigarette smoking. However, it is still not clear which
behavioral characteristics are vulnerability factors that can early be identified for
smoking initiation among adolescents. It is due in part to selective populations used in
studies. Participants in many studies previously mentioned were sampled from a special
population such as those who were in treatment for symptoms of ADHD and behavioral
problems (e.g., Milberger et al., 1997). In addition, behavioral characteristics were often
measured concurrently for adolescent populations or retrospectively for adult populations
in the majority of existing studies. For these reasons, for example, it is still not clear
whether symptoms of ADHD are uniquely associated with smoking above and beyond

the behavioral problems and vice versa (Lynskey & Hall, 2001).

In summary, the literature on adolescent smoking initiation suggest that an
individual vulnerability to cigarette smoking may be traced back to their prenatal
development. Negative affect, inattention, and unruly, and unrestrained behaviors may
be potential early behavioral characteristics that mediate the link between prenatal
exposure to maternal smoking and early onset of smoking in adolescent children. We

then ask naturally whether parental smoking and alcoholism are associated with early

16

unset ot'smokin
hsignititam dir
utergeneration:
pattuays ria pr
characteristics.
alcoholism to ea
neonatisms of;
present study.

This stud
Emission of i
prOSpectiye. IOY
adolesetms an

remission
hthilc‘ren b
model illust
smoking. at
onset and Il
More
transmission a
PaiWWO/Y'flg

“W .r
alt/473,1] parent‘-

onset of smoking above and beyond the more specific mediational pathways.
Insignificant direct paths in the presence of the mediating pathways would suggest that
intergenerational transmission of smoking is largely accounted for by the mediating
pathways via prenatal exposure to daily maternal cigarette smoking and early behavioral
characteristics. On the contrary, significant direct paths from parental smoking and
alcoholism to early onset of smoking would strongly suggest that there are unaccounted
mechanisms of parental smoking and alcoholism in leading to early smoking onset by the

present study.

The Current Study

This study seeks to advance the research literature of intergenerational
transmission of smoking and early smoking onset in a number of aspects using the
prospective, long-term longitudinal study of a population-based, high-risk sample of early
adolescents and their parents. Figure 1 illustrates a heuristic model of intergenerational
transmission of smoking where a familial risk structure is linked to early smoking onset
in children by mediators of individual vulnerability factors. Following the heuristic
model illustrated in Figure 1, the current study plans to separately examine parental
smoking, alcoholism, and early characteristics of child as risk factors for early smoking
onset, and then to investigate their roles played in early smoking onset simultaneously.

More specifically and first, it was hypothesized that intergenerational
transmission of smoking can be observed reliably. The following three factors related to
parental smoking were explored in association with early smoking onset in offspring of

smokers: 1) Parental smoking subtypes derived from a long-term prospective follow-up

17

eadyot‘ds'i

Second. P3“
As in parent
explored in «'
£57330 fror:
prints. and
for early snt

Tllllt
suing no
Internal cig
by age l4. I
an. reliably
characteristi

Fina

S’u'lllllmeou

Hauling sul
l.

ml": bthan
'1‘.

-l.,,r .
“‘tfiftllsm

{Pal-‘13 1 \ 3

study of cigarette smoking, 2) the number of smoking parents, and 3) the relative potency
of maternal versus paternal smoking as a risk factor for early smoking onset in offspring.
Second, parental alcoholism was expected to predict early smoking onset in offspring.

As in parental smoking, the same three factors related to parental alcoholism were
explored in association with early smoking onset: 1) Parental alcoholism subtypes
derived from a long-term prospective study of alcoholism, 2) the number of alcoholic
parents, and 3) the relative potency of maternal versus paternal alcoholism as a risk factor
for early smoking onset in offspring.

Third, it was hypothesized that prenatal exposure to daily maternal cigarette
smoking would be related to early smoking onset in offspring. Heavier exposure to
maternal cigarette smoking was hypothesized to trace to adolescents who have smoked
by age 14. In addition, it was hypothesized that adolescents who start smoking by age 14
can reliably be differentiated on measures of early temperamental and behavioral
characteristics as well as concurrent behavioral characteristics.

Finally, direct and indirect paths to early smoking onset were hypothesized and
simultaneously tested. Maternal and paternal smoking subtypes, parental alcoholism, and
prenatal exposure to maternal smoking were hypothesized to directly lead to early
smoking onset in children. In addition to direct paths, a mediational path from maternal
smoking subtype to early smoking onset was hypothesized via prenatal exposure and
early behavioral characteristics. Figure 2 illustrates a simplified model of pathways to
early smoking onset. Paths 1 — 3 indicate direct associations, with parental smoking and
alcoholism, and prenatal exposure to smoking predicting early smoking onset in offspring

(Paths 1 —- 3). The first path stands for intergenerational transmission of smoking, while

18

the second path r
melting. Path.3

ntemal Clearer:
V

smoking (Path 4

melting onset t

the second path represents a familial alcoholism as a risk factor for early onset of
smoking. Path 3 reflects the direct association between prenatal exposure to daily
maternal cigarette smoking and early smoking onset in children. And finally, Path 4
indicates indirect mediational pathway from parental smoking to prenatal exposure to
smoking (Path 4-1), to early behavioral characteristics (Path 4-2), and then to early

smoking onset (Path 4-3).

19

    

$ch
weEoEm beam

o.8.ouo..~.=O.. Iomlwoo. mlm

    

BEBE memo—08m Mo 56:50 genome—owe 5 “once wagon; base .3 3608 czar—do: <

 

VH u N" momm as momumtouowhdno

322269 805280

m - m wows as
moumtflumano #82232— beam

“583383 EEO

mutate SE5 wee—08m
3 2:8on 33:05

 

mEmgnooz $5382

 

 

._ 05mm

         
              

  

8&20
flamenozmm
ESSA

sac

BEES?

REP—am

 

wee—08m 3:85“

08:82 big .3 x03

lease ma 3:35

20

m4... 1:. :1" 93.9-31.4...J NJ

.09.?!- .IJ..LJ:.::_.J _....:v.->.:_.Jn.- rah-aluvnfixuv —::~:r.v-a~ ~m..7..-xav-nv.J-w.i~.-4:mu:v-~:nfi. auhfila :~l~W.. Huey-Twine N
. u . a

.825 wee—08m 3.80 8 @3353 no $on comm—95m <

              

 

 

 

    

    

     

 

.N Rama
“ammo
wagoem _ 5am
ram
l N 53
m 53
me 53
8338335 a w
38523 was 5mm 0.38 xo szonoofi ego—em
beam 335$ 3553 13:23
TV 53
mac—08m otoSIMB 9
moumtofilaalno _l_a.8. lg“: lam 0.58 x0 335.5 flags—Jam mg

21

 

Participant
and their sons (5'
four sat es of the
longitudinal Stu-I
1985 until mitt}
my designed to
lttiing lost ard. C

Pithtipathology

alcoholic and no

laniiies with of
diagnosis of FA
im‘ol

t'sment, an

literati e1 ,1

Almost ;
P‘PWS ( 54:

Pitcan.

ENC

 

METHOD

Participants

Participants for the current study were 281 families with both biological parents
and their sons (N = 281) and daughters (N = 88), who completed at least two of the first
four waves of the larger University of Michigan — Michigan State University (UM-MSU)
Longitudinal Study (Zucker et al., 2000) during the seventeen-year-span starting from
1985 until mid-2001. The UM-MSU Longitudinal Study is an ongoing longitudinal
study designed to understand the risk and protective factors that affect etiologic pathways
leading toward, or away from, alcohol abuse or dependence and co-active forms of
psychopathology from a high risk sample of families in the mid-Michigan area with
alcoholic and non-alcoholic fathers (Zucker et al., 2000). Maternal alcoholism was also
assessed, but it was neither a requirement for study inclusion, nor a basis for exclusion.
Families with children who manifested characteristics in the three areas required for a
diagnosis of F AS (e.g., prenatal or postnatal growth retardation, central nervous system
involvement, and characteristic facial dysmorphology; Cooper, 1987) were excluded

(Fitzgerald et al., 1993; N011, Zucker, Fitzgerald, & Curtis, 1992).

Almost all parents included in the present study were non-Hispanic European
Americans (542 of 555; 97.7%) with the exception of eight Hispanic Americans, three
Native Americans, one Asian American, and one of other ethnic heritage other than
Afiican American. All resided in the mid-Michigan area at initial assessment. Family
income levels mostly fell within the lower to low-middle class range, although there were
also some higher income families. Trained interviewers, who were blind to the family

diagnostic status, collected the data. Since many alcoholics are known to be smokers in

22

I Y I“I II
area. are pops: .
. . - |
3351?.1'115111111'3 I
a I .I .
afifflpmtfllm p.11.
30‘. t time. 311D
nt- \_1 w 0‘53" q' ‘ 3 .
aging .li'nll auOlt:

111111: 111':

«1'. :: ;
5‘“: Oi limli‘l he.

ili'lalf 1). Far

:31:fo Pilate .1)
.1. ' ~ .
.ea 1anth and

.

03 participating ta
titration on th
flit-there lZuclte
median year of as

“Shem for al

401": demlvd inf

general and population-based community alcoholics don’t seek out treatment, the
participants in the current study provide an opportunity to investigate the natural
developmental patterns of cigarette smoking and alcoholism among parents over a long
period of time, and the mechanisms of how they are related to early smoking initiation

among their adolescent and preadolescent children.

During the initial contact, all families were invited to participate in a long-term
study of family health and child development starting at the male child ages of three to
five (wave 1). Families then have been followed up once in every three years when the
male child’s age reached six to eight (wave 2), nine to eleven (wave 3), and twelve to
fourteen (wave 4). Parental information on the areas of psychosocial fimctions, including
their drinking and smoking has also been collected once in every three years. Daughters
of participating families were recruited into the study a few years later. More
information on the recruitment procedures and sample characteristics are available
elsewhere (Zucker et al., 1996; Zucker & Fitzgerald, 1991; Zucker et al., 2000). The
‘ median year of assessment for each of four waves and the corresponding age at
assessment for all participants included in the current study are presented in Table 1 (for

more detailed information on participants, see also Appendices A and B).

23

"Jl
(I;

“as I
Was 3
Was .‘
Media:
Ya.-

Wat's

Ware

Table 1

Descriptive Statistics of All Participants

 

 

 

 

 

Father2 Mother2 Son Daughter
E=275 E=280 fl=281 fl=88
Age
Wave 1 33.15 (5.11) 30.93 (4.23) 4.31 (1.00) 4.86 (0.90)
Wave 2 36.58 (4.97) 34.47 (3.91) 7.58 (1.00) 7.56 (0.86)
Wave 3 39.54 (5.00) 37.25 (4.09) 10.40 (0.94) 10.26 (0.87)
Wave 4 42.58 (4.95) 39.99 (4.29) 13.44 (0.93) 13.29 (0.80)
Year of birth
Median 1956 1958 1985 1986
Range 1938 - 1966 1943 - 1970 1979 - 1988 1981 - 1992
Yea_r of assessment
Wave 1 1989 1989 1989 1993
Wave 2 1993 1993 1993 1994
Wave 3 1996 1996 1996 1996
Wave 4 1998 1998 1998 1998

 

Note. The numbers in parentheses are standard deviations. Year of assessment is shown

in median (For more detailed information, see also Appendices A and B.).

 

2 The number of fathers and mothers included in the present study were different from the number

of sons because six fathers and one mother did not complete two of the first four assessments, failing to

meet the criteria of the study inclusion. The cases were not deleted listwise since not all subsequent

analyses required all inforrmtion.

24

Patel Measure
_._—.———

Parental s
51:12:: item quest

52337" RCSpong,‘

About one and 0
treble reflects
axial} used and
We: imposed 2
eille. Ial Disor
criteria Inform;
and Drug Histo
lichigan Alcol
he \NH Diag
1339}- The DH
consummation 0:

me.

The D1:

‘ I 161' “lens“.

as wel} as (“he
Ibf 1978 Nallo

1978}. the Am:

Parental Measures
Parental smoking at waves 1 through 4. Parental smoking was determined by a

single item question, “How frequently have you smoked cigarettes during the past 30
days?” Responses were 0 = Not at all; 1 = Less than one cigarette per day; 2 = One to
five cigarettes per day; 3 = About one-half pack per day; 4 = About one pack per day; 5 =
About one and one-half packs per day; 6 = Two packs or more per day. This seven-level
variable reflects quantity of smoking at the time of measurement. This question is very
widely used and accepted in the literature of smoking.

Parental alcohol use disorder at waves 1 through 4. Parental alcohol use disorder
was diagnosed at each assessment wave based on the Diagnostic and Statistical Manual
of Mental Disorders, Fourth Edition (DSM-IV; American Psychiatric Association, 1994)
criteria. Information for the diagnosis was obtained via administration of the Drinking
and Drug History Questionnaire (DDHQ; Zucker, Fitzgerald, & Noll, 1990), the Short
Michigan Alcohol Screening Test (SMAST; Selzer, Vinokur, & van Rooijien, 1975), and
the NIMH Diagnostic Interview Schedule (DIS; Robins, Helzer, Croughan, & Ratcliffe,
1980). The DIS and the DDHQ provided detailed information on past and current
consumption of alcohol and smoking patterns and problems related to excessive alcohol
use.

The DIS is a structured diagnostic interview that allows trained lay interviewers to
gather extensive information about physical, alcohol-related, and drug-related symptoms,
as well as other areas of psychiatric syrnptomatology. The DDHQ consists of items from
the 1978 National Institute on Drug Abuse Survey (Johnston, Bachman, & O’Malley,

1978), the American Drinking Practice Survey (Cahalan, Cisin, & Crossley, 1969), and

25

he Veterans Adn'
tSchzcl't't. 19'31.
tug use and path
set mouths and it
Iéhour period at
piticipant uheth
:fso. ’he freq uettt
1:» entory used to
muses.

L’strtg all
alcohol abuse dc]
itfeume. the pagt
milluses were t
alcohol dependt~
det’nce, In .
m1We time p

Re t

milled age Ofp
988nm] data on

Memes of

mutt-0W d
the 3~\S€SSmem V

the Veterans Administration Medical Center Research Questionnaire for Alcohol
(Schuckit, 197 8). The DDHQ gathers information about the informant’s alcohol/other
drug use and problems with regard to the amount of alcohol consumption during the past
six months and it also inquires about the largest amount of alcohol consumed during a
24—hour period at any point in the participant’s life. The instrument also asks the
participant whether s/he has experienced various problems as a result of alcohol use and,
if so, the frequency of these problems. The SMAST is a well-validated screening
inventory used to assess alcohol problems, consisting of 13 items with “Yes/ No”
responses.

Using all three sources of information, an experienced clinician established
alcohol abuse/dependence diagnoses for the following three periods: the subj ect's entire
lifetime, the past three years, and the current year leading up to the assessment.
Diagnoses were coded from zero to three: 0 = No diagnosis; 1 = Alcohol abuse; 2 =
Alcohol dependence without physical dependence; 3 = Alcohol dependence with physical
dependence. In the current study, alcohol use disorder diagnoses were made over four

exclusive time periods and the time prior to the first assessment.

Rearrangement of da_t-a on parental smoking and alcoholism. The current study
utilized age of parents as a time variable instead of assessment wave when analyzing
parental data on smoking and alcoholism. The interest of the current study with regard to
the measures of parental smoking and alcoholism was to see whether the change and/or
stability of the developmental patterns could be found over chronological age, not over
the assessment wave at which variables related to smoking and alcoholism were

measured. However, parental data were collected based on their son’s assessment

26

media
and but:
use or
353671. 1

.
nanny
r" “Au.

391?":

“-uAM

shroud:
ii". In

3.178711

schedule (i.e., age), resulting in a rather heterogeneous sample of parents in terms of ages
and birth cohorts. This scheme was originally designed to match parental functions to

those of their son.

The research questions in the current study, however, focus more on identifying
several clusters of parents based on their own functions over time. For that reason,
parental data on smoking and alcoholism were rearranged so that age rather than
measurement wave was the time variable for analysis. Avoiding aggregation of the data
across different ages for each measurement by grouping individuals by age, allowed one
to identify change and stability of smoking and alcoholism in adulthood as a function of
chronological age. In addition, we can gain knowledge about a longer period of the life
span in a shorter amount of time to maximize efficiency of the available data. In the
current study, parental smoking was investigated for the time period from ages 24 to 50.
Parental alcoholism was investigated for the time period from ages 14 to 54 for men, and

ages 14 to 49 for women.

There is one shortcoming of the design that needs to be noted here. The
rearrangement of the data results in a higher rate of missing values because the data were
converted from four measurements in nine to twelve years to twelve plus measurements
in as many years (The number of observed cases used in the TRAJ procedure is reported
in Appendix D.). The key to the issue was to find the balance between maximum
utilization of the available longitudinal data and integrity of the data. Given that both
smoking and alcoholism are phenomena of high stability and convergence and that the
analytical tool (TRAJ) utilized in the current study handles missing data well, it was

decided that the reliable findings were attainable in the current study.

27

Parental antisocial behavior at wave 1. Parental antisocial behavior was measured
by the Antisocial Behavior Checklist (ASBCL; Zucker, Noll, Ham, Fitzgerald, &
Sullivan, 1994) when parents were first recruited into the project. The ASBCL is a 45-
item questionnaire that assesses the frequency of aggressive and antisocial activity in
both childhood (e. g., lying to parents, being suspended from school) and adulthood (e. g.,
being fired, resisting arrest). Chronbach’s alphas were .832 and .834, respectively for the
subscales of childhood and adulthood in the current study.

Parental depression at wave 1. Parental depression was measured by the revised
Beck Depression Inventory (BDI; Beck, Rush, Shaw, & Emery, 1979) and the Hamilton
Rating Scale for Depression (HRSD; Hamilton, 1960). The BDI was designed for
assessment of the severity of depression in adolescents and adults (Beck & Steer, 1993).
The BDI has been one of the most frequently used instruments for assessing the intensity
of depression in psychiatric patients and for screening possible depression in normal
populations since its first introduction in 1961 (Piotrowski, Sherry & Keller, 1985; Steer,
Beck, & Garrison, 1986). The items represent symptoms and attitudes based on clinical
observations and descriptions of symptoms frequently mentioned by depressed patients,
and are rated on a 4-point severity scale ranging fi'om zero to three. In the current study,
the short version of BDI with 13 items was used (Chronbach’s alpha = .751).

The HRSD was originally designed for use with patients already diagnosed as
suffering from depression measuring behavioral and somatic symptoms of depression.
The HRSD has been utilized most frequently as an interviewer-based measure for patient
selection and later assessment (Grundy, Lunnen, Lambert, Ashton, & Tovey, 1994). The

HRSD was coded following administration of the DIS by the clinician who conducted the

28

, 93’1“."
.251 ' lb “

"‘3<.'"."

luck-U1» ‘-
I

1‘”: tr

nu. h:
5
'LJ' 1 '91
:0 A»!

‘ 1
“5.: r".
in -

'4
ch.

l
":03. '3'
i'l‘a 1‘.

was

.4. s

‘ t
C’Vl“
“A.“ ’

interview. The score was based on both the participant's responses to interviewer’s
questions during the DIS administration as well as the interviewer's judgment upon
completion of the DIS administration. The interviewer made both a Current Depression
rating and a rating of the level of the subject's depression at the point in their life when
they were most depressed (Worst Ever rating). The Worst Ever episode was selected on
the basis of the period with the largest number of depressive symptoms reported. Inter-
rater reliabilities obtained on this project were .78 for current depression and .80 for
worst-ever depression using a sample of 16 individuals (Reider, 1991).

Egrental education and occupation at wave 1. The current study included parental
education measured by years of education completed and parental occupation coded
using the US. Census occupation codes. Parental occupation was recoded based on the
revised Duncan Socioeconomic Index (RDSEI-TSEIZ; Stevens & F eatherrnan, 1981).
Higher scores on parental occupation indicate socially more prestigious occupations.
Child Measures

PrefingLal exposure to maternal smoking and drinking. The Health History -
Prenatal Form (HHPF; Carpenter & Lester, 1980) was used to obtain information on
mothers’ cigarette smoking and alcohol use during pregnancy. Mothers recalled how
many cigarettes they smoked per day and how many drinks they had per week during
pregnancy. The mean length of time between the child’s birth and the mother’s report on
their smoking during pregnancy was 4.18 years post-delivery for boys and 9.05 years
post-delivery for girls. Compared to a prospective measure of prenatal exposure to
smoking and drinking, the retrospective measure used in the current study was limited.

However, the time lag between the child’s birth and the mother’s recall was relatively

29

short in the current study and the retrospective report of maternal cigarette smoking
during pregnancy appeared valid (Griesler et al., 1998).

Child temperament at wave 1. Child temperament was measured by the
Dimensions of Temperament Survey (DOTS; Lerner, Palermo, Spiro III, & Nesselroade,
1982) rated by each parent at Wave 1. The DOTS is a 34-item questionnaire that
measures five dimensions of temperament: Activity Level, Attention Span/Distractibility,
Adaptability/Approach-Withdrawal, Rhythmicity, and Reactivity. The dimension scores
are based on sums of the items for that dimension (1 = True; 0 = False). Three of the
temperament dimensions were used in the present study as precursors of smoking
behavior in adolescence: Attention Span/Distractibility, Adaptability/Approach-
Withdrawal, and Reactivity. High scores on the dimensions reflect greater longer
attention span and higher persistence to distraction, higher approach, and greater
reactivity. Chronbach’s alphas were .803, .765, and .519, respectively. Activity Level
and Rhythmicity were not included for reasons of the very limited range of possible
scores (0-3) and irrelevance to child outcomes in earlier work (Tarter & Vanyukov,
1994), respectively.

Behavioral characteristics3 : Anxious/depressed. Attention problems. Delinquent
behavior. Aggressive behavior at waves 1 and 4. Based on the existing literature, four of
eight syndromes from the Child Behavior Checklist for Ages 4-18 (CBCL; Achenbach,

1991) were tested in the current study in relation to smoking onset in adolescence:

 

3 A term, behavioral characteristics was used in place of behavioral problems or syndromes in the
present study since only a few children scored over the borderline clinical cutoff score on each of four

syndrome scales.

30

humus
swim

.fl‘
A3 M
REES.)

D... .
r. :,
went».

v

that: ]
i”nnt~
' “Saul

35611

mu

vtere sl

({3}
f}!
._.
. -

mater;

filtttm

‘Hat‘e

Anxious/depressed, Attention problems, Delinquent behavior, Aggressive behavior.
Syndrome scale scores were computed by summing individual items. The CBCL
consists of 118 items measuring the prevalence and degree of child behavior problems.
Items were rated on a three-point scale (0 = Not True; 1 = Somewhat or Sometimes True;
2 = Often or Very True). The CBCL is well known for its robust construct and
discriminant validity, as well as its reliability (Achenbach, 1991). In the current study,
ratings from each parent were obtained between the time when the child was three to five
(wave 1), and twelve to fourteen (wave 4). Cronbach’s alphas for Anxious/depressed,
Attention problems, Delinquent behavior, Aggressive behavior were .725, .690, .560, and
.866 at wave 1 and .801, .785, .750, and .888 at wave 4, respectively.

Smoking onset. Smoking onset was determined by responses on two questions
from the Drinking and Drug History Questionnaire (DDHQ; Zucker et al., 1990) that
were slightly adapted for youth. The questions were asked during their regular wave 4 (at
ages 12, 13, or 14) assessment and also annually at ages 11, 12, 13, and 14 for the
majority of adolescents. The questions asked whether adolescents smoked during
lifetime and the past 12-month period (i.e., “Have you ever smoked cigarettes?” and
“Have you smoked cigarettes during the past 12 months?”). Their response was recoded
as “Never,” “Once or twice,” “Occasionally but not regularly,” “Regularly for a while
during this year, but not now,” and “Regularly now.”

Teenagers were grouped into three categories based on the following criteria.
First, adolescents who acknowledged their smoking (at least “Once or twice” or more) in
any time during ages between 11 and 14 were considered as having experimented with

cigarette smoking by age 14 (Smoking-onset; p = 79; 28.1% for boys; p = 21; 23.9% for

31

7'25). S

U

as

tu=llJ

1‘ ‘n ‘
refit).

those at
hest u
Lteiuie
to age 1
create a
suite t
he “Sn
assess:

Vin:
. “5‘1”
X

girls). Second, adolescents who never smoked by age 14 were grouped as “Non-smoker”
(p = 114; 40.6% for boys; 3 = 20; 22.7% for girls). For some adolescents of early birth
cohorts, data from later assessments (e. g., wave 5) were available. If they continued to
remain “Non-smoker” at age 15 or older, they were then grouped as “Non-smoker.” The
last category, “Smoking-Onset Unknown” (or smoking-onset remains to be seen) was for
those adolescents who have not smoked but younger than 14 years old at the time of the
latest assessment (3 = 88; 31.3% for boys; p = 47; 53.4% for girls). This category also
included a small number of adolescents who never smoked by the last measurement prior
to age 14 but whose annual information at age 14 was not available. The decision to
create a “Smoking Onset Unknown” category rather than to treat it as missing was made
since the category conveys meaningful implications. The higher percentage of girls in
the “Smoking Onset Unknown” category reflects their younger birth cohorts so that later
assessments have not been completed.
Missing Data Estimation

Although there is no rule of thumb for an acceptable rate of rrrissing data, the
current study included any participants who had completed at least two out of four
measurements. Missing data were handled at two levels. At the first level, the data on
parental smoking and alcoholism were used with missing data in the subsequent analyses
of trajectories of parental smoking and alcoholism. The SAS macro, TRAJ, developed by
Jones, Nagin, and Roeder (2001) uses all the available data while neither imputing nor
deleting the missing data. Once the trajectories of parental smoking and alcoholism were
identified, the remaining missing data were imputed separately for parents and adolescent

children using Schafer’s NORM version 2.03 (Schafer, 2000), the most accessible

32

ripen

thisrfr
tom
3100‘
in Wh
Nth

Simul

Patter

idols

implementation program of multiple imputation (MI) method (Graham & Hofer, 2000;
Schafer, 1997, 1999).

M1 is the technique that substitutes each missing value m times with a value
representing a distribution of possibilities in m datasets (Rubin, 1987). M1 is superior to
other popular procedures of single imputation because it provides a valid basis for
computing standard errors of the parameter estimates by treating missing data as an
explicit source of random variability (Graham & Hofer, 2000; Rubin, 1987; Schafer,
2001). Utilizing M1 for missing data imputation in an analysis usually involves three
steps: multiple imputation of m datasets, repeated analyses over _m_ datasets, and
combining results from gn_ datasets to obtain a single set of results.

The NORM program used in the current study is originally designed for normally
distributed data. However, it is shown to yield good results even with seriously non—
normal data (Graham & Schafer, in press; of. Graham & Hofer, 2000). Under the NORM
procedure, data augmentation procedure alternates between the imputation step (I-step),
in which missing data are simulated from their conditional distribution given the current
estimate of the covariance matrix, and the posterior step (P-step), in which new values are
simulated by drawing them from a Bayesian posterior distribution given the current
values of the data (Graham & Hofer, 2000; Schafer, 1997, 1999).

In the current study, parental data and adolescent data were imputed separately.
The number and percentage of cases with missing values and the matrix of missingrress
patterns for all variables used in the current study are presented separately for parents and
adolescents in Appendix C. Missing data on parental measures were very few (for more

information see Table C1 in Appendix C). Parental data needed only 38 iterations for

33

Expectation Maximization (EM) convergence. The diagnostics on the mean, and the
variance and covariance parameters showed that the autocorrelation reached near zero
instantly. Following the rule of thumb (i.e., doubling the number of iterations for EM
convergence) and the diagnostics, 760 total steps were specified in data augmentation
step. Although no more than five multiply imputed datasets are sufficient for most
occasions (Rubin, 1987; Schaffer, 1997), conservatively ten datasets were imputed per
every 76 steps of iterations for parents in the present study.

The adolescent children’s data included parental reports of behavior problems
assessed when the target child was between the ages of three and five (wave 1) and again
between the ages of 12 and 14 (wave 4). However, a substantial portion of adolescent
daughters in the current study had missed data collections at waves 1 and/or 4 (see Table
C3 in Appendix C for more information) due to the recruitment design of the UM-MSU
Longitudinal Project. Either they were not in the age range to complete wave 4 or they
were recruited to the study from wave 2 and on. Therefore, child behavior problems
reported by both parents at waves 2 and 3 were also included in the missing data
estimation procedure to assure a better solution, although they were not part of the
analyses in the current study. This is a generally recommended procedure for imputing
missing data because the imputed values do not depend on what is included in the
subsequent analyses (Little, 2001; Schafer, 2000).

Adolescent children’s data required 1,243 iterations for EM convergence, due in
part to the large number of variables included and to a higher proportion of missing
information especially for girls. Following the same procedure used for the parental data,

2,600 total steps were selected with an imputed data set per every 260 steps. Tables C2

34

and C3 in Appendix C present matrices of missingness pattern of the data used in the
subsequent analyses for sons and daughters, respectively. Results from daughters were
mostly considered exploratory in nature, due to a small sample size (N = 88) and a higher
rate of missing data (see Table C3 in Appendix C), with the exception of prenatal
exposure to maternal smoking. Results from daughters are briefly noted in the following
sections when applicable and tables and figures pertaining to daughters are presented in

Appendix G.

35

RESULTS

Results are organized as follows: 1) Intergenerational Transmission of Smoking
asks whether and how parental smoking plays a role of a risk factor for children’s
smoking, 2) A Family History of Alcoholism investigates alcoholism subtypes in relation
to early onset of smoking, 3) Early and Concurrent Characteristics of Adolescents with
Early Smoking Onset investigates prenatal exposure to maternal cigarette smoking, early
temperamental dimensions, and behavioral characteristics across different groups of
adolescents with different smoking onset status, and 4) Paths to Early Onset of Smoking
investigates direct and indirect paths to early onset of smoking in children. Each of the
four results addresses the mechanisms of early smoking onset from slightly unique
perspective and methods.

The first two sections of results address the notions of intergenerational
transmission of smoking and a familial alcoholism as a shared vulnerability factor for
early smoking onset. First, in order to identify subtypes of smoking and alcoholism
among parents in the first two sections of Results, a group-based semiparametric
modeling approach4 (Nagin, 1999; Nagin & Trembly, 1999, 2001) was utilized. Unlike
hierarchical and latent grth curve modeling analyses that focus on an average growth
curve under the assumption that all individuals belong to one homogeneous population,
this new approach was designed to identify qualitatively different, prototypical multiple
developmental growth patterns. Following identification of the number of groups,

posterior probabilities of group membership are estimated for all individuals. Multiple

 

4 It is also called Latent Class Growth Analysis (LCGA; B. Muthén, 2001).

36

grth patterns are conventionally dealt with by modeling a growth function separately
for a subset of the sample in hierarchical and latent grth curve modeling analyses,
based on a priori theories and hypotheses. However, this procedure is limited in the
sense that there is no statistical ground to test whether there exist multiple groups and
how many subgroups exist (Nagin, 1999).

In the current study, the SAS (SAS Inc., 1999) procedure, T RAJ (Jones et al.,
2001) was used to fit semiparametric mixtures5 of censored6 normal distributions of
parental smoking and alcoholism. Once the identification of group membership for all
parents was completed based on posterior probabilities, the current study investigated
whether these groups are different on psychosocial measures such as years of education,
occupation, conduct problems in childhood, antisocial behavior, and depression.
Multivariate Analysis of Variance (MAN OVA), followed by subsequent univariate
analysis of variance were conducted using S Plus (MathSoft Inc., 1998) and SPSS (SPSS
Inc., 1999) software programs.

Second, using the group membership identified in the previous step, Configural
Frequency Analysis (CF A; von Eye, 2001a) was conducted to see whether and how

parental smoking and alcoholism are related to early smoking onset in offspring. CF A is

 

5 A mixture refers to the situation where the measurements of a random variable are taken under
two or more different conditions, resulting in the distribution of the mixture of two or more subpopulations.
Instead of treating the distribution as bi- or multi-modal, in analysis of mixture models, a number of
subpopulations are identified and parameterized so that a relatively simple model can normally be applied
to each of subpopulations (Gelrnan, Carlin, Stern, & Rubin, 1995).

6 Censored data means that the upper and lower limits of the data have been truncated by various

reasons when the underlying distribution is normal.

37

a multivariate method for typological research that involves categorical variables in both
exploratory and confirmatory research. Using CFA, researchers ask whether cells
contain fewer or more cases than expected from some chance model (von Eye, 1990,
2002). CFA is the only method application of person-centered approaches that shows
whether a group of observations (or configuration) does not occur beyond statistical
significance (von Eye, personal communication, 2001b). Therefore, CFA application is
well-suited for the current research question.

Third, the current study investigated whether adolescents who started smoking by
age 14 can be identified based on measures of prenatal exposure to maternal daily
cigarette smoking and weekly drinking, child temperament, and child behavioral
characteristics. Concurrent behavioral characteristics measured at ages twelve to
fourteen were also investigated in relation to their smoking onset status. Multivariate
analysis of variance (MANOVA) was conducted using S Plus (MathSofl Inc., 1998) and
SPSS (SPSS Inc., 1999).

Fourth, using the selected key factors of intergenerational transmission of
smoking, manifest variable structural equation modeling (SEM) analysis was conducted
using LISREL program version 8.51 (J oreskog & Sorbom, 2001 ). Given that at least a
few key variables were either categorical (e.g., smoking onset status) or ordinal (smoking
subtypes), other numerical variables were recoded into ordinal variables while
maintaining as much information as possible. Due to the ordinal nature of the variables
and the moderate sample size in the current study, a limited number of variables were
selected to test direct and indirect paths of intergenerational transmission of smoking

Simultaneously.

38

U
u
L)

STICK

etpre

mp8:
ftpet

"31131

11her

1'6

N0 11

itdg)

INTERGENERATIONAL TRANSMISSION OF SMOKING

Developmental Patterns of Smoking in Adulthood
A mixture modeling was used to identify distinctive clusters of parental smoking
patterns using a customized SAS macro, TRAJ developed by Jones et al. (2001).

Parental smoking measured in the current study was censored normal. Smoking behavior
in adulthood was modeled by means of a latent variable y? , measuring potential for
smoking for individual i's age at time t given membership in group j. Following Nagin
(1999)’s notations, for example, a quadratic relationship between smoking and age is
expressed:

yf/ = 65 + Bf Age” + B! Age-‘3 + 6.7
where Age" and Age; are the subject i's age and the square of the age at time t,
respectively, and E is a residual term assumed to be normally distributed, with the
expected mean of zero and constant variance 0]. The expected value of the m

variable, yzj , is y? = 65 + 61’ Age" + [32" Agej. The expected value of the observed
variable, E ( Ynj ) , assuming group membership j observed, is expressed:

Em! ) = spasm... + M. (cw... — we... ) + awe... — 45:...) +(1 — we... )Sma.
where <I> and ¢ denote cumulative normal distribution function and the normal density
function, respectively, and Sm and Sm, denote the minimum and maximum possible
scores on the measurement scale, respectively (For more information, see Nagin, 1999).

The determination of the optimal model of parental smoking trajectories involved
two important issues: 1) determination of the number of groups to compose the mixture;

and 2) determination of the order of the growth patterns (e. g., linear, quadratic, cubic,

39

etc). The de‘
or the Bates
McCall. and
aloe is gene
addition to t1
numher ot~ g
estntates for
to the order 1
pmteter gi
unpotls tn
30M).

L'sir.
0fparental s
555 Parents
mltoes. (
313161th 5
0“ the BIC
as he best.
{hen assign
maximUm 1
resulted in
,\

The
156457)

etc.). The determination of the number of groups that best describes the data was based
on the Bayesian information criterion (BIC), following the lead of D’Unger, Land,
McCall, and Nagin (1998) and Nagin (1999). The model with the smallest absolute BIC
value is generally selected as the best-fitting model, with BIC rewarding parsimony. In
addition to the BIC criteria, two additional criteria were used to determine the optimal
number of groups to compose the mixture in the present study: 1) significant parameter
estimates for growth terms, and 2) at least 5% of participants in each group. With regard
to the order of growth polynomial, the models were specified using a linear grth
parameter given that studies of natural history of smoking have reported stability or small
drop-offs in adulthood (Anthony & Echeagaray—Wagner, 2000; Chassin et al., 1996b,
2000).

Using these procedures, one-, two-, three-, and four-group models of trajectories
of parental smoking were tested for ages 24 to 50. Non-smokers (p = 293; 52.8% of the
555 parents in the analyzed sample) were a priori identified and excluded from the
analyses. Gender and birth cohort were not meaningful factors when added as covariates;
therefore, subsequent analyses were conducted across genders and birth cohorts. Based
on the BIC criterion as well as other additional criteria, a three-group model was selected
as the best-fitting model. The results are summarized in Tables 2 and 3. Individuals are
then assigned to the group that best fits their observed smoking behavior according to the
maximum posterior probability of group membership.7 This model selection procedure

resulted in the best-fitting model with adequate sample sizes for subsequent analysis.

 

7 The formulas for derivation of likelihood for each individual can be found in Nagin (1999, pp.

156-157).

40

Table 2

Model Comparisog of Pgental Smoking (I! = 262)

 

 

Model BIC A BIC
1 -1667. 17
2 -1532.86 134.31
3 -1530.50 2.36
4 -1530.64 -0.14

 

Note. Bolded row indicates the model selected. BIC: Bayesian information criterion.

Table 3

Growth Parameter Estimates for Each of Three Smoki_ng Groups

 

 

Group Parameter Estimate SE _t_

1. Light Smoker (p = 57) Intercept 3.68 1.08 3.42*
Linear -.09 .03 -3.11*

2. Heavy Smoker (a = 186) Intercept 2.21 .51 4.35*
Linear .06 .02 3.87*

3. Heavy-to-Light Smoker (a = 19) Intercept 9.44 2.01 4.70*
Linear -.18 .06 -3.15*

 

Note. * p < .05, 293 of 555 parents (52.8% of the total parents in the sample) were non-

smokers. SE = Standard error. There were 25 men and 32 women in the Light smoker

category; 105 men and 81 women in the Heavy smoker category; and 13 men and 6

women in the Heavy-to-light smoker category.

41

Table 3 shows growth parameter estimates for each group, and three distinctive
growth patterns of smoking in adulthood are illustrated in Figure 3. Solid lines represent
observed means of smoking whereas dashed lines denote expected value of smoking. All
parameters were significant in each of three smoking groups. The first group of Light
smokers (a = 57; 22.1% of 262 smoking parents; 10% of the 555 smoking and non-
smoking parents) was characterized by a lower level of smoking in quantity and the
gradual decline throughout adulthood (linear grth parameter = -.09, p < .05). Their
level of smoking was confined within one to five cigarettes per day.

This group of smokers may also be called “tobacco chippers” who smoke less
than five cigarettes per day for a long-term without developing nicotine dependence
(Shiffrnan et al., 1994a, 1994b). It is estimated that five percent (0. F. Pomerleau et al.,
1993), or five to ten percent (USDHHS, 1988) of smokers are tobacco chippers or light
smokers in the population at large. A “chipper” was originally used to describe casual
opiate users who use opiates in moderation for a long term without developing addiction
(Shiffrnan et al., 1994a, 1994b). Based on their history and pattern of smoking, Shiffrnan
and his colleagues speculated whether “chippers” are the “smoking equivalent of a
moderate social drinker (Shiffrnan et al., 1994b).” They smoke cigarettes regularly and
they find cigarette smoking reinforcing, but their smoking is often tied to social contexts
and positive affect. Furthermore, their cigarette smoking is not driven by dependence to
compulsive use.

As expected, the majority of smokers belong to the second group, Heavy smokers
(13 = 186; 71% of smoking parents; 33.5% of all parents in the sample). They showed a

persistent, high level of smoking throughout adulthood, ranging from a half pack per day

42

to one and a half pack per day, in average. Moreover, their smoking showed a slight
increase over years in smoking quantity (linear grth parameter = .06, p < .05). This
group of smokers may overlap with habitual smokers with nicotine dependence (with and
without physiological dependence) based on the DSM-IV criteria (American Psychiatric
Association, 1994) in the literature.

The third group (a = 19; 7.3% of smoking parents; 3.4% of all parents in the
sample) consisted of a smaller number of people whose smoking was indistinguishable
from Heavy smokers until early 303 but who either quit smoking or reduced their
smoking under one to five cigarettes per day (linear growth parameter = -.18, p < .05).
Shiffinan et al. (1994b) named a small number of smokers who were previously nicotine
dependent but could not be distinguished from pure “chippers” based on current pattern
of smoking as “converted chippers.” Results confirmed the notion of heterogeneity of
smokers. However, the results from the current study indicate that after late 20s smokers
can be differentiated mostly by the quantity of their cigarette smoking, and that smokers

consist of three groups: Heavy, Light, and Heavy-to-light smokers.

43

53 5a 808 no 838 93 u o 5%
Ba $.03 £2-28 can one u m See on x93 28 n v See com 308 32-25 H M £3 eon 8303mm”. o>m 9 one .I. N Sac Ba 88830
28 SE“ 33 u g Em 3 “oz n o ”wee—2cm >23 2: me 325/ 600833 E wee—95 .«o mEoan 35:80—30“. o>uo=3m=o 085.

.m 2%;

H§HIOHI§8m' I I yam I 34 on...-

 

 

om 3. 8 on
_ > er L h p r h b — - h n n p h n p n —L h bl h P b 1P h P l—
. V‘f t O
I! < Inna-IR” u
_aasuutooc- v
x z. a. F... rilwuHUnI
III .. H
I v
I .
I ) .1 N S
I . m
I o
T m Ur.
.. ¢
5 ‘ 1 .
.. m
4
.. m

 

 

 

_Qllagcteristics of Subtypes of Smokers

I next attempted to determine whether smokers of different subtypes differ from
each other on dimensions other than smoking, such as educational level, occupation,
conduct problems in childhood, antisocial behavior in adulthood, and depression. In
addition to smoking subtypes, gender was also investigated as a fixed factor since
smoking subtypes were originally generated fiom one sample pool across men and
women.

Multivariate analysis of variance (MANOVA) was repeated ten times for each of
ten multiply imputed data sets. Means and standard deviations from each of the ten
imputed datasets were averaged, and test statistics such as E statistic and Wilk’s L were
reported in range.8 Main effects of smoking subtype and gender, and the interaction
between the two were significant at p < .05 (see Table 4). Smoking, gender, and their
interaction explained 20-21%, 21-22%, and 3% of the variance, respectively. This part of
the MANOVA computation was done using S Plus, version 4.5 (MathSoft Inc., 1998).

At the next step, each of seven variables was examined separately using SPSS MANOVA

module, version 10 (SPSS Inc., 1999) and the unique contribution of each of three

 

8 Averaging parameters of interest is a common practice to obtain a combined set of results from
multiply imputed datasets. However, standard errors (and accordingly test statistics) need to be calculated
based on both between and within variances of parameters across m datasets (Rubin, 1987; Schafer, 1997).
Using the standard error and 1 statistic calculated, the null hypothesis, a parameter of interest = 0 can be
tested as in regression analysis or path analysis (see p. 96 in the present study for more information).
However, there is no practical application to obtain a single set of combined results from multiple mean
comparisons. Therefore, alternatively, the range of E statistic and Wilk’s A was reported here in the current

study.

45

sources (smoking subtype, gender, and interaction) was investigated utilizing Type IH
sum of squares.9 This procedure takes into account that all three factors (smoking
subtype, gender, interaction between the two) contribute to each of seven variables, and
that a significant F statistic represents unique input of each factor. A descriptive
summary including standard deviations for each measure is provided in Tables El and E2
in Appendix E.

Post-hoc tests were conducted separately for men and women using the
Bonferroni method to protect familywise error rate.10 Each of ten multiply imputed
datasets was serially analyzed, with results combined to a single set of results using
Rubin’s method (1987) to compute standard error (for more details, see p. 96 in the
following section of Results in the present study). Although the Bonferroni method tends
to be more conservative than other methods, it was adopted in the present study given the
small number of comparisons conducted (three to six comparisons per post-hoc test in the
present study) and the reliance on the 1 distribution for combining results from multiply
imputed datasets.

Figures 4 — 10 reveal results of post-hoe tests. Bars that do not share letters in
both men and women indicate that means differed at p < .05 (a Bonferroni-adjusted alpha
per post-hoe test = .05/6 = 0.0083). Among men, Heavy smokers were different from

others on years of education, occupational status, conduct problems, antisocial behavior,

 

9 Type III sum of squares is a method to divide variance by attributing a unique portion to each
source. It does not depend on the entry order of sources. On the other hand, Type I sum of squares divides
variance sequentially by the entry order as in hierarchical regression analysis.

10 Familywise error rate is the probability to reject at least one true null hypothesis, where the family

refers to the collection of all pairwise null hypotheses.

46

and both current and worst-ever depression. Heavy smokers had a fewer year of
education, a lower level of occupational status, a higher level of conduct problems and
antisocial behavior, and a higher level of both current and worst-ever depression,
compared to non-smokers. On the other hand, on all measures, neither Light nor Heavy-
to-light smokers differed from non-smokers. Among women, Heavy smokers had a
fewer year of education, a lower occupational status, a higher level of conduct problems
and antisocial behavior, and worst-ever depression, compared to non-smokers. Heavy
smoking women had a fewer year of education than Light smokers. With the exceptions
of conduct problems and antisocial behavior that smokers of all types showed higher
levels but no differences among them, neither Light nor Heavy-to-light smokers differed
from non-smokers.

The results suggest that Heavy smoking in both men and women were associated
with a higher level of psychopathologies and a lower level of socioeconomic status.
Although, the direction of the relationship is not clear, it is plausible that these men and
women pose smoking-specific as well as common risks for their children. In addition,
there was a “dose-response” pattern between smoking subtypes and variables tested,
albeit insignificant statistically: Heavy smokers, followed by Light and Heavy-to—light
smokers manifested an elevated level of antisocial behaviors and depression, and a more

disadvantageous socioeconomic status.

47

503838“ £23.88
:38 me some .8.“ 852.5, mo o\cm.m 98 £66 .Xmé .o\%.mm £99 ...\&.3 .xxcw.m~ 8:398 388 2E. .688 :33 .8.“ Sun 28 n

Hoecow 8m 5% 23 fl £28885 can @8093 3325 no.“ Sum 28 m one? 88» m “firearm—S Sm 8039a mo gunmen .3. v a s .802

 

 

 

.53 t of %. t m... .. s; t .53. 9% t a: x85 858:8 seam
.53. t $2. on .. a. a8 t .33. $3 I a? “no; t 3325 Saga:
3:. .. .. Em ea. .. mm. 1.3.4 t ME :36 t .36 “850 t Sesame Sign:

:43 t .58 :3 t a? some t *3: send t .82 82223 5 8328 3832
song t an”: .52 .. :3 of .. o: song t :93 Baguio a masses 8880
:3 t *2; own I 8.2 rook .. is *5: t 3.2 causaeoo
.32 t *Nofl 3. t we. 9% t mam 1.2.8 t 3.2 was.» a 52383
:33. 85885 8200 use; wee—08m 053:3

 

cotspwaeoa 033:: we 88H

 

 

 

$3. I swam. ion. t sown. Ron. I 13.. A $53
39m 1 Led *mmgm .. .3”ch snmdm t sowdfi Gem 5w. 0893532
226885 6980 2585 wcflofim
850m

 

 

 

 

5980 98 08,55 macaw 3 momwofifimmongmm wad moumtoaogno oEdfldeoD no 938% <>OZ<2
e 28¢.

48

15.00

 

14.50 <

14.00 -

13.50 -

Years of education
:3
'o
O

 

Figure 4.

 

Non-smoker

14.24 [3 Men

     

 

 

 

Light smoker Heavy smoker Heavy-to—light
smoker

Smoking subtype

Years of education and smoking subtype. In both men and women, means that do not

share letters differ at p < .05 using the Bonferroni method.

 

Occupation

 

Figure 5.

 

Non-smoker

 

Light smoker Heavy smoker Heavy-to-light
Smoking subtype “Wk“

Occupational status and smoking subtype. In both men and women, means that do not

share letters differ at p < .05 using the Bonferroni method.

49

 

  

 

 

 

 

 

   

14
12.31
12 _ BMen
I Women
a 10 -
2 7 98
.0 7.64
8
O-
‘6
:3
'0
C!
o
O
Non-smoker Light smoker Heavy smoker Heavy-to-light
smoker
Smoking subtype
Figure 6.

Childhood conduct problems and smoking subtype. In both men and women, means that

do not share letters differ at p < .05 using the Bonferroni method.

 

 

 

 

 

 

12
0 El Men
1 ..
H I Women
.2
E
o
.o
'3
'5
o
.9
Non-smoker Light smoker Heavy smoker Heavy-to-light
smoker
Smoking subtype
Figure 7.

Adulthood antisocial behavior and smoking subtype. In both men and women, means

that do not share letters differ at p < .05 using the Bonferroni method.

50

 

    
 
    
       
     
      

  

. BM” 7.16
7 Women §\\\:

 
 

\\

Non-smoker Light smoker Heavy smoker Heavy—to-light smoker

Smoking subtype

/ .

Current depression
#

 

//

  

 

 

Figure 8.
Current depressive symptoms and smoking subtype. In both men and women, means that

do not share letters differ at p < .05 using the Bonferroni method.

 

 

 

 

25
:1 20 ~
.9
§
§. 15 4
'U
33
3 10 -
g
3 5 —

o .

Non-smoker Light smoker Heavy smoker Heavy-to-light
. smoker
Smoking subtype

Figure 9.
Worst-ever depressive symptoms and smoking subtype. In both men and women, means

that do not share letters differ at p < .05 using the Bonferroni method.

51

Parental Smoking Subtype and Early Smoking Onset in Adolescent Children

Parental smoking was examined in relation to adolescent children’s smoking
onset. Maternal (M) and paternal (P) subtypes of developmental pattern of smoking were
crossed with adolescent children’s smoking onset (S) separately in CF A analysis,
yielding a 4 x 3 cross-tabulation for each pair, M x S and P x S. The data were analyzed
under the assumption of total independence (i.e., main effect model) which dictates that
all three classifications are not related at all. If the assumption is violated, types and/or
antitypes should emerge. If the assumption is met, neither types nor antitypes are
expected. Lehmacher’s test (L; Lehmacher, 1981) was used for significance testing of
types and antitypes with a Bonferroni—adjusted alpha level (00" = 0.05/12 cells =
0.004167). The Bonferroni adjustment of alpha was adopted to control for inflated alpha
first due to, simultaneous multiple testing of types and antitypes and second, the mutual
dependency of tests (see von Eye, 1990, 2002).

This analysis resulted in Pearson’s X3 = 21.08 for g = 6; p < .05 for maternal
smoking subtypes (M x S), and Pearson’s Xi = 23.37 for if = 6; p < .05 for paternal
smoking subtypes (P x S). Types and antitypes were found in the same configurations
for maternal and paternal smoking subtypes. In both maternal and paternal smoking
subtypes, parental Heavy smoking was associated with early onset smoking in adolescent
children, L = 2.84, 2 < .05 for maternal; L = 4.27, p < .05 for paternal. Fewer cases than
expected whose smoking onset was unknown were found among children of Heavy
smokers, L = -2.91, p < .05 for maternal; L = -3.36, p < .05 for paternal. Among children
of non-smoking parents, more cases were found than expected of unknown smoking

onset, L = 2.66, p_ < .05 for maternal; L = 2.97, p < .05 for paternal. Conversely, fewer

52

cases than expected started experimenting with cigarettes, L = -4.05, p < .05 for maternal;
L = -3.75, p < .05 for paternal.

The results of pair-wise CFA supported the notion of heterogeneous risks of
parental smoking for adolescent children’s smoking onset. In particular, parental long-
term heavy smoking presented an elevated risk for their children’s early smoking onset.
In contrast, neither Light nor Heavy-to-light smoking parents elevated the chances of
their children’s early smoking onset. Parental smoking had the same effects on children’s
smoking onset, regardless of whether a smoking parent was a mother or father.

At the next step, cumulative risks of parental smoking were investigated by
examining maternal and paternal smoking subtypes altogether. Parental smoking
subtypes were simultaneously crossed with adolescent children’s smoking onset (M x P x
S in 4 x 4 x 3 cross-tabulation). Due to a small number of parents in the Heavy-to-light
subtype (six mothers and thirteen fathers), expected frequencies in a couple of cell
configurations fell below 0.5 in this analysis. Therefore, “Heavy-to-light” category was
collapsed with “Heavy” in subsequent analyses.ll Both maternal (M) and paternal (P)
smoking patterns were crossed with adolescent son’s smoking status (S), yielding the 3 x
3 x 3 cross-classification (M x P x S). Table 5 shows the observed and expected
frequencies and types and antitypes from CFA analysis for boys, and Figure 10 illustrates

the three-way associations among maternal and paternal smoking, and smoking onset of

 

11 Alternatively, “Heavy-to-light” category was collapsed with “Light” category. Results were
almost identical to those results reported in the text, with the exception of one antitype found in cell

configuration 131.

53

adolescent boys”. Again, the data were analyzed under the assumption of total
independence (i.e., main effect model).

As expected, the data on sons showed a poor fit. The Pearson _)_(z = 80.81, for _d_f =
20, p = .00, suggests that there were associations above and beyond main effects. Four
types and three antitypes emerged. Types were found in configurations 111, 113, 232,
and 332. The first two types (i.e., 111 and 113) indicate that there were more cases than
expected of non-smoking parents with adolescent boys who have never smoked by age
14 or whose smoking onset remains to be seen, respectively. The latter two types (232
and 332) suggest that maternal smoking, regardless of its type, when paired with Heavy
smoking on the paternal side was more often associated with early smoking onset in
offspring. Antitypes were found in configurations 133 and 313, indicating that among
adolescent boys whose smoking onset was unknown, fewer cases than expected were
found when they had one Heavy smoker parent and one non-smoker parent.

Figure 10 also captures the associations between subtypes of smoking and early

smoking onset in a mosaic display.13 A glance at Figure 10 reveals that smoking

 

12 Data on daughters showed a similar pattern to that of sons in the sense that there were more cases
of a combination of two Heavy smoking parents paired with daughters who started smoking (Configuration
332), and of a combination of non-smoking parents with non-smoking daughters (Configuration 111; for
more information see Table 01 in Appendix G).

13 The mosaic display, proposed by Hartigan and Kleiner (1981, 1984) is a graphical method for
examining cross-tabulated data. A mosaic, defined as the collection of tiles or rectangles for the n-way
contingency table is formed by dividing a square n times vertically and then horizontally until all cell
configurations are displayed. All mosaic displays in the current study were generated using MOSAICS

developed for the SAS software (SAS Institute, 1999) by Friendly (1992, 1994). More detailed information

54

subtypes of parents were strongly associated in a way that not only was smoking of one
parent was related to the other parent’s smoking, but smoking type was also related to the
other parent’s smoking type. For example, among the three clustered columns of
maternal smoking, the two columns on the right represent smoking mothers. Of smoking
mothers, only a fraction were paired with a non-smoking spouse (the bottom clusters of
rectangles). Of Heavy smoking mothers (the column of clusters of rectangles on the
right), the majority of spouses were also Heavy smokers (the elongated oblong in the
middle of the clusters in the upper right comer). In addition, Figure 10 exhibits that
adolescent sons with early smoking onset were more often found in families with both
parents being Heavy smokers or with Heavy smoking father and Light (and Heavy-to-
light) smoking mothers (i.e., Types 332 and 232; see the two hatch-marked oblongs in the
upper right corner). On the other hand, in families where neither parents smoked, more
adolescent boys than expected were found to have never smoked by age 14 or to remain
non-smoker (i.e., Types 111 and 113; see the two hatch-marked oblongs in the lower left
comer). Two antitypes marked by cross-hatched rectangles (at configurations 133 and
313) suggest that fewer than expected were cases of one Heavy smoking parent and the
other non-smoking parent with sons whose smoking status was unknown or remains to be

8661'].

 

about utilizing mosaic displays for CFA is available elsewhere (Mun, von Eye, Fitzgerald, & Zucker,

2001).

55

Smoking onset

 

 

 

 

 

 

 

 

 

   

No Yes DK N Y D No Yes DK

.5
E

{'53
a :1
OD
:
% a
E is
m E
E l
2 §
:6
a.

Non—smoker Light Heavy
Maternal smoking subtype
Figure 10.

Parental smoking subtypes and adolescent children’s smoking onset. Yes = Smoking

onset by age 14, N0 = Non-smoker, DK = Smoking onset unknown or remain to be seen.

56

Table 5
Configurations of Parental SmokingLPattems and Early Smoking Onset Among

Adolescent Sons

 

 

MPS Obs. Freq. Exp. Freq. L Type/Antitype
111 46 30.88 3.79* Type
112 18 20.59 -.75

113 45 23.65 5.84* Type
121 7 5.94 .52

122 1 3.96 -1.67

123 6 4.55 .78

131 17 27.32 -2.67

132 9 18.21 -2.75

133 7 20.92 -3.95* Antitype
211 2 6.14 -1.93

212 0 4.09 -2.24

213 4 4.70 -.36

221 1 1.18 -.17

222 3 .79 2.57

223 3 .90 2.28

231 5 5.43 -.21

232 9 3.62 3.10* Type
233 4 4.16 -.09

 

57

(table continues)

Table 5 (Continued)
Configurations of Parental Smoking Patterns and Early Smoking Onset Among

Adolescent Sons

 

 

MPS Obs. Freq. Exp. Freq. L Type/Antitype
311 8 16.43 -2.58

312 4 10.95 -2.48

313 3 12.58 -3.23* Antitype
321 2 3.16 —.72

322 1 2.11 -.81

323 1 2.42 -.98

331 23 14.53 2.70

332 29 9.69 7.20* Type

333 12 l 1.13 .31

 

Egg, M = maternal smoking pattern; P = paternal smoking pattern; S = smoking onset
by adolescent boys. Numerals in MPS column represent ordered triples of variable
categories. Response categories for parental smoking were 1 = Non-smoker, 2 =
Light/Heavy-to-light smoker, and 3 = Heavy smoker for parental smoking, and options
for adolescent smoking onset were 1 = Never smoked, 2 = Smoked by age 14, and 3 =
Smoking-onset unknown. L stands for Lehmacher’s test statistic (1981); Bonferroni-

adjusted alpha, of“ = 0.00185 was used; * significant at or" = 0.00185.

58

The results indicate that parental smoking poses different levels of risks for their
children’s smoking onset. Smoking type appears to be a very useful concept for studying
parental smoking as a risk factor for adolescent smoking onset. Long-term Heavy
smoking by parents, regardless of whether it was maternal or paternal was tied to early
smoking onset among adolescent children. It was well demonstrated in pair-wise
associations of adolescent children’s smoking onset with maternal and paternal smoking
subtype. However, when both parental smoking subtypes in two-parent families were
crossed with children’s smoking onset it was the presence of both smoking parents that
sufficiently elevated the likelihood of early smoking onset among adolescent children.
Although one spouse’s smoking was highly related to the other’s smoking, parental
smoking did not influence sons’ smoking onset above and beyond statistical significance,
when only one parent smoked. There also appears to be specificity of parental smoking
in the relation to early onset of smoking in adolescent children. Whereas Light smoking
by fathers did not pose much risk, Light smoking by mothers, was a significant risk factor

when linked to paternal Heavy smoking.

59

A FAMILY HISTORY OF ALCOHOLISM AS A RISK FACTOR

Developmentgl Patterns of Alcoholism

Natural developmental patterns of alcoholism were modeled in the present study
using the same procedures as in the analysis of parental smoking trajectories. Males and
females were analyzed separately for a number of theoretical and empirical reasons.
First, in the literature of alcoholism, the etiology of women’s alcoholism is considered to
be somewhat different from that of men, although it is uncertain as to what extent as well
as to what kind (Babor et al., 1992; Cloninger, 1987; Fitzgerald, Zucker, Puttler, Caplan,
& Mun, 2000; Zucker, 1987; Zucker et al., 2000; Zucker et al., 1995). Second, the age
range covered for men and women in the current study was different, with women’s
alcoholism documented over a shorter span of adulthood (e.g., ages 14 to 49 for women
versus 14 to 54 for men). Separate analysis by gender was necessary in the context of
unequal end point of observations since the prevalence of both alcoholism and drinking
tend to dwindle with increasing age (Anthony & Echeagaray—Wagner, 2000; Zucker et
al., 2000) and therefore, it was likely that the different end point of data observations for
men and women prevent one from revealing the true developmental patterns of
alcoholism in adulthood, if analyzed in one sample. Third, men and women in the
ongoing larger UM-MSU project were recruited based on different criteria. Women’s
alcoholism was neither a requirement nor a basis for exclusion, whereas alcoholic men
were deliberately recruited in the study. Fourth, gender was a significant covariate when
all participants were pooled and analyzed together in the initial TRAJ procedure.

Because the natural courses of alcoholism and alcohol use show patterns of

gradual decline afier peaking during the twenties (Anthony & Echeagaray-Wagner, 2000;

60

Zucker et al., 2000), the models were specified using linear and quadratic growth
parameters. One-, two-, and three-group models of trajectories of alcoholism were tested
separately for men and women. Non-alcoholic men (n = 73; 26.5% of 275 men in the
sample) and women (Q = 171; 61.1% of 280 women in the sample) were a priori
identified and excluded in the subsequent analyses. Birth cohort was not a meaningfirl
factor when added as a covariate; therefore, subsequent analyses were conducted across
birth cohorts. A two-group model was selected as the best-fitting model for both men
and women. The results are summarized in Table 6. Table 7 shows growth parameter
estimates for each group and Figures 11 and 12 depict distinctive developmental patterns
of alcoholism for men and women in adulthood, respectively. Solid lines represent
observed means of alcoholism whereas dashed lines denote predicted value of alcoholism

diagnosis.

Table 6

Model Comparisons of Parkergl Alcoholism

 

 

 

 

Men (p = 202) Women (3 = 109)
Model BIC A BIC Model BIC A BIC
1 -1976.79 1 -944.98
2 -1885.94 90.85 2 -899.67 45.31
3 -2221.47 -335.53 3 -957.70 -58.03

 

Note. Bolded row indicates the model selected. BIC: Bayesian information criterion.

61

608658:

333—293: :23 8:28:08: 3:82 u m ”00:35:26 329$: 3055 00:26:08.6 3:82. .1. N woman—a 3:82 n _ 869%an 02

u o ”magma: QD< Mo 8252 8628:: .5286 um: 3:83. u £833: DD< :08 mafia 8:38? we 8:23: _Sfianofifin

u QWHOHHRVZ III...

om 3

h hip PHrL blthPi—ibibb b bbb

Hnuzonoozlllll

 

 

 

trfi

I'I’YYIVV

Y'TV'V'I

 

 

1P 011V

srsousn

.: use:

62

005308: 328—293:
5:5 85:88: 3:003. n m ”00:02:08: 323—06.3: 305:.» 85:58: “0:00? n N ”0030 3:002 u fl ”magma: 02 n
o 6.60:me Qb< .«0 m020> 860:3: 8:804: 000 3:003. u magma: n5< .5803 30:8 530:8? :0 2.530: 3:08:0_0>0Q

.2 2%:

m aflfiogooz Ill-II H EHOAHOOE 'll

 

 

 

0w<
om o: om ON 9.
h Luir h P bl l- .P P b h .1 h — P P b r L — BI h h .- b L h P D — h h I P I I P D r
_ . «l .. y . n o
1911.! inn—Hull . D ‘ _-
VA: 4 «o’fi .
O .
0000 .
.. . w V
.o. u H
o n G
09 90 m D...
W cm
I’D... 0 W N w
.I-roolu..olo-IIIIIIOI- 9-
fl

 

 

IVITYTV'

 

 

 

 

 

63

Table 7

Growth Parameter Estimates for Each of Two Alcoholisms

 

 

 

Group Parameter Estimate SE t

M311 (2 = 202)

1. Alcoholism I (n = 103) Intercept -21.01 2.80 -7.51*
Linear 1.29 .22 5.88*
Quadratic -.02 .00 -5.68*

2. Alcoholism H (n = 99) Intercept -30.30 2.63 -l 1.50*
Linear 1.88 .18 1024*
Quadratic -.03 .00 -8.98*

m (n = 109)

1. Alcoholism I (n == 79) Intercept -12.93 3.00 -4.32*
Linear .52 .22 235*
Quadratic -.Ol .00 -2.34*

2. Alcoholism II (I_‘1 = 30) Intercept -24.37 5.86 -4.16*
Linear 1.31 .46 2.86*
Quadratic —.Ol .01 -1.78*

 

Note. "‘ p < .05. 73 (13.2%) of 275 men and 171 (30.8%) of 280 women never met a

positive alcoholism diagnosis. SE = Standard error.

In both men and women, two types of developmental patterns of alcoholism
emerged as expected from the literature of alcoholism. All linear and quadratic growth
parameters were significant (see Table 7). The first group of alcoholic men (Alcoholism
I, n = 103; 51.0% of alcoholic men; 37.5% of all 275 men in the sample) showed an
idiosyncratic developmental pattern characterized by a less severe kind of alcoholism
diagnosis (i.e., alcohol abuse) over the course of adulthood, with a peak at late 205 and
gradual decline thereafter. The second group of alcoholic men (Alcoholism I], n = 99;
49.0% of alcoholic men; 36% of all men in the sample) revealed a developmental course
characterized by a severe type of alcoholism diagnosis (i.e., alcohol dependence) for the
most of adulthood life span that peaked at mid 30s but gradually declined afterwards (see
Figure 11).

Among women, the first group of alcoholic women (Alcoholism I, n = 79; 72.5%
of 109 alcoholic women; 28.2% of all 280 women in the sample) showed a pattern of
alcoholism that was confined within the diagnosis of alcoholism abuse. Although there
was no sharp peak or drop-off in their alcoholism pattern, the pattern of alcoholism
diagnosis at ages 14 and 49 for this group of women showed a combined flat shape of
linear (.52, p < .05) and quadratic (-.Ol , p < .05) components. The second group of
alcoholic women (Alcoholism II, n = 30; 27.5% of 109 alcoholic women; 10.7% of all
280 women in the sample) revealed a developmental course characterized by a severe
diagnosis of alcoholism over time with a slight decline in 405 (-.01, p < .05; one-tailed),
without a clear drop-off but as illustrated in Figure 12. It can be attributed to the scarcity
of data observations after age 40 among women, causing relatively large standard error as

indicated by the fluctuations after age 40 depicted in Figure 12. It remains to be seen

65

whether Alcoholism H among women tapers off after late 40$ fi'om future follow-up
studies.

Although there are existing terms for the subtypes of alcoholism (e. g., Types I and
H (Cloninger, 1987), Types A and B (Babor et al., 1992), and Antisocial alcoholism and
Non-antisocial alcoholism (Zucker et al., 1996), Alcoholism I and Alcoholism H were
used throughout the present study. The rationale is as follows. In theoretical and
empirical studies of typology of alcoholism, different samples (and populations) of
alcoholics as well as various methods were used to derive alcoholism subtypes. Although
there are some convergence in the literature in that one type (Type II, Type B, and
Antisocial alcoholism) is generally regarded as a more severe expression of alcoholism
than the other, with other co-active psychopathologies and a denser family history of
alcoholism, it is not clear that, to what extent, the two types of alcoholics in the present
study are equivalent to the types of alcoholism in the extant literature. Therefore, instead
of adopting the existing terminology, Alcoholisms I and H were used in the current study
to differentiate the two kinds of developmental patterns of alcoholics.

Characteristics of Subtypes of Alcoholism.

Since subtypes of alcoholism were derived separately for men and women, they
were separately tested on the following measures: Education in years, occupational
status, conduct problems in childhood, antisocial behavior in young adulthood, and
depression. In both groups of men and women, alcoholism subtype was a significant
factor (see Table 8). It explained approximately 25% of the variance in both groups of
men and women (Wilk’s lambda = .764 — .775; p < .05 for men, and .742 — .761; p < .05

for women). Univariate analysis on each of the seven measures revealed that in all

66

measures with the exception of self-reported depression, there was a group difference
across subtypes of alcoholism for men (see Table 8 and Figures 13 — 19). As for women,
with the exception of occupational status, alcoholism subtype differentiated
developmental patterns of alcoholism (see Table 8 and Figures 13 — 19).

Post-hoc tests were conducted separately for men and women using the
Bonferroni method. Ten multiply imputed datasets were analyzed separately and then
results were combined following the previous procedure used in post-hoc tests of
smokers. In both men and women, bars that do not share letters indicate that means
differed at p < .05. Among men, alcoholics had a fewer year of education, a lower
occupational status, a higher level of conduct problems, antisocial behavior, and worst-
ever depression. Among alcoholic men, men with Alcoholism II were associated with a
higher level of antisocial behavior and depression, compared to Alcoholism 1. Among
women, alcoholic women had a higher level of conduct problems childhood and
antisocial behavior, and worst-ever depression, compared to non-alcoholics. Among
alcoholic women, women with Alcoholism II had a higher level of conduct problems and

antisocial behavior, compared to women with Alcoholism I.

67

Table 8

MANOVA Results on Demographic Characteristics and Psychopathologies of Two

Alcoholisms for Men and Women

 

Men

Women

 

 

Multivariate E

Wilk’s A

11.07* —— 11.76*

.764* — .775*

12.17* —13.53*

.742* — .761*

 

Variable

Univariate analysis

 

Education in years

Occupation

Conduct problems in childhood
Antisocial behavior in adulthood
Hamilton depression — current
Hamilton depression - worst

Beck depression index

1055* — 11.69*

928* — 1022*

17.78* — 19.21*

2908* — 3072*

9.02* — 1000*

14.12* —15.92*

2.55 — 3.22*

324* -— 3.39*

.39 — .47

2952* -— 34.16*

2999* — 3354*

4.00* —- 4.81 *
6.78* — 8.39*
4.44* - 6.00*

 

Note. * p_ < .05. Degrees of freedom for multivariate analysis of variance for men and

women were 1, 273 and l, 278, respectively; degrees of freedom for univariate _E for men

and women were 2, 272 and 2, 277, respectively. Alcoholism subtype explained 7.7%,

6.8%, 12.0%, 17.9%, 6.5%, 10.0%, and 2.1% of variance respectively for each of seven

measures for men. Alcoholism subtype explained 2.4%, 0%, 18.6%, 18.7%, 3%, 5.1%,

and 3.8% of variance respectively for each of seven measures for women.

68

 

[3 Men
Women

13.00 -

12.50 ~

Years of education

12.00 5

11.50 -

 

 

 

 

 

 

 

1 1.00
Non-alcoholic Alcoholism l Alcoholism ll

Alcoholism subtype

Figure 13.
Years of education and alcoholism subtype. In both men and women, means that do not

share letters differ at p < .05 using the Bonferroni method.

 

Occupation

 

 

 

Non-alcoholic Alcoholism 1 Alcoholism ll

Alcoholism subtype

Figure 14.
Occupational status and alcoholism subtype. In both men and women, means that do not

share letters differ at p < .05 using the Bonferroni method.

69

 

E] Men
10 ‘ 1 Women

Conduct problems

 

 

 

 

 

 

Non-alcoholic Alcoholism l Alcoholism ll
Alcoholism subtype
Figure 15.
Childhood conduct problems and alcoholism subtypes. In both men and women, means

that do not share letters differ at p_ < .05 using the Bonferroni method.

 

Antisocial behavior

 

 

 

Non-alcoholic Alcoholism l Alcoholism ll

Alcoholism subtype

Figure 16.
Adulthood antisocial behavior and alcoholism subtypes. In both men and women, means

that do not share letters differ at p < .05 using the Bonferroni method.

70

 

Current depression
'J’l

 

 

   

 

4 .
3 -
2 - _ _‘iw'
. ' ' v‘ \-’1 . V. ‘-
l- (2:;
0 _ -- \- $5
Non-alcoholic Alcoholism l Alcoholism ll
Alcoholism subtype
Figure 17.

Current depressive symptoms and alcoholism subtype. In both men and women, means

that do not share letters differ at p < .05 using the Bonferroni method.

 

   

       
  
  
  

 

 

25
E3 Men
20 q . Women 19.23
g 17.70 _
H < :53 :3
e '5 ss‘
‘5 {‘3‘ ix:-
§
9 10 -
8
3 5 _
0 ..
Non-alcoholic Alcoholism l Alcoholism ll
Alcoholism subtype
Figure 18.

Worst-ever depressive symptoms and alcoholism subtype. In both men and women,

means that do not share letters differ at p < .05 using the Bonferroni method.

71

 

 
  
  

5.00
4.50 ‘ D Men
8 4.00 < Women
'5
(I)
8
8‘
'U
'8
1:
o
5
4L
3
m

 

 

Non-alcoholic Alcoholism l Alcoholism ll

Alcoholism subtype

Figure 19.
Self-reported depressive symptoms and alcoholism subtype. In both men and women,

means that do not share letters differ at p < .05 using the Bonferroni method.

72

Commonality Between Subtypes of Smokers and Alcoholics

The commonality between subtypes of smokers and subtypes of alcoholics was
investigated using a two-sample CFA with the z-approximation of the binominal test
(_z_*). The major purpose of the two-sample CFA analysis was to see 1) whether smoking
types were associated with alcoholism types, and 2) whether gender discriminated the
relationship between smoking and alcoholism. Four categories of smoking types (S)
were crossed with three categories of alcoholism types (A). This categorization scheme
yielded the 4 x 3 cross-classification (S x A) for men and women. Table 9 and Figure 20
show the observed and expected frequencies and types and antitypes from CF A analysis
for both men and women. Following a two-sample CFA, I subsequently analyzed the
data under the assumption of total independence and Lehmacher’s test was used for
significance testing of types and antitypes with a Bonferroni-adjusted alpha level (or* =
0.0042).

As expected, the data showed a poor fit, Pearson X2 = 69.01, for fl = 6, p = .00
for men; Pearson 52 = 37.38, for g = 6, p = .00 for women, suggesting associations
above and beyond main effects. Two types and two antitypes emerged for both men and
women at the same configurations. Types were found at configurations 11 and 33 while
antitypes were found at configurations 13 and 31. The results indicated that regardless of
gender, there were more cases of neither smoker nor alcoholic than expected
(Configuration 11; a hatch-marked rectangle in the lower left comer in Figure 20) and
that Heavy smokers were more often than expected Alcoholism 1] (Configuration 33; a
hatch-marked rectangle in the upper right comer in Figure 20). Antitypes indicate that

cases of Heavy smokers who were non-alcoholic (Configuration 31: a cross-hatched

73

rectangle in the lower right corner) and cases of non-smokers with Alcoholism II
(Configuration 13: a cross—hatched rectangle in the upper left comer) were found less
frequently than expected.

In addition, a two-sample CF A test showed that men and women had unequal
number of cases in five configurations (see Table 9). Five discrimination types (DT)
were found to differentiate men from women. More cases of women than men were
found in configurations ll, 21, and 31, while more observations of men than women
were found in configurations 13 and 33. In more detail, there were more women than
men in the categories of Non-smoker and Non-alcoholic, z* = -4.63, p < .05
(Configuration 11), of Light smoking and Non-alcoholic, z* = -2.91, p < .05
(Configuration 21), and of Heavy smoking and Non-alcoholic, z“ = -4.54, p < .05
(Configuration 31). On the other hand, there were more cases of men than women in the
categories of Heavy smoking with Alcoholism I], z“ = 4.63, p < .05 (Configuration 33)
and of Non-smoker with Alcoholism 11, z’“ = 3.56, p_ < .05 (Configuration 13).

The discrepancy between men and women can also be shown in Figure 20
Although types and antitypes were found at the exactly same locations for men and
women, the number of cases (i.e., the size of rectangle) was visibly and statistically
different for men and women in the five configurations identified as discrimination type
(DT) above. For example, a configuration 31 (a cross-hatched rectangle in the bottom
right) can be interpreted as follows: Although there were fewer cases than expected of
Heavy smokers who were non—alcoholic in both men and women, these rare cares were

observed more often in women than men. Results illustrate that an overall pattern of

74

association between smoking and alcoholism holds true for both men and women, with
some gender-specific characteristics.

The discrepancies between men and women in the observed frequencies of two-
way associations in the present study were not due to the recruitment criteria; one-way
marginals of both alcoholism types and smoking subtypes were taken into consideration
in examining two-way associations across gender. The results confirm the well-known
association between alcoholism and smoking. Furthermore, the results support the notion
that the nature of the relation between alcohol and tobacco-related phenomena may
depend on levels of involvement with alcohol and cigarette smoking (use versus
dependence), with dependence more linked to two specific but related factors (Prescott &
Kendler, 1995). The common thread between Alcoholism H and Heavy smoking in the
present study was dependency. Unique pathways for each of these dependences as well

as the related risk factors and mechanisms remain to be studied.

75

 

 

 

 

 

 

 

 

 

 

 

 

 

      

 

 

 

 

 

 

 

 

 

  

 

 

 

 

 

E
‘0
t?
Z
o _
s g
.D
a 2
E
.2
'6'
..= _
.93 1% r
< g ——
I __
g a Type
_ Antitype
Nm—srmler LQKITUQI Heavysrmlcer Haida—light
Smoking subtype
Women
S
E — I:
E . _
a __
E
B a
.c:
a 3 .
<2 ' '
Q
= Type
- Antitype
. . ’I . ’ '-
hbn—srmkar Lign srmler Heavysrmker Heevy—to—hgtx
Smoking subtype
Figure 20

Associations among subtypes of smoking and alcoholism

76

Table 9

Associations Among Subtypes of Smokers and Alcoholics

 

 

 

 

Men (E = 275) Women (fl = 280)
SA OF EF L T/A OF EF L T/A DT
11 60 35.04 6.81* Type 112 98.33 3.38* Type DT
12 47 49.44 -.61 43 45.43 -.65
13 25 47.52 -5.65* Antitype 6 17.25 -4.39* Antitype DT
21 6 6.64 -.30 21 19.54 .56 DT
22 14 9.36 2.01 10 9.03 .41
23 5 9.00 -1.75 1 3.43 -l .47

31 7 27.87 -5.86* Antitype 36 49.47 -3.63* Antitype DT

32 38 39.33 -.34 23 22.85 .04
33 60 37.80 5.73* Type 22 8.68 5.67* Type DT
41 O 3.45 -2.22 2 3.66 -1 .41
42 4 4.87 -.51 3 1.69 1.20
43 9 4.68 2.55 1 .64 .48

 

Egg 8 = smoking subtype; A = alcoholism subtype. Numerals in SA column represent
ordered doubles of variable categories. Response categories for smoking were 1 = Non-
smoker, 2 = Light smoker, 3 = Heavy smoker, and 4 = Heavy-to-light smoker. Options
for alcoholism subtype were 1 = Non-alcoholic, 2 = Alcoholism I, and 3 = Alcoholism H.
OF = Observed frequency, EF = Expected frequency, T/A = Presence of Type or
Antitype, L = Lehmacher’s test statistic (1981), DT = Discrimination type. Bonferroni-

adjusted alpha, (1* = 0.0042 was used; * significant at or“ = 0.00185.

77

Parental Alcoholism Subgges and Early Smoking Onset in Adolescent Children

 

To address whether parental alcoholism subtypes are associated with adolescent
children’s smoking onset, frequencies of parental alcoholism and smoking onset were
examined. Three categories of maternal (MA) and paternal (PA) alcoholism patterns were
crossed with adolescent children’s smoking status (S). This cross-tabulation yielded the
3 x 3 x 3 cross-classification (MA x PA x S)”. Table 10 shows the observed and
expected frequencies and types and antitypes from CFA analysis for sons. The same
assumption of total independence, and the same statistical procedure used in the MPS
data were adopted. Adolescent sons’ data showed a poor fit, Pearson X3 = 60.73, for g =
20, p = .00, suggesting that there were associations among the three classifications. Two
types and one antitype emerged (see also Figure 21). Types were found in cell
configurations 111 and 332. Results indicated that there were more cases of non-
alcoholic parents whose adolescent son has not tried smoking (Configuration 111; a
hatch-marked rectangle in the lower left comer), and that there were more cases of
Alcoholism II by both parents whose adolescent son started smoking cigarettes by age 14
(Configuration 332; a hatch-marked oblong in the upper right corner). There was one

antitype at configuration 133 indicating that cases of adolescents whose smoking onset

 

14 Paternal alcoholism and maternal alcoholism were separately examined in their associations with
smoking onset among adolescent children (3 x 3 cross classification). Results revealed that maternal
alcoholism subtypes were not associated with children’s smoking onset (Pearson’ X} = 2.11, p = .72).
However, paternal alcoholism subtypes were related to children’s smoking onset (Pearson’ 2? = 13.79, p <
.05) and one type and one antitype were found. There were more cases than expected of early smoking
Onset with paternal Alcoholism 11, L = 3.01; p < .05. Fewer cases than expected were found among

adolescents with paternal Alcoholism 11 whose smoking onset was not lmown, L = -2.70; p < .05.

78

was not known in two-parent families in which mother was non-alcoholic mother but
father had Alcoholism H were fewer than expected (see the cross-hatch marked rectangle
on top in Figure 21).

Results suggest that the salience of paternal alcoholism over maternal alcoholism
as a risk factor for early smoking onset in offspring. Furthermore, results point out that
the risks of early smoking onset were higher for sons of fathers with Alcoholism II. A
milder form of alcoholism over long-term did not appear to pose much risk to children.
In a striking resemblance to the results of smoking parents, results appear to emphasize
the importance of both parents in two-parent families in the sense that it took two
alcoholic parents to steer children to the smoking path. Even with a severe type of
alcoholism (i.e., Alcoholism H), it was the combined effects of alcoholism on both

parents’ side that elevated the risks for adolescent sons’ smoking onset.”

 

15 Data on daughters was inconclusive, due to the small observed and expected frequencies. One
antitype was found at configuration 113, indicating that there were fewer cases than expected of adolescent
daughters, whose smoking onset was unknown, had non-smoking parents (for more information, see Table

oz in Appendix G).

79

Smoking onset

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No Yes OK No Yes DK N Y D

.5 E

g :
é . E

E

g _ E
e E, E
.8 a: E
o :
.2 .__
«3
'5'
E .9
8 E
(6
°‘ r

8

2 III

 

 

 

   

 

Nan-Alcoholic Alcoholism I Alcohol ism ll

Maternal alcoholism subtype

Figure 21.
Parental alcoholism and smoking onset among adolescent sons. Yes = Smoking onset by

age 14, N0 = Non-smoker, DK = smoking onset unknown.

80

Table 10

Configurations of Parental Alcoholism Patterns and Early Smoking Onset Among

Adolescent Sons

 

 

 

MAPAS Obs. Freq. Exp. Freq. L Type/Antitype
111 30 17.95 3.64* Type
112 11 11.96 -.34

113 22 13.74 2.73

121 20 25.53 -1.47

122 18 17.02 .30

123 24 19.55 1.29

131 17 24.77 -2.09

132 17 16.51 .15

133 7 18.97 -3.52* Antitype
211 4 8.11 -1.65

212 0 5.41 -2.57

213 3 6.21 -1.44

221 14 11.53 .86

222 5 7.69 -1.10

223 12 8.83 1.22

231 15 11.19 1.34

232 12 7.46 1.88

233 10 8.57 .56

 

81

(table continues)

Table 10 (Continued)
Configurations of Parental Alcoholism Patterns and Early Smoking Onset Among

Adolescent Sons

 

 

 

MAPAS Obs. Freq. Exp. Freq. L Type/Antitype
311 0 3.14 -1.93

312 1 2.09 -.80

313 0 2.40 -1.66

321 3 4.46 -.77

322 2 2.97 -.61

323 3 3.42 -.25

331 8 4.33 1.97

332 8 2.89 3.26* Type

333 4 3.31 .41

 

flog MA = maternal alcoholism pattern; PA = paternal alcoholism pattern; S = smoking
onset by adolescent sons. Numerals in MAPAS column represent ordered triples of
variable categories. Response categories for parental alcoholism were 1 = Non-alcoholic,
2 = Alcoholism I, and 3 = Alcoholism H. Options for adolescent smoking onset were 1 =
Never smoked, 2 = Smoked by age 14, and 3 = Smoking-onset unknown. L stands for
Lehmacher’s test statistic (1981); Bonferroni-adjusted alpha, of" = 0.00185 was used; *

significant at d“ = 0.00185.

82

In summary, so far results indicate that there are differential risks involved, with
dependent types of usage patterns of cigarette smoking and alcoholism related to a higher
risk for their adolescent sons’ smoking onset. However, in both cases of parental
smoking and alcoholism, it was the combination of both parents that exerted any impact
on adolescent children’s smoking onset. As for the relative importance of paternal versus
maternal, it appears that at least in the case of alcoholism, paternal alcoholism is a more
important factor for early smoking onset in offspring.

From the previous results on parental smoking and alcoholism where dependent
types of smoking and alcoholism were found associated with one another, it is plausible
that children’s risks for early smoking onset would accordingly increase, as parental
dependence on smoking is stacked upon Alcoholism H. Unfortunately, it was not
possible to investigate all five factors together (i.e., maternal smoking and alcoholism
subtypes, paternal smoking and alcoholism subtypes, and adolescent children’s smoking
onset) in the current study since it requires at least 243 cell configurations (3 x 3 x 3 x 3
x 3). However, a large scale national longitudinal data may provide further insights into
this issue of aggregated risks of parental substance abuse for their children’s smoking

onset and usage, and the specific patterns of risks in the future.

83

EARLY AND CONCURRENT CHARACTERISTICS OF ADOLESCENTS WITH

EARLY SMOKING ONSET

In this section, adolescents who start experimenting with cigarette smoking early
were compared on measures of prenatal exposure to maternal smoking and drinking,
early temperament and behavioral characteristics as well as concurrent child behavioral
characteristics. In particular, it was hypothesized that adolescents with early smoking
onset would be characterized by 1) a higher level of prenatal exposure to maternal daily
cigarette smoking and/or weekly drinking, 2) early temperament dimensions assessed at
ages three and five, and 3) a higher level on each of the four CBCL syndrome scales rated
by both mother and father at child ages three and five, and once again at child ages
twelve and fourteen: Anxious/Depressed, Attention problems, Delinquent behavior, and
Aggressive behavior. MANOVA was conducted for each of ten multiply imputed
datasets, and the results are presented in average parameters (i.e., average mean and
standard deviation) and the range of test statistics.

Prenatal Exposure to Maternal Smoking and Drinking

Prenatal exposure to maternal smoking and drinking was examined for both sons
and daughters. Information on their prenatal history was obtained for almost all children
included in the current study. Therefore, prenatal exposure to maternal smoking and
drinking was investigated and presented for both sons and daughters. Prenatal exposure
to daily maternal cigarette smoking was related to smoking onset of sons, E(2, 278) =
3.29 - 4.26, p < .05, and daughters, E(2, 85) = 6.12 — 7.81, p < .05 (see Figure 22 and

Tables F1 and G3 in Appendices F and G).

84

Post-hoc tests using the Bonferroni method revealed that sons who started
cigarette use by age 14 had a higher level of prenatal exposure to maternal daily smoking,
compared to those whose smoking onset was unknown. No significant difference was
found between sons with early smoking onset and those who never smoked. Daughters
who started smoking by age 14 had a higher level of maternal smoking, compared to all
others. Results supported the latest research findings that prenatal exposure to maternal

smoking is a risk factor for early smoking onset in offspring.

 

7 < 1]] Sons
E Daughters

Maternal smoking during pregnanc

 

 

 

Smoking by age 14 Non-smoker Smoking onset unknown

Figure 22.
Prenatal exposure to maternal smoking and adolescent children’s smoking. Means that

do not share letters differ at p < .05 using the Bonferroni method.

In contrast, results on maternal weekly drinking during pregnancy as a risk factor
for early smoking onset were inconsistent. It turned out to be a significant factor for

early onset of smoking in only four of ten multiply imputed datasets among sons, E(2,

85

278) = 2.28 — 4.47; p = .104 — .012. It was not a significant factor in all ten datasets

among daughters, E(2, 85) = 1.80 — 3.08, g.

Precursors and Concurrent Characteristics of Adolescents of Early Smtiing Onset

In this analysis, multivariate analysis of variance was conducted to see whether
early smoking onset status was related to the host of early measures, including prenatal
exposure to daily smoking and weekly drinking, child temperament and behavioral
characteristics measured at ages between three and five, and behavioral characteristics
measured at ages between twelve and fourteen. Sons of various smoking onset status
were different, multivariate E = 2.44 — 3.52, p < .05 with maternal ratings; E = 2.59 ——
3.66, p < .05 with paternal ratings (see Table 11 and Figures 23 - 30). Approximately
11% — 15% of the variance for variables entered in MAN OVA was explained by smoking
status of adolescent sons.16

Overall, maternal ratings of early child temperamental characteristics and
behavioral characteristics proved to be a better indicator of early smoking onset among
adolescent sons than paternal ratings. Based on maternal ratings, early measures of child
temperament and four behavioral syndromes and concurrent measures of behavioral
syndromes distinguished adolescents who subsequently started cigarette use early from
those who did not (see Figures 23 — 26). On the other hand, paternal ratings were not as

predictive as maternal ratings (see Table 11). Based on paternal ratings, a temperament

 

16 The data on daughters was inconclusive, albeit some apparent resemblance to results of sons, due
to a small sample of younger cohorts. MANOVA results are presented in Appendix G (see Table G3), and

descriptive statistics in Tables G4 and G5.

86

dimension. A
Anxious dcp
were predict
Post
Bonferroni
again Map
(For more
H.160 M 3
those win
Who Stan
Remy SIT
01 adnle
results 3
(Meter
Reacti'

‘0 “Eu

adole
“Ere
Prob
lhre.

Cha]

dimension, Approach-Withdrawal at ages three to five, and concurrent measures of
Anxious/depressed, Attention problems, Delinquent behavior, and Aggressive behavior
were predictive of smoking onset in sons.

Post-hoc tests were conducted on each of significant measures using the
Bonferroni method of pairwise comparisons. The same procedure used previously was
again adopted to combine results of post-hoe tests from ten multiply imputed datasets
(For more details, see p. 96 in the present study). Adolescents who never smoked were
rated by mother more attentive (and less distractible) at ages three to five, compared to
those whose smoking onset was unknown (see Figure 23). In addition, adolescent sons
who started cigarette use by age 14 were rated more reactive, compared to those who
never smoked, based on maternal ratings. There were no differences across three groups
of adolescents on a temperamental dimension, “Approach-Withdrawal.” However,
results from paternal ratings were different from those of maternal ratings in that no
differences were found on both temperamental dimensions of Attention span and
Reactivity. Yet, those who started cigarette use by age 14 were rated more approaching
to new stimuli and people at ages three to five by father (see Figure 24).

Early behavioral characteristics were markedly different across three groups of
adolescents based on maternal ratings. Adolescents who started cigarette use by age 14
were rated by mother as being more anxious and depressed, having more attention
problems, and more often displaying both delinquent and aggressive behavior at ages
three and five (see Figures 25 -— 28). Paternal ratings of child early behavioral
characteristics did not differ across three groups of adolescents (see Figures 29 — 30).

Concurrent behavioral characteristics rated by mother revealed that adolescents with

87

early onset of:
aggressor bet
characteristic
delinquent at
smoking ons
characteristi

those \th r

In
to matem
Precursor
Were D31
011561, a
IE'mper:

exist 51
be‘ttax
qUesr

OHSer

early onset of smoking had more often attention problems, delinquent behavior, and
aggressive behavior at ages 12 to 14 (see Figures 26 — 28). Parental ratings of behavioral
characteristics converged more on measures of concurrent behaviors. Paternal ratings on
delinquent and aggressive behaviors were discriminating adolescents of different
smoking onset statuses (see Figures 29 — 30). In all measures of behavioral
characteristics, adolescents whose smoking onset was unknown were not different from

those who never smoked by age 14.

In summary, results in this section support the hypotheses of 1) prenatal exposure
to maternal cigarette smoking and 2) early discriminating child characteristics as
precursors of early smoking onset. Overall, results support the notion that children who
were parentally exposed to maternal daily cigarette smoking are at risk for early smoking
onset, and that there are identifiable early precursors of early smoking onset
temperamentally and behaviorally. Results so far point to the possibility that there may

exist some early constitutional vulnerability manifested in child temperament and

 

behavioral characteristics as well as familial vulnerability. The next section addresses

 

questions related to the mechanisms of how these factors contribute to early smoking

onset.

88

D

A\M.\m.

Table 11
MANOVA Results on Precursors and Concurrent Characteristics of Adolescents with

Differing SmokingOnset Status

 

 

 

 

 

 

 

Maternal rating Paternal rating
Multivariate .13 2.44* — 352* 259* - 3.66*
Wilk’s A .854* — .894* .849* — .888*
Variable Univariate analysis
Prenatal emosure
Daily maternal smoking 329* — 426*
Weekly maternal drinking 2'28 " 447*
EarlLtemperament (Ages 3 — 5)
Attention span 326* — 435* 2.00 — 3.30*
Approach/Withdrawal .27 — .76 3.35* — 4.40*
Reactivity 3.08* — 356* 1.65 —— 2.65

Early child behavior problems (Ages 3 — 5)

Anxious/Depressed 6.42* — 8.67* 1.33 — 2.17
Attention problems 377* — 524* 1.31 — 1.60
Delinquent behavior 7.15* — 9.65* 1.24 — 1.69
Aggressive behavior 2.85 — 5.35* 1.23 -— 1.56

Concurrent adolescent behavior problem_s (Ages 12 - 14)

Anxious/Depressed 2.87 - 620* 1.38 — 9.62*
Attention problems 530* — 7.99* 3.74* - 9.47*
Delinquent behavior 17.02* — 22.92* 1322* — 18.98*
Aggressive behavior 8.31* — 11.80* 5.32* — 686*

 

Note. * p < .05. Degrees of freedom for multivariate analysis of variance 1, 279 for both

maternal and paternal ratings; degrees of freedom for univariate F test were 2, 278.

89

Fig
Pat:

511a

 

El Attention span
6 — 5-73 IApproach/withdrawal
5 7;; ll] Reactivity

  

 

 

 

Smoking by age 14 Non-smoker Smoking onset
unknown
Figure 23.
Maternal ratings of child temperament and smoking onset among sons. Means that do

not share letters differ at p < .05 using the Bonferroni method.

 

5_17 El Attention span
3 Approach/withdrawal
llIl Reactivity 4.23
3.74

  

 

 

Smoking by age 14 Non-smoker Smoking onset
unknown

Figure 24.
Paternal ratings of child temperament and smoking onset among sons. Means that do not

share letters differ at p < .05 using the Bonferroni method.

90

Ma

1101

M

“0

 

 

 

 

 

3.60 A
3-5 B .130 a 3.40 A
.0 3.0 .
Q)
3 A
o 2.5 . 2.47
’51. A 2.04
2 0
g A 2.00
.2 1.5 4
g 1 0 +Smoking by age 14
' +Non-smoker
0.5 + Smoking onset unknown
0.0 r
Ages 3-5 Ages 12-14

Figure 25.
Maternal ratings of Anxious/depressed and smoking onset among sons. Means that do

not share letters differ at p < .05 using the Bonferroni method.

 

 

 

 

 

4.5
4.0 4 3 70 444.13 B
B k 3.83 AB
3.5 .
U)
AB 3.16
E, 3.0 4
'° A
g 2.5 « A 2.56 2'66
E 2.0 «
’5'
g 1'5 i +Smoking by age 14
1'0 ‘ +Non-smoker
o, 5 . + Smoking onset unknown
0.0 T
Ages 3-5 Ages 1214
Figure 26.

Maternal ratings of Attention problems and smoking onset among sons. Means that do

not share letters differ at p < .05 using the Bonferroni method.

91

3.5

 

 

 

 

 

3.21 B
3'0 ‘ 2.70
h B
,9 2.5 A
is
.1:
.8 2-0 1 A 1.83
E >\: 1.74 A
5:" 1‘5 1 A 1.52 1.35 A
.E
a 1-0 ‘ +Smoking by age 14
O 5 + Non-smoker
' + Smoking onset unknown
0.0 1
Ages 3-5 Ages 12-14
Figure 27.

Maternal ratings of Delinquent behavior and smoking onset among sons. Means that do

not share letters differ at p_ < .05 using the Bonferroni method.

 

 

 

 

14
12 J B 12.16
0.. A 10.00
.2 10 - B
E: A 9.79 9'68
.1:
a 8 ~
g 7.24 A
a 6 4 6.41 A
31.-’6
20 4 4 +Smoking by age 14
+Non-smoker
2 1 + Smoking onset unknown
0 1
Ages 3-5 Ages 12-14
Figure 28.

Maternal ratings of Aggressive behavior and smoking onset among sons. Means that do

not share letters differ at Q < .05 using the Bonferroni method.

92

3.0

 

 

 

 

 

2.84 B
2.5 -
2.13
.§ A
> 4
,2 2.0 A ‘37
_g A fl 1.72 A
g 1.35 A
o1
'5 l 0 3
g ’ +Smoking by age 14
0 5 3 +Non-smoker
'i— Smoking onset unknown
0.0 ,
Ages 3-5 Ages 12‘14

Figure 29.
Paternal ratings of Delinquent behavior and smoking onset among sons. Means that do

not share letters differ at p < .05 using the Bonferroni method.

 

 

 

 

12
A 1042 10.67
10 . A '
15 A 9.20 B
.... 9.37
is 8 1 7.87 A3
E
E 6 3 6.38 A
§
% 4 . +Smoking by age 14
< +Non-smoker
2 —4
+ Smoking onset unknown
0 r
Ages 3-5 Ages 12-14

Figure 30.
Paternal ratings of Aggressive behavior and smoking onset among sons. Means that do

not share letters differ at p < .05 using the Bonferroni method.

93

obscn
resul‘
adol:
pane
ciga‘
511111
of d
ado
me;
€211
101

of

PATHS TO EARLY SMOKING ONSET

The current study so far demonstrated that early smoking onset was robustly
observed among adolescent boys in families where _b_oth parents are smokers, with at least
one parent is a Heavy smoker for their entire adulthood (see Figure 10). In addition,
results showed that parental alcoholism also played a role in early smoking onset among
adolescent sons (see Figure 21). When both parents were alcoholics with a lifetime
pattern of dependence, adolescent sons were more likely to experiment with or smoke
cigarettes. The current study also showed that the smoking subtype and alcoholism
subtype were associated with one another, and that these subtypes were related to a host
of demographic and psychopathological measures. In addition, findings showed that
adolescents with early smoking onset could be distinguished from others, based on the
measures of prenatal exposure to cigarette smoking, child temperament, and behaviors as
early as three to five as well as concurrent adolescent behaviors at ages twelve to
fourteen. However, due to the co-existing nature of risk factors involved in the process
of growing up in families with smoking parents, it is desirable that the unique role of
each of the risk factors related to smoking onset in children be investigated
simultaneously.

Manifest variable structural equation modeling (SEM) analysis was conducted.
Based on the previous results of the current study, the following key nine variables were
selectively chosen: Maternal smoking subtype, paternal smoking subtype, parental
alcoholism composite score, prenatal exposure to daily maternal cigarette smoking, four
syndrome scale scores of Anxious/depressed, Attention problems, Delinquent behavior,

and Aggressive behavior rated at ages three to five, and smoking onset status. The

94

1me
the 011
Gene

13.1 ge

01h
.‘Xlt
PT:

111;

variables were carefully selected in number and kind due to the overall sample size and
the ordinal scale level of the variables. Asymptotic covariance matrix with the
Generalized Weighted Least-Squares (WLS) estimation method normally requires very
large samples for reasonably robust estimation (J oreskog & Sbrbom, 1996, pp. 21-23).

Maternal and paternal smoking subtypes were coded on a four-point ordinal scale:
0 = Non-smoker, l = Light smoker, 2 = Heavy-to-light smoker, and 3 = Heavy smoker.
Parental alcoholism composite score was created by adding maternal and paternal
alcoholism subtypes, yielding a five-point ordinal scale (0 = Neither parent alcoholic, 1 =
Only one alcoholic parent with Alcoholism I, 2 = One parent with Alcoholism II and the
other non-alcoholic parent, or both parents with Alcoholism I, 3 = One parent with
Alcoholism II and the other with Alcoholism I, and 4 = Both parents with Alcoholism H).
Prenatal exposure to maternal daily cigarette smoking was originally measured in the
number of cigarettes smoked per day but recoded on a six-point ordinal scale: 0 = None,
1 = One to five cigarettes per day, 2 = Six to ten cigarettes per day, 3 = 11 to 19 cigarettes
per day, 4 = 20 cigarettes (or one pack) per day, and 5 = More than one pack per day.
The cutoff points for each category of prenatal exposure were decided based on the
smoking literature, and the frequencies of responses. Four CBCL behavioral syndrome
scales rated by mother were recoded on a three-point ordinal scale: 0 = Lower quartile
(bottom 25 percent), 1 = Middle 50 percent, and 2 = Upper quartile (top 25 percent).
Smoking onset status was coded as follows: 0 = Never smoked, 1 = Smoking-onset
unknown, and 2 = Smoked by age 14.

Due to the ordinal level scale of variables, a polychoric correlation matrix and an

asymptotic covariance matrix were created for each of ten multiply imputed datasets by

95

PREL]
pohtl
mere:
1996‘

M111

Scha

eSUr

“in
the

V211

su
pr
Sn
th

e)

b1

It

PRELIS using LISREL, version 8.51 (Joreskog & Sorbom, 2001). Once all ten sets of
polychoric correlation matrices and asymptotic covariance matrices were calculated, they
were analyzed using LISREL with the WLS estimation method (J oreskog & Sorbom,
1996). A polychoric correlation matrix using all ten imputed datasets is presented in

Table 12. Statistical inferences on parameters followed Rubin’s suggestion (1987;

Schafer, 1997, pp. 108-110). The overall estimate (a ) is the average of the individual

 

estimates (Q). The overall standard error (J7: ) is Jl—f +(1 + -1—J* B ,
m

where within-imputed variance ((7 ) is the mean variance (i.e., squared standard error in

the current situation) of _rr_1 parameter estimates from m datasets, and the between-imputed

variance (B) is the variance of imputed estimates (Q) from the overall estimate (5 ).

* _
The overall degrees of freedom is calculated by df = (m — 1)* 1+ —-£—U—- .
(m + 1) * B

It was hypothesized that there are direct links from maternal and paternal smoking
subtypes to early smoking onset in adolescent sons. In addition, parental alcoholism and
prenatal exposure to maternal daily smoking were hypothesized to directly link to early
smoking onset. The most interesting part of the hypothesized structural relationships was
the indirect path from maternal smoking subtype to early smoking onset via prenatal
exposure to maternal daily cigarette smoking and early child behavioral characteristics
(i.e., Anxious/Depressed, Attention problems, Delinquent behavior, and Aggressive
behavior). Four early child behavioral syndrome scales were allowed to covary, and
three exogenous variables (parental smoking subtypes and parental alcoholism) were

automatically set to correlate with each other.

96

direc

abox

Whe

ex;

011:

011

BI

The hypothesized model addresses several research questions: 1) whether the
direct links from parental smoking to early smoking onset in offspring stays significant
above and beyond other direct and indirect paths to early smoking onset in offspring; 2)
whether parental alcoholism plays an important role in leading children toward paths to
early smoking onset above and beyond other more specific paths; 3) whether prenatal
exposure to maternal daily smoking plays a direct and immediate role in early smoking
onset; and finally 4) whether the indirect path from maternal smoking to early smoking
onset via prenatal exposure and early child behavioral characteristics is significant above
and beyond the effects of other direct mechanisms of early smoking onset.

Reports of the analyses followed the guidelines suggested in the literature (Hoyle
& Panter, 1995; Raykov, Tomer, & Nesselroade, 1991). The hypothesized model fit the
data very well, 32 (20) = 14.664 — 18.213, g; Root Mean Square Error of Approximation
(RMSEA) = 0.000; Standardized Root Mean Square Residual (SRMR) = 0.038 — 0.042;
Goodness of Fit Index (GFI) = 0.996; Comparative Fit Index (CPI) = 1.000. Goodness-
of-fit indices and residuals were all within the acceptable range. Modification indices of
un-estimated path coefficients were very small, ranging from .00 to 6.31. Only one
modification index for the path coefficient fiom Aggressive behavior to prenatal
exposure to maternal daily smoking was slightly bigger than the critical value of 3.84
(expected amount of 33 change significant for one degree of freedom) in seven of ten
analyses. The hypothesized model was not modified due to the improbable nature of the
direction of the un-estimated path. Overall, excellent goodness of fit statistics, and small

modification indices and residuals suggest that the hypothesized mechanisms of early

97

smoki

relatic

$13311
into

mat:

int

01‘

smoking onset were all in the expected directions and that there were no other substantial
relations left out in the hypothesized model (see Figure 31).

All direct links from exogenous factors to early smoking onset were not
significant. Neither parental smoking subtypes nor parental alcoholism were directly
involved in intergenerational transmission of smoking. In addition, prenatal exposure to
maternal daily cigarette smoking was not directly related to early smoking onset in
adolescent sons. However, there was an indirect link between maternal smoking subtype
and early smoking onset via prenatal exposure and observed child behavior of
Anxious/Depressed at ages three to five.

The results suggest that when all things were considered simultaneously, it was an
indirect path via early child behaviors of Anxious/depressed that led to early smoking
onset in adolescent sons. Although prenatal exposure to maternal daily cigarette smoking
was related to all four domains of child behavior problems at child age three to five, it
was only via the Anxious/depressed route that led to early smoking onset in adolescent
sons. Young children as early as three to five who were rated high by their mother on the
CBCL items such as “Lonely,” “Cries,” “Unloved,” “Fearful,” and “Worries” were more
likely to engage in early smoking experimentation and smoking. Results suggest that the
link between parental smoking and early smoking onset in offspring can be accounted for
by mediating factors of prenatal exposure to cigarette smoking and negative affect as
early as age three to five. In addition, paternal smoking or parental alcoholism do not
appear to have true associations with early smoking onset in offspring. Their ties to early
smoking onset in offspring in the first two sections of Results can be mostly attributed to

their shared variance with maternal smoking type.

98

5.. ~ I.“ .4..th

 

.3383 83%: :2 =m mama: 8398 33 $38 $5. .802

 

 

83 mm; 33 SS :2 Ed :3 22 :3 626 manoegam .6
83 was a? 22 Ed ES 53 RS 8328 6382»? .m
83 $2 82 ammo 23 $2 83 832838325 .5
83 $2 a; :2 ES SS 866368 882:... .6
83 2.3 ago 62.0 83 882669885. .m
83 8.3 23 ӣ6 8:82 265836 3265 .4
83 :2 :23 86:28; 385 .m
83 god 6.5%... mice... 388$ .N
coo; cabana wee—08m 38032 A
a w a o m 4 m N a

 

$.52 cote—obou otonobom 3385mm

2 2an

99

7—

.. _ $3.... <

 

 

 

 

 

31263

 

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

no “ohm onus—um :32 “com ”a 6.2.23 1 v8.3 n 88 AIM .339 oumtflm m 08 momofizoaq E Eon—:52 .mfima “5&8me

 

 

.83 1 EB sees E 86 86580 ad 1 2&25 8:25.63
-mo: 88:85 8:: 8on .8362: 3.8323 a we commemogmaomg ”mat—08m mo :ommflfimaab _acoufiouowuoafi mo _ocoE <
.3 65mm
3 ow< >m A 8NCE. r
28.5 1 so.-. 5.- ..
wEonm A .. Emzonooz
‘.... ...... ...... .3... ..... —NH:DHNL
fl. 4.. 80¢
. we: SSEom .- .-
.. 0338mm? / Ca V m: m
80.: Z. wombnsm
600 mm. :EoEm
. . 8320mm . _mEBmm
Eoscczoo
9&8 cm. 25032
was oh 9 22895 A .. onbnsm
E m .8255 $o.m© a. mac—08m
cos—~82 €58 S. 1 35082
$5.8 VN.
vommocmoa
\wzomxé
3:83 Ea mac—08m
m 1 m mow< 325$

2:30; RENEE

palm
smo‘:
our

farm

per

the

PO}

SN
Ca

1'6

DISCUSSION

The present study examined initiation of cigarette use among adolescents from a
pathological perspective where individual sensitivity and vulnerability to cigarette
smoking is emphasized as a mechanism of early smoking onset. This perspective leads
our attention to early observable factors with some constitutional basis in children and
familial vulnerabilities. These emphases sharply contrast to a widely popular epidemic or
exposure model of adolescent smoking where exposure to smoking by peers and family
members are construed as a key factor in the context of modeling and pressure. The
perspective of individual sensitivity (and vulnerability) to smoking as a key is not new in
the literature of nicotine dependence and intervention for cessation of smoking in adult
populations. However, it has not been a focal point in the studies of smoking onset and
in preventive efforts to control smoking initiation.

In line with individual differences in sensitivity to cigarette smoking, the present
study examined whether the phenomenon of intergenerational transmission of smoking
can reliably be observed in the manifestation of early smoking onset. Although some
researchers suggested that habitual smoking runs in families (e. g., Bierut et al., 2000;
Prescott & Kendler, 1995), parental smoking as a risk factor for adolescent smoking
initiation has been relatively underexplored. To fill the gap, the present study
investigated smoking onset among adolescents from four major angles. While
intergenerational transmission of smoking can be attributed to numerous factors, the
present study limited its focus to early risk factors in children and parents. Each of four
major results sections provides insight to the phenomenon of early onset of smoking

among adolescents. Findings of the current study are recaptured and discussed in here in

101

that
2m

fina'

cha
init
smt

OllE

par
Oni

Sm

far

f0]

pr:

Pa
(t.

0f

the order of 1) pathways to early smoking onset in a high-risk population of adolescents,
2) heterogeneous developmental patterns of smoking and alcoholism among parents, and
finally 3) limitations and future directions.

Pathways to Early Smoking Onset

Parental smoking, Results revealed that Heavy smoking by either parent,
characterized by a long-term, high-level smoking was related to children’s smoking
initiation, when either paternal or maternal smoking was studied alone. However, when
smoking by two parents was simultaneously considered in their association with smoking
onset in offspring, it was the presence of both smoking parents that was related to an
increased likelihood for children to start cigarette use early. Heavy smoking by either
parent alone was not sufficiently related to an increased likelihood of early smoking
onset. It was the combination of heavy smoking by fathers, with either light or heavy
smoking by mothers that was associated with an increased risk of early onset of smoking
in children.

The results suggest that parental smoking be studied on both parents in two-parent
families. In the presence of heavy smoking by one parent, smoking by the other parent
becomes a vital factor for children’s smoking onset. Although it is uncertain how the risk
for early smoking onset increases as both parents in a two-parent family smoke, the
presence of a non-smoking parent appears to ameliorate and counter-balance the
facilitators of early smoking onset in children, whereas the presence of two smoking
parent exacerbates them. In addition, the results point to th_ep§ttern of cigarjette smoking
(i.e., quantity and duration of smoking) by parents as an important factor for early onset

of smoking in offspring. The current study hints that a long-term habitual heavy smoking

102

(or ni
susce
risk l

smo
repr
and
smc
attr

Srn<

fiJr
Stu
eat
ap‘
bu
re]

Pa

Sit

(or nicotine dependence) by parents may be a significant marker for an underlying
susceptibility to cigarette smoking in children, and also a potent marker for an array of
risk factors related to growing up in families with a higher level of parental
psychopathologies and a lower level of socioeconomic status.

Parental alcoholism. Among many possibilities, one pathway to early onset of

 

smoking investigated in the present study was via a common familial vulnerability
represented by parental alcoholism. Given the high co-occurrences of habitual smoking
and alcoholism, it is natural to speculate that the intergenerational transmission of
smoking may potentially be marked by a parental alcoholism diagnosis that can be
attributed in part to the common correlates of a positive family alcoholism and habitual
smoking.

The results lend support to the studies that found a relationship between parental
alcoholism and early smoking initiation in offspring (i.e., Hanna & Grant, 1999), and
further provide some insight on the reported relationship between the two. The current
study particularly points to the specific nature of parental alcoholism as a risk factor for
early smoking onset in offspring. A long-term dependence on alcohol by both parents
appears to link to an increased likelihood of early smoking onset in offspring. Paternal
but not maternal alcoholism subtype, when crossed with adolescent smoking onset, was
related to an increased the likelihood of children’s early engagement in cigarette use, with
parental Alcoholism II (a long-term dependence on alcohol) closely tied to early child
smoking onset. However, when alcoholism subtypes by both parents were

simultaneously crossed with children’s smoking onset, it was the presence of both

103

fa

alcoholic parents with Alcoholism H that were associated with early onset of smoking in
offspring.

The results for alcoholism were parallel to those of parental smoking as a risk
factor for early onset of smoking in the sense that 1) both results revealed that in two-
parent families, it involved two parents to increased a chance for their child to start
cigarette use at an earlier age, 2) out of the two subtypes of alcoholism only Alcoholism
II, representing a more dependent type of alcoholism was related to an increased
likelihood of early onset of smoking in offspring, 3) Alcoholism II was related to a higher
level of psychopathologies and a lower level of socioeconomic status. Furthermore,
Heavy smoking and Alcoholism II were related to one another. The strong relationship
between smoking types and alcoholism types and the compatibility between parental
smoking and alcoholism in their associations with early smoking onset in offspring
prompt many more questions than answers.

Among many others, we can ask what it is about Heavy smoking and/or
Alcoholism II by parents that might set children at risk for early onset of smoking? And
what are possible synergistic influences on children’ smoking onset when parents smoke
and drink so heavily for the majority of their adulthood during while their children move
into adolescence. In addition, how does the risk for children’s early smoking onset step
over the threshold as both parents exhibit a dependent type of smoking and alcohol use?
It is plausible that heavy smoking and Alcoholism II by both parents reflect the increased
likelihood that the child has some constitutional susceptibility to cigarette smoking
among other problems, with one possible source resulting from prenatal and postnatal

exposure to cigarette smoking. At the same time, co—existing psychopathologies and lack

104

atte

ear

of resources in families may play indirect roles in promoting early smoking onset in
offspring via numerous pathways, including parenting behaviors, coercive family
environment, and association with deviant peers. The current study focused on early
observable characteristics of children as antecedents of early smoking onset in
adolescence.

Earlxcharacteristics of adolescents with early smoking onset. One can shift the
attention to children and ask whether there were early identifiable factors in children with
early smoking onset. Results revealed that adolescents could be traced back to their early
development for their vulnerability to early cigarette use. Adolescents with early
cigarette use were different, on several characteristics typically considered as early risk
factors for many later developmental outcomes. Adolescents with early onset of smoking
had more prenatal exposure to maternal daily cigarette smoking. In addition, their
mother viewed them as more reactive, and the father viewed them as more approaching
to new stimuli and people as early as three to five. In addition, adolescents who never
smoked by age 14 were observed by their mother as less distractible at ages three to five.
Behaviorally, they were rated high at ages three to five by mothers on the CBCL scales of
Anxious/Depressed, Attention problems, and Delinquent and Aggressive behaviors.
These behavioral characteristics were again confirmed at ages 12 to 14, with the
exception of Anxious/Depressed. Paternal ratings produced slightly different results,
showing adolescent sons with smoking onset rated high on Delinquent and Aggressive
behavior at ages 12 to 14.

Results lend support to the recent findings of the literature of child behavioral

characteristics and smoking. There are burgeoning debates on 1) which aspects of

105

ADH
and l
and ‘
et al
issu
h}?

It“

rel
aft

W

ADHD (e.g., inattention versus hyperactivity) are related to smoking (Burke et al., 2001),
and 2) whether ADHD is truly related to smoking, regardless of conduct disorder (CD)
and vice versa among various populations of adolescents (Milberger et al., 1997; Whalen
et al., 2002). Although the current study is neither designed nor equipped for these
issues, it supplies evidence that these co-existing conditions of attention problems,
hyperactivity, and CD-related behaviors are manifested and observed as early as three to
five years of age among children with early smoking onset.

In addition, results indicate the importance of Anxious/Depressed behavior in
relation to early smoking onset. The relationship between smoking and negative
affect/depressive symptoms is well established in adult populations of normative (e. g., M.
Windle & R. C. Windle, 2001) as well as clinical samples (e.g., Pomerleau et al., 1997;
Shiffman et al., 1994a). However, Anxious/Depressed behavior has not been studied in a
sample of young children in relation to smoking. The current study highlights that early
negative affect measured using questions such as “Lonely,” “Cries,” “Fearful,” and
“Worries” can meaningfully be observed in children as early as three to five, and that it
appears to be a significant early risk factor for early smoking onset.

The current study also showed that maternal and paternal ratings had low
convergence on child temperament and behavioral problems, yielding distinct results
based on maternal and paternal ratings of child characteristics. Correlations between
maternal and paternal ratings on child temperament ranged in average from .269 to .373,
and from .198 to .384 for four behavioral syndrome scales at ages three to five. Parental
agreement on children’s behavior appeared improved at child’s ages 12 to 14, with

correlations ranging from .250 to .527. Low to moderate agreement between maternal

106

SOL

PSI

R.

si‘

SI“.
1h
9X
th.
alc

eat

and paternal ratings is nothing new. Substantially different findings, depending on the
source of information (e. g., ratings of self, parents, peers, or teachers), have been noted in
psychiatric and family research (O’Connor & Rutter, 1996). Multiple sources of
information have been suggested as a strategy to obtain a robust solution that can reliably
be generalized, especially when concurrent relations are of interest in a study (O'Connor
& Rutter, 1996; Rothbart & Bates, 1998). In the current study, however, early child
temperament and early behavioral characteristics temporally preceded smoking onset in
young adolescence, therefore minimizing possible reporting biases that one’s perception
of problematic behavior clouds judgment of other behaviors. In the current study,
maternal observations and perceptions of child temperament and behavioral problems
were overall a better indicator than paternal ratings for later smoking onset in children.
Results were consonant with a summarized report that maternal observations are, in most
situations, relevant and reliable (see Rothbart & Bates, 1998).

m to early smokinggaet, The primary purpose of the manifest variable SEM
analysis was to provide some clues to the mechanisms of intergenerational transmission
of smoking at the stage of smoking onset in offspring. It was hypothesized that parental
smoking and alcoholism are risk factors for early smoking onset in children, not because
they have direct influences on it, but because they reflect indirect paths via prenatal
exposure to maternal cigarette smoking and early child behavioral characteristics. With
the exception of maternal smoking subtype, neither paternal smoking type nor parental
alcoholism nor even prenatal exposure to smoking was a sufficient condition to cultivate

early smoking onset in children.

107

Maternal smoking, however, appeared to be a causal spark in a series of chains
leading up to early smoking onset via mother’s continued smoking during pregnancy.
Prenatal exposure to maternal cigarette smoking was then related to anxious/depressed
behavior (negative affect), attention problems, aggressive and delinquent behaviors at
child ages three to five. Of these negative affect predicted early smoking onset many
years later. Paternal smoking subtype and parental alcoholism were remotely related to
early smoking onset only due to the fact that they shared variance with maternal smoking
subtypes. Therefore, roles of paternal smoking and parental alcoholism appear to be
spurious in nature in relation to early smoking onset in offspring. The existence of heavy
smoking and Alcoholism H by both parents in two-parent families may be suggestive of
perilous undercurrents of heavy smoking by mothers and accordingly exposure to a
higher level of cigarette smoking during prenatal development.

The origin of individual vulnerability to cigarette smoking may be rooted in a
long-term heavy smoking by mother that is directly and closely related to continued
smoking during pregnancy. Indeed, maternal smoking type and prenatal exposure to
maternal smoking were almost inseparable constructs, as indicated by the path coefficient
of .91 in Figure 31 and also by the correlation coefficient of .888 in Table 12. And the
exposure to maternal cigarette smoking during prenatal development may have altered, to
some extent, the function and structure of the brain of the child (Wakschlag et al., 1997;
Weitzman et al., 1992). Although we do not know the full details about prenatal
exposure to cigarette smoking, it appears that it certainly leads to an increased level of
behavior problems (Anxious/Depressed, Attention problems, Delinquent and Aggressive

behaviors) in young children as early as three to five. Among these early child

108

behavioral characteristics, at least one domain, Anxious/Depressed appears to be directly
associated with early smoking onset in adolescent offspring. Anxious/Depressed
syndrome or negative affect in a more general term has not been studied in the literature
in relation to prenatal smoking exposure nor to smoking onset. The current study
provided evidence that negative affect shown by young children is an important mediator
of early smoking onset.

There are several important implications to these findings with regard to
preventive efforts to control cigarette initiation among adolescents. The report of the
Surgeon General in 1994 concluded that school-based programs, coupled with youth-
oriented mass media campaigns and tobacco tax increase are effective measures to
prevent tobacco use among youth (USDHHS, 1994). Recently, the Task Force on
Community Preventive Services reconfirmed the previous conclusions and recommended
that increase the unit price of tobacco products and long-term, high-intensity mass media
campaigns are an effective deterrence to smoking initiation among youth (CDC, 2000).
Findings of the current study suggest that prevention programs targeted for special
populations at risk for early smoking onset may also be effective. Adolescents with
increased susceptibility to cigarette smoking may benefit more with preventive and
intervention programs tailored uniquely for them. Findings of the current study showed
that adolescents with early smoking onset are different from others from very early on,
tracing back to as early as their prenatal development followed by differences in
temperament and behavioral characteristics at ages three to five. Therefore, children’s
susceptibility to cigarette smoking may be identified much earlier before their first

cigarette.

109

Heteflgeneous Developmental Patterns of Smokingand Alcoholism

Although it was not the focal point of the current study, a pathological model of
smoking could also be examined in a high-risk population of alcoholic and control
parents. This served two purposes. It provided parental information in connection with
adolescent children’s smoking initiation, and a parent equivalent portrait of vulnerability
to smoking. The former has been discussed; the latter, the second focus of the current
study is discussed in this section. The investigation of parental smoking and alcoholism
in the present study was unique in many aspects, including utilization of population-
based, prospective longitudinal data and new analytical techniques. Its implications are
discussed as follows.

Smoking types. The current study asked whether the findings of two major types
of smokers from Chassin and her colleagues (2000) could be extended into the age period
that goes beyond age 31. Results supported the notion of the heterogeneity of smokers,
and confirmed the general pattern of decline in smoking prevalence (Anthony &
Bcheagaray-Wagner, 2000). However, the number and the kind of distinctive smoking
patterns over time proved to be convergent more with the literature on smoking by
habitual smokers in adult clinical populations, including Shiffman and his colleagues
(1994a, 1994b) than with findings of Chassin et al. (Early Stable, Late Stable,
Experimenter, and Quitter). Three distinctively different types of smokers (i.e., Heavy
smokers, Light smokers, and Heavy-to-Light smokers) were identified in the present
study, based on their long-term patterns of cigarette smoking during adulthood ranging

from ages 24 to 50.

110

attrib
bent

OllCE

thei:

“a
We

de'

)‘C

0C

Cl

1C

(4"

Differences in results between the current study and Chassin et al. (2000) can be
attributed to the differences in measurement of smoking, the age period, and participants
between two studies. First, in Chassin et al’s study smoking was measured in a way that
once participants smoke weekly, they were differentiated into three categories based on
their quantity of smoking: 10 or fewer cigarettes per day, 11 — 20 cigarettes a day, and 20
or more cigarettes a day. Therefore, it is plausible and acknowledged by authors that
Light smokers and/or chippers may have been included in both Late and Early stable
smokers in Chassin et al’s study (2000). In the present study, on the other hand, smoking
was based on a finer quantity measure of seven categories that captured light smoking as
well as heavy smoking. Therefore, the current study was able to show the distinctive
developmental pattern for Light smokers.

Second, in Chassin et al’s study, smoking was measured from adolescence to
young adulthood (up to age 31) where smoking initiation and experimentation most often
occur, with frequent changes in both smoking status and quantity of smoking. Therefore,
a lot more fluctuations in smoking can be expected and consequently captured in Chassin
et al’s study. However, the age period covered in the present study ranged from ages 24
to 50, well past the time of smoking initiation and experimentation. Three types of
smokers found in the present study, accordingly, reflect stabilized, long-term habitual
smoking in adulthood. The patterns of stabilized habitual smoking in mid 203 to 50 in
the present study generally match with findings of epidemiological studies of nation-
wide, representative, cross-sectional data of all ages on tobacco use and dependence

(Anthony & Echeagaray-Wagner, 2000).

111

If}

h!

Third, participants in the present study were population-based alcoholic and non-
alcoholic men and their spouse, whereas Chassin et al (2000) featured population-based
adolescents. Given a higher rate of smoking among alcoholics, it is more likely that
smokers were over-represented in the present sample than in populations at large. In fact,
47.2% of smokers in the present sample were substantially more than 39.1% of smokers
(including quitters and experimenters) in Chassin et al’s study. In addition to a higher
number of smokers among alcoholics, alcoholics tend to smoke more heavily. Therefore,
results from the current study may be over representative of heavy smokers than in
populations at large.

However, findings of the current study hold up the notion that there are DE M

of smokers. The majority belongs to a group of long-term regularaheavy smokers who

 

are chronically dependent on nicotine, and the second group consists of a small

proportion of long-term regular but light smokers without nicotine dependence (i.e.,

 

chippers or light smokers) (O. F. Pomerleau et al., 1993). A much smaller number of
smokers are suggested to be in transition from heavy smoking to reduced light smoking
(e. g., converted chippers) or even to smoking cessation. Findings from the current study
appear to fit into these three categories of smokers described in the literature in terms of
prevalence and patterns of smoking, with Heavy smokers equivalent of “regular and
heavy” smokers, Light smokers of “regular but light” smokers or chippers, and Heavy-to-
light smokers of converted chippers.

The notion of two major types of smokers was empirically supported in the
present study for the first time, to the author’s knowledge, from population-based, long-

term prospective longitudinal data. New revelations made by the current study were that

112

firs
tth
the
ant
rel;
me
me
bot
nu:

lht‘

lit:
C l.
‘31
P0
CU]
9X]

011

YES

Lin

first, there were substantially more smokers (22. 1%) in the category of Light smokers
than previously thought (an estimated 5% of smokers); second, although non-significant,
the types of smokers showed systematic differences on measures of psychopathologies
and socioeconomic status in a dose-response pattern; third, the specificity of the
relationship between smoking types and other characteristics varied across gender, with
more differences found between Heavy and non-Heavy smokers among men, whereas
more differences found between Non-smokers and smokers among women; fourth, in
both men and women Heavy smokers were statistically different from all others in the
number of years of education and conduct problems and antisocial behavior. Findings by
the current study should prompt initiatives to examine typology of smokers and its related
antecedents, concurrent relations, and health outcomes in future studies.

Alcoholism types. There have been many theoretical and empirical studies in the
literature that point out the heterogeneous nature of alcoholism 63abor et al., 1992;
Cloninger, 1987; Zucker, 1987; Zucker, Fitzgerald, & Moses, 1995). However,
typologies of alcoholism were rarely derived from an empirical study based on
population-based long-term prospective longitudinal data. Empirical results fi'om the
current study appear convergent with the literature on alcoholism although more
extensive studies are needed to calibrate the nature of these two alcoholic groups. Based
on findings of the current study, Alcoholism H appears to overlap, to certain extent, with
Antisocial alcoholism, Type H, or Type B in the literature. However, results were limited
by the modest sample size of alcoholics in the present study, and the generalization of the
results beyond the homogeneous population featured in this study should be cautioned.

Limitations of the current study are later discussed in more detail.

113

me
am
1m
lor
M

be

“‘2

M

nil

l6

di

(1

As expected, types of smoking and alcoholism were highly associated in both
men and women. Results support the long-held notion using a different analytical
approach and perspective that smoking and drinking are related. Non-smokers were very
unlikely to have a long-term severe type of alcoholism (i.e., Alcoholism H). Likewise,
long-term Heavy smokers were very unlikely to be non-alcoholics but more likely to
have Alcoholism H. In addition, there were gender-specific patterns of association
between alcoholism and smoking. While it was unlikely that heavy smokers were non-
alcoholics, those cases were found more often in women than men. Similarly, while it
was unlikely that non-smokers were those who showed Alcoholism H, more men than
women belonged to the category.

Although the co-occurrence of smoking and drinking or the co-morbidity of
nicotine dependence and alcohol dependence were outside the focus of the present study,
results nonetheless provided interesting issues for future studies. First, there may be
different mechanisms underlying the link between smoking and drinking for men and
women. The present study showed that there are patterns of gender-specific as well as
general in the way that smoking is related to alcoholism. Second, there may be some
shared psychosocial characteristics unmeasured in the present study. The current study
used only a limited number of variables to test whether empirically driven types of
smoking and alcoholism could be distinctively discriminated. However, it is highly
plausible that people with dependence on nicotine and alcohol may share some other
psychosocial and personality characteristics, including impulsivity and sensation seeking

(Little, 2000). Similarly, nicotine and alcohol may share some neuropharrnacological

114

sn
dr

int

513
thi:
chi
uni
ahe

mve

First,
tralec
allaiy:

the pl]

aICQhol :

”mmmi

properties as expressed in hypotheses of cross—tolerance and cross-reinforcementl7 (O. F.
Pomerleau, 1995). More studies need to follow from all directions to reveal the
mechanisms of co-occurrence of smoking and drinking/alcoholism.

Quantitative differences versus qualitative typology. In the current study, parental
smoking and alcoholism were utilized as typologies as opposed to quantitative
differences over time. Subtypes of smoking and alcoholism were examined as an
indicator for an unmeasured risk structure typified as a lower level of socioeconomic
status and resources, and a higher level of antisociality and depression. An alternative to
this approach is to study fluctuations of parental functions across time as a predictor for
child outcomes. It is natural to assume that familial and child outcomes may vary over
time as parental functions fluctuate (DeLucia, Belz, & Chassin, 2001). Although this
alternative approach did not turn out as expected in DeLucia et al’s study, it merits future
investigations on both approaches to parental functions over time as predictors of child
developmental outcomes.

Limitation_s_ and Future Direction_s

The findings of the present study should be seen in light of its several limitations.
First, findings of the present study regarding the natural developmental patterns (or
trajectories) of smoking and alcoholism need to be replicated using a full longitudinal
analysis that matches the same age period and descriptions of the population analyzed in

the present study. The current study utilized an available four-wave longitudinal data to

 

17 Cross-tolerance refers to the possibility that nicotine increases tolerance to the aversive effects of
alcohol and vice versa, whereas cross-reinforcement refers to the possibility that nicotine increases

reinforcing effects of alcohol and vice versa.

115

its '

alc

be

QB

(in

its maximum efficiency, with possible compromise on robustness of the data. Parental
alcohol use and smoking were assessed up to four times with a three-year interval
between measurements in this study. Information on parental smoking and alcoholism of
each individual were then placed and overlapped with others over age, creating
developmental patterns over a much wider age span. This method produced incomplete
data with gaps in between ages per person that pertained to a relatively smaller age period
per participant.

The current study design can be construed as a blend between a full longitudinal
study and a cross-sectional study in that results were derived from four-wave longitudinal
data stretched over an age span at least twice as long. In addition, analysis on parental
alcoholism utilized a retrospective report on the age of first alcoholism diagnosis,
expanding the time period up to 40 and 35 for fathers and mothers, respectively. One of
the drawbacks was that in both extreme ends of the age period, there were only a few
observations made, resulting in relatively large standard errors. It especially influenced
analysis on women’s alcoholism. In an ideal situation where no limits are placed on time
and financial resource, a more controlled forty-year longitudinal study may suit better to
investigate developmental patterns of smoking and alcoholism. However, its merits are
also traded off with the hefty price tag of such a longitudinal study and other related
issues, including a substantial subject attrition rate. In the present study, the issue of
robustness and reliability of the data was weighed against the efficient use of the data.
Although the present study had a higher rate of unobserved data points across
chronological age, it was justified given the highly convergent nature of the phenomena

of our interests (i.e., smoking and alcoholism).

116

ax
dc
Sn
311

the

Second, one should exercise caution when generalizing the findings from this
racially and geographically homogeneous sample of the current study. Participants in the
present study were mostly Caucasian families residing in mid-Michigan area. In
addition, when they were initially recruited, families were intact. Although a number of
families were separated, divorced, and/or remarried, the UM-MSU Longitudinal Study
followed up almost all biological and stepparents who were separated from or added into
the families. In the present study, three stepparents met the criteria of the present study
(e. g., completion of at least two of the first four assessments). However, in all three cases
information on biological parents’ smoking and alcoholism was available and therefore
used in the present study. Consequently, parental smoking and alcoholism in the present
study refer to those of biological parents in two-parent families who reside with the
children in all but three cases. It is important for firture studies to investigate parental
smoking and alcoholism with adolescent children in families of different structures. The
roles of parental smoking and alcoholism may differ among children in single-parent
families or two-parent families with stepparents. In addition, it is important for future
studies to replicate the present findings with adolescent children and their parents fiom
other geographic locations, and from other racial, ethnic and cultural groups.

Third, there may be more factors that lead children of smoking parents to and
away from cigarette smoking that were not included in the current study. By no means
do results from the current study imply that the tested path model is the only way to early
smoking onset in a high-risk population of adolescents. On the contrary, the current
study aimed to show that some children are at increased risks for early smoking onset and

their susceptibility to cigarette smoking exists prior to children’s first cigarette smoking

117

from the perspective of the multiple pathways to substance abuse, with smoking included
(Chassin, Curran, Hussong, & Colder, 1996a; Zucker, 1994). Although the research
focus could not be extended due to sample size and the nature of the data, it is more
reasonable that there may be multiple routes to early onset of smoking via diverse
combinations of risk factors and mechanisms (i.e., equifinality; Cicchetti & Rogosch,
1996; Cicchetti & Toth, 1998). For example, it is very plausible that families with
parents of certain subtypes of smoking and alcoholism may foster smoking-friendly
environment in which children take on more favorable attitudes toward smoking, and
have an easy and early access to cigarettes. In addition, findings of the current study
point out that children in these families are exposed to a higher level of parental
psychopathologies and a lower level of socioeconomic resource. Future studies can
address whether there exist additional mediating pathways to early smoking onset in a
high-risk population of adolescents.

Likewise, maternal heavy smoking and prenatal exposure to cigarette smoking
may also lead to multiple outcomes, with early smoking onset being just one outcome of
many (i.e., multifinality; Cicchetti & Rogosch, 1996; Cicchetti & Toth, 1998). In
addition, parental smoking subtypes and alcoholism may serve as a proxy for other
developmental outcomes in children other than cigarette smoking. Considering that
Heavy smoking and Alcoholism H were associated with other known risk factors for
children’s less advantageous developmental outcomes, it is highly likely that children of
those parents may be at risk for other developmental outcomes, including early onset of
drinking, higher behavioral problems, and lower academic achievement. Children of

smoking parents have not been considered as a risk population in the literature. However

118

the current study provided evidence that they may be at increased risk for other aspects of
development as well. Further efforts are necessary to investigate the specificity as well

as aggregation of the risk factors identified in the current study in relation to a wide range
of developmental outcomes.

Fourth, data on daughters were not sufficient to ensure reliable findings for many
analyses conducted in the current study, due to the recruitment design of daughters.
Daughters were brought into the UM-MSU Longitudinal Study systematically later than
sons and it resulted in either their first participation starting at later assessment wave or
younger birth cohorts at the first measurement wave. Due to this design, assessments of
either the first wave or the 4th wave were not carried out, resulting in a higher rate of
missingness. Since many analyses of the current study required information collected
when the target child was three to five and twelve to fourteen years old, it was decided
that data on daughters data be exploratively used with caution. Although results from
daughters showed similar trends to those of sons in some analyses, they were not as
statistically reliable. Comprehensive research is needed in the future to clarify whether
there exists an equivalent mechanism of intergenerational transmission of smoking for

boys and girls.

119

APPENDICES

120

APPENDIX A

Table 1A

Number of Adolescents by Birth Cohort

 

 

 

 

Count
Cohort Son Daughter
1992 O 1
1991 0 l
1990 0 6
1989 0 12
1988 24 12
1987 50 1 l
1986 37 18
1985 57 7
1984 25 l 1
1983 31 5
1982 22 l
1981 20 3
1980 10 0
1979 5 0
Total 281 88

 

121

Table 2A

Number of Parents by Birth Cohort

 

 

 

 

 

Count Count

Cohort Father Mother Cohort Father Mother
1938 2 0 1955 21 18
1940 l O 1956 22 25
1941 1 O 1957 24 21
1943 l l 1958 23 26
1944 3 0 1959 20 33
1945 2 0 1960 16 19
1946 2 O 1961 19 18
1947 5 3 1962 13 23
1948 4 0 1963 6 22
1949 6 3 1964 10 9
1950 10 5 1965 3 6
1951 12 9 1966 l 4
1952 15 4 1967 0 1
1953 13 14 1970 0 l
1954 20 15 Total 275 280

 

122

 

 

m 33

fl 2 ow?

fl m cm 53

_ v o 3 mm?

v N. 3 mm $3

m o n m VN wwa

o 2 3 3 cm mwg

m a 3 N. hm cw?
w R 3 av $2
N. 2 mm wwa
Sow ooom 33 mag 33 83 mag .33 moi Nam“ $3 33 $3 wwa 53 m fiasco

 

uflDEmmOmmm MO .30?

 

 

meow J 963 “w Eoflwflommxx me We“; wmeo< tonoo Sam .3 533.5me

m Nun—Zambia.

fim 2an

123

 

m fl v 33

N m N a owa

m m c N 3 5a—

m o n v NN Nwa

h N. 3 m 3 $2

5 n m N cm go—

_ 3 3 3 a L mm mwofi
w 2 v mm 33

m a 2 on $2
_ 9 : mwa

 

SON oooN 33 woa 32 cog mag .33 32 mag 33 33 $3 wwg 32 £2 £2 a tozeu

uCDF—mmowmw MO 50>

 

 

mcom ”v o>mi>> Hm Eofimmomm< we flour $83. tonou 5am mo 533.5me

mm 2an

124

 

_ N m ~wm—

_ ~ Nwa

H _ _ m 33

~ ~ dwa—

~ ~ 39

m m 33

fl N m w 53

m m m a £3

1 n 2 $2

m M L m 092
_ y 33
~ ~ Nag

 

Sow ooow 83 mag .33 33 mag vae 33 Nae 33 82 $3 $3 53 £3 mwa m tonoU

 

ufiUEmemwm MO hfl0>

 

 

 

magma—mn— ; 02:5 3 EoEdmomm< mo .80». 383‘ tonoo 5am .«o aowsnwama

mm 023%

125

 

_ N m _wm~

fl _ Nwofl

v _ m mwmfi

N o m __ vwa~
m m n mwa~

m m v_ emo—
_ w swa—
v wwafi

o mwmfi

o oao~

o _aa~

o Naofi

 

flooN ooON aao~

wma~ noo~ ooafi mao_ vao_ mmofi Noa_ _aa~ ooa~ awa~ wwa~ nwm~ owafi mwa~ pu ”fiasco

 

HGDEmmommm MO umo>

 

 

 

825st a. 335 um EoEtwmomma‘ mo 36>..mim8o< fiasco :Bm we eager—“Elm.

cm 2an

126

 

$2.568 ofimfl

 

Nma g

33

ome

v-rNNN

'—

N
N
M
q.
'—

Nmmfi
_mm_
cma_
ava~
wvmfl
nvafi
away
mva_
veafi
mwmfi
_vm~
ovo~
wma~
tenoo

N
.—o
M

N
ON
II—Ilv—t

v—i
N—«HMMNF‘WVO

El

32 mwa swag £3 32

 

HGDEmmommm m0 30>

 

 

956$ ; 233 m «anamomflx 3.48% 883‘. tenou :Em we cos—Saba

mm 033.

127

 

L L ooaL

 

m m mooL
o N L L OL «.03
m L L L o momL
v v N L L L mL NcaL
m m N L w m wL LoaL
N L. v N L 0L 08L
m o N N L m N L ON one
N m m L N N L w NN wmaL
v m m m m N MN 33
o N w L N m m NN ommL
m Lu N o L m L LN mmmL
L m m w m L oL VmaL
L m m L N m L mmmL
NooL LooL ommL owoL wwaL nme owaL mme m fiasco

 

LCDEmmommm mo 30>

 

 

865an ”L o>m>> 31.58808... 3&6.» $804. toaou 5mm .3 :oLLspEmLQ

62.5888 8 2,5

128

 

605588 2%:

 

v-‘Mv—‘VM

0‘

mmaL
NmaL
meL
ommL
mme
wme
NVmL
ovoL
memL
vwaL
mme
wmaL

NNr—‘NF‘
Ov—
v-dv—‘

N—‘Nv-‘v-«Vm'srh

 

maaL

vmmL

moo L

tonoo

21

N09 LamL omaL

 

 

 

£05an ”N 0235 8 Language... .Lo 30> mmeo< tenoo 5mm .3 coaznsma

cm oLnLaH

129

 

 

.w.1..ac..‘:r~..\\ ”\NN (<4.an

 

 

L L comL

m m mooL

N m N a vcaL
v L L o momL

v m N m NL NomL

m m n L N wL LcmL
v n m VL ooaL
m m N m L vL amoL
N w L m N L 5L wmaL
N v o m L L oL hmaL
N v m w L m 5L omaL
N v m N L m mL meL
L m m m L ML VmaL
mooL vaoL mamL NomL LoaL ommL m 98:00

 

 

 

maniam ”N 02.35 a EoEmm0mm< .Lo 30> 000.8... fiasco LLth .3 :oLSnLmeLnL

902588 00 2.3

130

 

625350 2%:

 

L

N
N
N
'—

NmaL
LmaL
cmmL
ome
wme
hvaL
cvaL
mme
vme
mw¢L
LvaL
ovaL
L wmmL

v-‘Mv-‘M
v—i
va—‘M
0
Fl!

v—i
Nv—dv—Av—d—an—tmfi'fl'w

 

80L wooL SQL 003

m g L v3 L mom L N3 L LaaL oaaL : 95:00

 

Hfioamwommm MO 30>

 

 

 

fiefimm ”m o>w>>fi Eofimmowm< .3 dew mmo~o< tosoo seam .Lo :oLusnLmeLnL

hm Bank

131

 

 

L L oomL

L L N momL

N L w L a vcaL

L L L L L v moaL
m m v N NL NcmL

v m N m L N 5L LcmL

v m N N L NL ooaL

v m m N N L hL ¢mmL

m m m m L L L mL wmmL

L N v w L N mL nmaL

L L m m m N L L N 5L omaL
N L m L c N 5L mmaL

N N N m v N L 0L VmaL

L L L o N L NL mmaL
oomL wamL 3m L oomL m mo L v3 L mg L NmmL LaoL oamL m 95:00

 

LGDEmmomwm .«O 30>

 

 

Egan Hm 9455 a Eofimmomflw .Lo au> 383‘ tonoU gm .3 scannEanL

885608 B 23

132

 

 

mm 93:

 

 

628350 2%:

 

 

L L m L N L LL mmmL
N L N L L N N LL NmaL
m N N N L N NL meL

L N L m L N oL OmaL
L N L L m ame

L L N wme

L L L L v nme

L L N ovaL

L L N mvaL

L L N wwaL

L L mvaL

L L LvaL

L L ome

L L wmmL

LOON OOON mg L moo L 3a L 82 mmaL vaaL mmmL m toLLOU

 

LGOEmm0mwm MO 30>

 

 

£2:an ”v 963 3 ~58ng .Lo mgr mm80< tonoo 25m .3 :oLSnEmLQ

mm 033.

133

 

 

«EUEZCZOLV mam .UNQLWK

 

 

 

 

L L ccoL
N N mcaL
N m L o vomL
L L L L v mcaL
v m m N NL NoaL
v N N m N L vL LoaL
N m N L w ocaL
N L0 m L L N L N 0L 03L
N m L m L N N m BL wmmL
m N m m m L L L mL hmaL
m w m L m L mL omaL
m m N v L N N 5L meL
N L m N m N m N mL VmoL
LOON OOON mooL waaL 30L ooaL mmaL vaoL moaL m tonoU
208000000 00 30>

 

 

 

8050a ”v 033% E0Ewm0mm< .Lo 30> 0080< .0200 505 .3 533305

@3088 mm 2.3

134

 

 

3: .0}th

 

 

602550 030:

 

 

 

a v N L N L aL ocmL

L w n m N m L N Lm ammL

v L: L L N v N LVN mmmL

m m L m N m N LN 33

N m m c m L L v N ommL

N w L. m N wL mmmL

v N N N N L L 3 $2

L L L N m L L V «L mmaL

L L L L v NmaL

L N N L L N a LmaL

L L L L L m omaL

L L L m avoL

L L L m hvoL

L L mvaL

NaoL LaoL oooL ome wme hwaL owaL mme m 0.5500
508000000 .3 80>

 

 

00:82 ”L 0>m>> Hm EOE0|0000< .Lo a0> 0080< fiasco gm .Lo 5355an

mm 030B

135

7....0LL:—£ .— :32er .3 :..::3.Jp..J.J<La~.:.‘.~\f VUPCL< “-002:er cukmmknv CA».::A~.NL~V<Q

2522.223. V» 3a 072:.

 

 

 

 

 

 

L L onmL

L L homL

L N L w coaL

L m L L o momL
N N N N L a $3

oL m m v N NN momL

L0 w o m N N L L N NcaL
m m m w L L N BL LcmL
NmmL LaoL ommL ome waL nwoL owoL mwaL m 98:00

308000000 .3 30>

 

 

000502 “L 02.03 $080000l0< .Lo a0> 0080< fiasco 505 .3 ”895035

602088 8 fig

136

 

 

I. 1 It

QNQ 0:205

 

 

00:50:00 030:

 

 

m o m L L vL oooL
v o a m N m 5N mmoL
L n m m L N 0L wmoL
v m w v N wL bmoL
N v m n L N aL omoL
N m m N vL mmaL
m m m cmoL

N m N L m LL mmmL

L L L m NmmL

L N L N o LmaL

L L N v onoL

L L N ame

L L nvaL

mmmL cam L mmmL Nam L LoaL omaL m tonoo

 

Lfi0Emm0mmm MO 30>

 

 

000522 “N 0>m>> a 0:0:100000040‘00 30> 0080< fiasco .LthL .3 :oLLBmeLIQu

oLm 07th

137

 

 

p- . -111.I\I_II‘P,\\ ~\-N F\\h\~.‘f~\

 

 

 

 

L L one
L L no.2
L L N oomL
L m L m 33
N L N L L n vca L
m a v m m NN mcaL
v c m w L L LN Noa L
N m m N L LL LomL
mmmL voaL mmoL NmmL LmaL oaoL m tonou
308000000 ,Lo 80>

 

000502 ”N 02$? 00 80800002 .3 80> 0080< 08:00 gm .3 noun—“L505

803088 Sm 0:3

138

 

 

L L Q DEL:

‘

 

 

$025.80 030:

 

 

L w v v L L L 0L oomL
v m w w v m L Nm mmoL

N N o m m N L aL wmmL

m v N h N wL nmoL

N L m c m v N LN omoL

v N o N wL mmoL

L m L m N L L L mL VmaL

N L N N L L N LL mmmL

L L L m NmaL

L N L v w LmaL

L L N v omaL

L L N mme

L N m hwoL

goo L woaL 30L ooaL mooL VomL mam L NamL LaaL ommL m tonou

 

“Guammommm MO 30>

 

 

000502 ”m 02.35 a 0:08m0000< .Lo 30Q000< tonou 505 .3 5:33:05

L Lm 0305

139

 

...“1!.-\\ - uhh ais<tr>

 

 

002

v
v
wmaL

nomL

VMOMv-d

ommL

ogL
$2
oomL
meaL
veaL
80L
NcmL

N L LoaL
m mm L vmaL mooL NamL meL omaL m fiasco

«GOEmmOmmw MO 30>

F4

F—g

v—fi
OOWVv-‘F‘

N
c L
c
m

F...
'—
v—I

\OQQ

H—‘v—l

 

 

 

000502 ”m 0>0>> 00 E0Em0000< .Lo :00.» wouo< tonoo 5.3m .3 533505

3850800 _ _m 2.3

140

 

30:50:00 05003

 

 

 

m m m L L L N L 5L oomL

N m 0 w N m N L wN meL

N m N m m L m N L NN 33

N v L c m m aL bmmL

L w v m v N N L LN omaL

N m c m N 0L mmaL

L N L m N N L NL 33

N N L L m L N L ML mmmL

L L L L v NmmL

L m L L N L a meL

L L L L L m cmaL

L L N ame

L L L m “.me

L L mme

LOON OOON 30L womL nmmL comL momL vamL moaL m 95:00
208000000 .3 30>

 

 

 

000502 6 0>m>> 0&0800003‘ .Lo 30> 0080< 98:00 5.3m .Lo 2032:0005

NLm 030%

141

 

CVULZLCEOLV N ~ MN .3335

 

 

L L o3 L
L L 33
L N L v oomL
L L L m moa L
N N m L L m Luca L
L m o L m N w L momL
m m w v N N L m L NcoL
v m L N L L L m L Lem L
LOON ocoN mom L wooL hmoL oooL mamL vamL momL m tonov

 

Hfi0Emm0mme MO 50>

 

000502 6 006340 E0~mw0000< 0%»; 0080< 05:00 .LthL .3 0033505

6350800 S0 25

142

.082: 03000800 zoom u Lam U006 08% ”02000800 :oLLLsam

M 31803 LL0>0L 80.58 82000800 aoLLLE0m n 0180: 6005300 5 830:0: L0Loo0LLq0 u L0800LL=< LuooLLLuLEo E 080308

8:980 n 83:80 80309800 n .000 ”0000» 5 00:00:00 n .oLLLumL ”259:0 E0LLoLLooL0 u LoLLooL< ”259:0 $5880 u waLuLoEm

 

.5038 00080an 05 LE? 083030000 .8 898:: M 8:00 .LwcL00LE M 0 6030000 H Lv 008:8 000LLwLLL00LE .8 8.502 .082

 

 

 

30 E0 E0 0L: 03 NS N; o o 00me02 .x.
0L ML ML 0 o v v o o 0:02: LL
o 0 LL L L o o L L N

o L L o o o o L L N

o L L o o L L L L w

o L L L L L L L L L

L o LL L L L L L L LL

L L L L L L L L L 80

80 31:00 01:00 38002 32:50 .80 .260 L382 032:0 :58

 

 

Ammm n we 00.50.0003an 80 803an 000LW£00L$L .8 85%

D Nun—733.50

LU 030a.

143

$00588 0503

 

 

L L L L L L o o o o o o o o o o o o o o o o o L L L L L

L L L L L L o o o o o o o o L L L L o o o o L L L L L N

L L L L L L o o o o L L L L o o o o o o o o L L L L L L

L L L L L L o o o o L L L L o o o o L L L L o o L L L L

L L L L L L o o o o L L L L o o o o L L L L o L L L L w

L L L L L L o o o o L L L L o o o o L L L L L L o L L L

L L L L L L o o o o L L L L o o o o L L L L L L L L L mm
L L L L L L o o o o L L L L L L L L L L L L o L L L L N

L L L L L L o o o o L L L L L L L L L L L L L L L L L 0N
L L L L L L L L L L o o o o L L L L o o o o L L L L L m

L L L L L L L L L L L L L L o o o o L L L L L L L L L m

L L L L L L L L L L L L L L L L L L L L L L o L o L L m

L L L L L L L L L L L L L L L L L L L L L L o L L L L mL
L L L L L L L L L L L L L L L L L L L L L L L L o L L N

L L L L L L L L L L L L L L L L L L L L L L L L L L L moL
0N 0N mN 0N MN NN LN ON 3 0L 2 0L 2 0L mL NL LL 2 a m n o m w m N L L550

 

 

LLmNnmU 008%.2 .88 8000083‘ 80 E038 000 0L00L$L .8 x502

 

N0 030%

144

0:30 .8 0800: 3:030: 0:0 3303: u 0N I NN L0 0>0B L0 02000 088080 A0m0 88 05 .8 0830: 33038 n LN I wL LL

0>0>> 0 02000 0:800:80 A0m0 88 05 .8 0800: 33038 n 0L I 0L ”0 0>0B 0 0208 2:80:30 L0m0 88 05 .8 0800: 3303:
n mL I 2 LL 0>03 L0 CoL>0LL0LL 0>L0000ww0 0:0 .8L>0LL0LL 3:088:00 080308 838030 000008038383 0030 0:88:30 40m0
0:8 05 8 0800: 0:83: u m I 0 ”83:30 080308 8 088an0 30:08.8 80:08 n 0 ”0.8088 803:0 30:08 n v ”08088

80880 30:08 8 33:03 N m ”08088 880.0 30:08 n N ”0:30 300:0 80880 n L 0:0 0:3 0:0 5 000:3: 00LLL0L.L0>

 

05. .8038 000880 20 L005 08308008 .8 808:: M 8:00 .3503 u 0 002008 n LL 050308 00088003: .8 5302 .082

 

 

 

0 0 0 0 m 0 LLL LLL LLL LLL LLL LLL LLL LLL L0 L0 L0 L0 LL LL LL LL mm N LL L LL wage;
LL LL LL LL LL LL L L L L LL LL LL LL L L L L LL LL LL LL LL LL LL LL L L

LL LL LL LL LL LL L L L L LL LL LL LL L L L L LL LL LL LL L L LL L L L

LL LL LL LL LL LL L L L L LL LL LL LL L L L L LL LL LL LL L L L L L N

LL LL LL L L L LL LL LL LL L L L L LL LL LL LL L L L L L L L L L L

LL LL LL L L L L L L L L L L L L L L L L L L L L L L L L L

L L L LL LL LL LL LL LL LL L L L L LL LL LL LL L L L L L L L L L L

R

0NmNONMNNNLNONaLwLnLOLvaLmLNLLLoLaw 00 vamNL 3300

 

 

LLwNua 0080002 .0L8m 8000083» :8 80308 000 :L00LSL ,8 8.302
00:83:08 N0 0LLL0.L.

145

1. . -4 4:31:55 25...:L..:I....:v..:L\L:..:u. :CCLBLLLLL :52205323

 

 

50280800: .de 0:0 .de .vLN .wnL .th .th .mw.wm £de
.mw.wm .mw.wm .0m.m .0m.m .0m.m .0m.m .NNdL .NNdL .NNdL .NNdL .La.m .Lmd .Lm.m .Lm.m .ovNL .Lhd .mm.~ find .06 0:03 ND 030.0 :L

m0LnL0L.L0> mm 80 8000 88 8600: .8 8080808 0LL.LI 6330000: 0:0 40308852008880 5230005088080 :oL80s0v 8080:0888

146

Table C3

= 88)

 

 

Pattern for Adolescent Dgghters’ Megures (1:1

 

Matrix of Missin ess

 

22 23 24 25 26 27

19 20 21

9101112131415161718

8

Count

 

00001

1

1

0 0 0 0
0 0 0 0
O 0 0 0

000000000000

0000000000000000

1

1
1

1
1

1 1
1

1

1
1

0000

1
1
1

26

0000

1
1

1
1
1

147

00000001

00000000

000000000000

00000000000000000000

111

1

00000000

0000

1
1

l
1
l

l

1

1

1

l

1

000000

1
1
1

000000

1
1

111000011
000011

1

1
1

0000
0000

1
1

27

000000

1

1

1

 

(table continues)

00LLL0LL0> mm ,8 0000 .88 w:L00LL: .Lo 0w0800008 088. 0930000: 0:0 40.308003200000880 5030000088080 88033 80500882
0080 .Lo 0800: L0E0L08 0:0 L0E0L0E H um I mm 0 0>03 0 0208 0:80:80 LumU 0:08 05 .Lo 0800: 8:088 u LN I wL LL

0>03 0 02000 0:880:80 ILUmLU 0:08 05 .Lo 0800: L0E0L08 n 0L I 0L L0 903 0 0230 0:80:80 Aomo .508 05 .Lo 0800: L0E0L0E
n mL I 3 LL 0>0B L0 CoL>0LL0LL 0>L000wa0 0:0 830:0: 80:LL:LL00 .080308 :28000 000080008083 03000 0800080 Lumu
0:08 08 80 08000 L0E0L08 n 8 I 0 888:0 L0E0L0L: 0L 0::008LL0 30:08 .8 888080 N n “0800808 w:L0L:L.L0 30:08 N 0 ”080085

8on80 L0L0:08 Lo 380:0 n m ”0800800 8on80 00:08 N N "0880 L008 83080 n L 0:0 033 05 E 000208 00LLL0L.L0>

 

00% 80:08 000L080 05 L005 0:0LL0E00LLLL .Lo 0008:: M 8:00 .8808: n 0 6030000 .I. LL 0:00:08 00088008: .Lo 8002 .082

 

L0 L0. L0 mm mm mm m0 m0 m0 m0 m0. m0 m0 m0 mm hm hm 8m. 8m an am mm m N v N o 8:00:50

 

 

nmommNVNMNNNLNomoLwLnLoLvaLmLNLLLoLaw nomLVmNL 8:00

 

 

 

g n 3 0080002 .00: :0Q 80000Lo0< .88 E0008 000 :LO0LSL .8 88.0.2.

880088 8 2000

148

50.500800: .8000 0:0 .8000 .8000 .wL.m0 .wL.m0 .wL.m0 .0L.Lm .0L.Lm .0L.Lm

.0L.Lm .0w.w0 .0w.w0 .0w.w0 .0w.m0 .0980 .mo.m0 £0.80 .mo.m0 .Nm.00 .Nm.00 .Nm.00 .Nm.00 .L0.m .nmd .000 KNN .06 0003 m0 0308. :L

149

APPENDIX D

Table D1

Number of Cases in the TRAJ Analysis of Smolglg (_11 = 262)

 

 

Age Number of cases Mean Standard Deviation
22 2 4.50 .71
23 5 3.80 .45
24 11 2.91 1.51
25 11 3.55 1.57
26 17 3.53 1.70
27 27 2.85 1.63
28 29 3.90 1.45
29 32 3.09 1.69
30 32 3.75 1.34
31 52 3.02 1.99
32 49 3.41 1.82
33 59 3.25 1.65
34 58 3.52 1.65
35 55 3.31 1.86
36 69 2.99 1.87
37 47 3.30 1.86
38 51 3.12 2.03

 

(table continues)

150

Table D1 (Continued)

Number of Cases in the TRAJ Analysis of Smoking (3 = 262)

 

 

Age Number of cases Mean Standard Deviation
39 43 3.16 1.76
40 44 2.89 2.03
41 28 3.00 1.83
42 30 2.37 2.06
43 19 3.21 2.32
44 17 3.06 2.19
45 1 1 4.00 2.14
46 9 3.22 2.59
47 8 2.25 2.43
48 7 3.14 2.67
49 2 2.50 3.54
50 6 3.67 1.75
51 3 4.67 1.53
52 2 5.50 .71
53 1 4.00 .00
54 2 6.00 .00
55 1 5.00 .00

 

151

Table D2

Number of Cases in the TRAJ Analysis of Alcoholism: Men (Q = 202)

 

 

 

Age Number of cases Mean Standard Deviation
14 200 .02 .21
15 199 .09 .48
16 192 .14 .62
17 183 .29 .84
18 163 .61 1.13
19 123 .58 1.10
20 94 .56 1.10
21 73 .56 1.11
22 57 .46 1.04
23 49 .33 .90
24 45 .40 .94
25 42 .83 1.19
26 39 .87 1.20
27 39 1.03 1.25
28 35 .89 1.21
29 41 .93 1.29
30 56 .73 1.09
31 58 .88 1.23

(table continues)

 

152

Table D2 (Continued)

Number of Cases in the TRAJ Analysis of Alcoholism: Men (Q = 202)

 

 

Age Number of cases Standard Deviation

32 64 .81 1.08
33 69 .84 1.21
34 67 .72 1.07
35 73 .82 1.21
36 79 .65 1.05
37 77 .57 1.04
38 77 .68 1.14
39 64 .66 1.12
40 59 .78 1.25
41 45 .73 1.12
42 32 .63 1.13
43 25 .96 1.24
44 25 1.04 1.21
45 19 .89 1.20
46 15 .33 .72
47 16 .69 1.01
48 13 .38 .77
49 7 .29 .76

 

153

(table continues)

Table D2 (Continued)

Number of Cases in the TRAJ Mews of Alcoholism: Men (g = 202)

 

 

 

Age Number of cases Standard Deviation

50 9 .33 1.00

51 6 .00 .00

52 7 .57 .98

53 4 .25 .50

54 4 .00 .00
154

 

 

Table D3

Number of Cases in the TRAJ Analysis of Alcoholism: Women (_n = 109)

 

 

Age Number of cases Mean Standard Deviation
14 109 .04 .30
15 107 .19 .72
16 100 .07 .36
17 95 .12 .50
18 89 .65 ‘L10
19 63 .32 .82
20 54 .22 .72
21 51 .31 .86
22 47 .11 .37
23 46 .13 .54
24 47 .47 .97
25 39 .15 .54
26 35 .02 .51
27 41 .32 .79
28 47 .47 .95
29 46 .22 .70
30 47 .34 .87
31 47 .53 'L04

 

(t_able continues)

155

Table D3 (Continued)

Number of Cases in the TRAJ Analysis of Alcoholism: Women (Q = 109)

 

 

Age Number of cases Mean Standard Deviation

32 53 .66 1.09
33 48 .50 .90
34 52 .42 .87
35 51 .57 1.10
36 45 .47 .94
37 33 .39 1.00
38 34 .56 1.05
39 30 .43 .97
40 24 .67 1.24
41 20 .55 1.10
42 15 .00 .00
43 10 .00 .00
44 6 .00 .00
45 7 .43 1.13
46 7 .00 .00
47 4 .00 .00
48 3 1.00 1.73
49 1 .00 .00

 

156

 

.8. v m .. .082

 

 

 

 

 

wm.m N0; 08d 00.0 800 08; 00.m 0N.N x02: seamen—00 xQom
$8 002 N0.: 30— 052 0~.0_ 08.8 08.8 “who? I 558500 cozmfimm
N00 80.0 00.8 00.0 2.0 000 _00 000 €0.58 I 88.0.8000 cow—:52
00.0 00 .8 05$ 80.: 00.0 00.0 _0.0 08.0 moo—0300 E .8323 30802
00.0 80.0 005 00.2 $0 00.x. 00.0 005 0000220 E man—088 8:980
0NAMN mmgmm 5N2 $ng 5.03 00:8 58.88“ 00.20 nova—800
mud 00.0 0 00.0 08.2 0~.m 0N.0_ m0N 00.00 88% E cosmosvm
Qm :82 Om :82 Om :32 90 532 203.2;
Emzéfigmom bonm .933 Moo—08m Ems .50—95-82

 

m 0392ng

 

 

:02 8813030 msfloEm 00 30380 o>untomoa

0m 2an

157

 

8. v m .. .eoz

 

 

 

00.0 00.0 08.0 00.0 00.0 88.0 00.0 0.0 x005 8000800 080m
00.00 5 _ .83 00.: 00.53 00.00 00.03 00.: 08.2 00.83 I 8600800 85503
30.0 00.5 00.0 00.5 00.5 00 .5 2.0 05.0 E0080 1 02000800 8580i
00.0 50.5 00.0 00.0 00.0 00.0 05.0 00.0 0005300 E 830303 388050
00.0 5 3 . S 08.0 05.0 50.0 08.5 00.0 05.0 0003230 E 08030.8 8:080
00.00 00. _50 80.08 00.300 00. 3 0_ 00.000 00.503 50.00 8009800
05. 50.03 00.3 00.03 00.0 00.03 58; 05.0_ 0000.» E 80003000
00 0002 Q0 2002 Q0 0002 Om 000—2 03000>
302-00-3003 000380 .9030 000380 303 000380-:02

 

 

 

20:83 ”0086330 000.880 .8 803% 03308009

mm 030.0

158

 

3. v m .. .eoz

 

 

 

 

 

$0 80 32 N2 42 80 0.005 80.280 080
n 0 . : 5.2 $2 2.2 80 we.» as; I 88280 85.50
a? 08 oi. 21. 8.... $0 8.5.. I 80.280 85.50
w: 8.2 :0 00.0 8.0 3.0 08530.. 5 8328 38.09..
00.0 8. : 8.» 8.2 :0 05.0 08:03.0 5 0.8038 3.680
8...: 3.08 80: 8.30 8.08 00.09 80888
:0 8.2 80 52 20 Si 38» a 808.60
am 502 mm :82 am :82 O§E>
a 826082 H 8.2882 £280-82

 

 

:02 ”0085300 8022302 .80 000.00% 0&8180Q

0m 030.0

159

 

8. v m .. .802

 

 

 

 

 

00.0 00.0 00.0 2.0 8.0 00.0 0.005 80.280 080
2.2 2.2 00.2 E2 002 2.2 as; .. 88230 80:50:
0 2 .0 m: 00.0 3.0 2.0 00.0 0550 I 80.20% 85800
00.0 as 5.0 00.0 2.0 00.0 085300. 5 82.000 08.090
00.0 02 _ 00.0 00.0 02 8.0 080020 a 082080 02.080
2.02 8.000 8.02 00.8... 00.02 00.000 8083.8
02 00.2 02.0 202 00.2 20.2 0.0% a 808000
80 082 am :82 mm 9.02 2080.»
a 8.00082 2 5.0.582 00880.82

 

 

00:83 “0085300 828382 80 000000.010 03808002

0m 030.0

160

APPENDIX F

Table Fl

Precursors and Concurrent Factors of Early Smoking Onset: Maternal Ratings of Sorg

 

 

 

 

 

 

 

Smoking onset Smoking onset
by age 14 Non-smoker unknown
Variable Mean SD Mean SD Mean SD
Prenatal exposure to maternal smoking and drinking
Daily smoking 6.02 9.51 3.89 7.47 2.64 6.62
Weekly drinking .24 .53 .51 1.37 .22 .57

_E_arly temperament (Ages 3 — 5)

 

Attention span 4.97 2.74 5.73 3.03 4.62 2.72
Approach/Withdrawal 3.79 1 .90 3 .84 1.89 3.57 1.93
Reactivity 3.48 1.51 2.87 1.63 3.11 1.71

Early child behavior problems (Ages 3 — 5)

Anxious/Depressed 3.30 3.37 2.04 1.86 2.00 2.33
Attention problems 3.70 2.79 3.16 2.40 2.56 2.25
Delinquent behavior 2.70 2.84 1.83 1.68 1.52 1.19
Aggressive behavior 12.16 5.87 10.00 5.75 9.79 5.11

Concurrent adolescent behavior problems (Ages 12 —- 14)

 

Anxious/Depressed 3.40 3.68 2.47 2.47 3.60 2.90

Attention problems 4.13 3.48 2.66 2.55 3.83 3.14

Delinquent behavior 3.21 2.76 1.35 1.53 1.74 1.83

Aggressive behavior 9.68 6.18 6.41 4.56 7.24 5.14
PM; * p < .05

161

Table F2

Precursors and Concurrent Factors of Early SmokinLOnset: Paternal Ratings of Sons

 

 

 

Smoking onset Smoking onset
by age 14 Non-smoker unknown
Variable Mean SD Mean SD Mean SD

 

Early temperament (Ages 3 — 5)

Attention span 5.02 3.16 5.17 3.03 4.23 2.86
Approach/Withdrawal 4.16 1.67 3.74 1.67 3.39 1.91
Reactivity 3.52 1.53 3.12 1.45 3.49 1.44

Early child behavior problems (Ages 3 — 5)

Anxious/Depressed 2.79 2.89 2.14 2.16 2.43 2.46
Attention problems 3.62 2.87 3.01 2.36 3.1 1 2.40
Delinquent behavior 2.13 1.71 1.70 1.46 1.87 1.96
Aggressive behavior 10.67 6.50 9.37 5.48 10.42 5.39

Concurrent glolescent beh_avior problems (Ages 12 — 14)

 

Anxious/Depressed 2.63 2.87 . 2.56 2.63 3.92 3.41

Attention problems 3.71 3.01 2.70 2.51 4.24 3.78

Delinquent behavior 2.84 2.39 1 .35 1.51 1.72 1.77

Aggressive behavior 9.20 6.76 6.38 4.73 7.87 5.83
Note. * p < .05

162

APPENDIX C

Table Gl
Configurations of Parental Smokm Patterns and Early Smoking Onset Among
Adolescent Dauglfirs

 

 

MPS Obs. Freq. Exp. Freq. L Type/Antitype
111 13 5.92 3.54 Type
112 3 6.24 -1.44

113 20 14.03 2.24

121 2 1.08 .45

122 O 1.13 -.66

123 2 2.55 -.04

131 2 4.31 -1.07

132 2 4.54 -1.18

133 6 10.20 -1.64

211 1 1.30 .19

212 2 1.37 .12

213 2 3.09 -.41

221 0 .24 .56

222 0 .25 .52

223 2 .56 1.33

231 O .95 -.50

232 3 1.00 1.62

 

(table continues)

163

Table 61 (Continued)

Configurations of Parental Smoking Patterns and EarlySmcfiing Onset Among

 

 

 

Adolescent Daighters
MPS Obs. Freq. Exp. Freq. I_. Type/Antitype
233 1 2.25 -.58
31 1 O 2.73 -l .56
312 0 2.87 -1.63
313 3 6.45 -1.52
321 O .50 .01
322 0 .52 -.03
323 2 1.17 .34
331 1 1.98 -.38
332 10 2.09 5.76 Type
333 7 4.69 1.03

 

1:193; M = maternal smoking pattern; P = paternal smoking pattern; S = smoking onset
by adolescent girls. Numerals in MPS column represent ordered triples of variable
categories. Response categories for parental smoking were 1 = Non-smoker, 2 =

Li ght/I-Ieavy-to-li ght smoker, and 3 = Heavy smoker for parental smoking, and options
for adolescent smoking onset were 1 = Never smoked, 2 = Smoked by age 14, and 3 =
Smoking-onset unknown. I_. stands for Lehmacher’s test with continuity correction
(1981); Bonferroni-adjusted alpha (0.00185) was used. Pearson’s X_2 = 70.35 for df = 20,

Q = .00.

164

Table G2
Configurations of Parental Alcoholism Patterns and Early Smoking Onset Among

 

 

 

Adolescent Daughters
MAPAS Obs. Freq. Exp. Freq. L Type/Antitype
111 8 3.85 2.57
112 0 4.05 -2.15
113 16 9.11 2.95 Type
121 3 3.85 -.21
122 4 4.05 .27
123 9 9.11 .18
131 1 3.85 -1.45
132 7 4.05 1.48
133 3 9.11 -2.59
211 2 1.73 -.20
212 1 1.83 -.27
213 1 4.11 -l.55
221 3 1.73 .64
222 1 1.83 -.27
223 5 4.11 .23
231 1 1.73 -.20
232 2 1.83 -.27
233 7 4.11 1.43

 

flgble continues)

165

Table GZ (Continued)

Configurations of Parental Alcoholism Patterns and Early Smoking Onset Among

 

 

 

Adolescent Daughters

M APAS Obs. Freq. Exp. Freq. L Type/Antitype
311 O .75 -.31
312 0 .79 -.35
313 0 1.79 -1.10
321 1 .75 -.31
322 2 .79 .85
323 0 1.79 -1.10
331 0 .75 -.31
332 3 .79 2.05
333 4 1.79 1.47

 

N_ot_§_:_._ MA = maternal alcoholism pattern; PA = paternal alcoholism pattern; S = smoking
onset by adolescent girls. Numerals in MAPAS column represent ordered triples of
variable categories. Response categories for parental alcoholism were 1 = Non-alcoholic,
2 = Alcoholism I, and 3 = Alcoholism II. Options for adolescent smoking onset were 1 =
Never smoked, 2 = Smoked by age 14, and 3 = Smoking-onset unknown. L stands for
Lehmacher’s test with continuity correction (1981); Bonferroni-adjusted alpha (0.00185)

was used. Pearson’s _Xi = 45.58 for df = 20, p = .00.

166

Table G3
MANOVA Results on Precursors and Concurrent Characteristics of Adolescents with

Different Smokim Onset Status Among Daughters

 

 

 

 

 

 

Maternal rating Paternal rating
Multivariate E 1.29 — 2.63* .83 -— 3.26*
Wilk’s A .684* — .812 636* -— .872
Variable Univariate analysis
Prenatal exposure
Daily maternal smoking 612* —- 7.81 *
Weekly maternal drinking 1.80 - 3.08
Early temperament (Ages 3 — 5)
Attention span .18 — 2.52 .10 — 1.62
Approach/Withdrawal .03 — 3.52* .04 - 4.28*
Reactivity .06 -— 2.00 .20 - 2.50

Early child behavior problems (Ages 3 - 5)

Anxious/Depressed .02 - 3.77* .08 — 3.05
Attention problems 1.04 —- 3.50* .06 — 2.46
Delinquent behavior 2.61 — 10.37* .87 -— 5.78"“
Aggressive behavior 1.74 — 6.15* .37 — 6.07*

Concurrent ad_olescent behavior problem_s (Ages 12 — 14)

Anxious/Depressed .59 — 2.05 .16 -— 3.15*

Attention problems 1.61 — 8.74* 2.13 — 883*
Delinquent behavior 2.89 - 6.06* 3.02 — 11.42*
Aggressive behavior 1.25 — 3.62* 2.09 — 7.37*

 

167

Note. * p < .05. Degrees of freedom for multivariate analysis of variance 1, 86 for both

maternal and paternal ratings; degrees of freedom for univariate E test were 2, 85.

168

Table G4

Precursoi and Concurrent Factors of Early Smoking Onset: Maternal Ratings of

 

 

 

 

 

 

 

 

 

Daughters
Smoking onset Smoking onset
by age 14 Non-smoker unknown
Variable Mean SD Mean SD Mean SD
Prenatal exposure to maternal smoking and drinking
Daily smoking 7.58 9.70 .32 1.31 2.78 5.48
Weekly drinking .48 1.22 0.09 .35 .09 .32
Early child behavior problems (Ages 3 — 5)
Attention span 4.37 2.94 5.85 2.76 5.38 3.75
Approach/Withdrawal 3.83 1.78 3.38 2.03 3 .68 1.98
Reactivity 2.96 1.61 2.81 1.71 2.71 1.49
Child behavior problems at wave 1 (Ages 3 — 5)
Anxious/Depressed 2.81 2.88 1.97 1.75 2.33 2.00
Attention problems 3.1 1 2.38 2.38 2.00 2.07 1.75
Delinquent behavior 2.29 1.85 1.79 1.58 1.22 1.28
Aggressive behavior 10.57 5.49 7.86 5.08 7.63 4.82
Concurrerfldmlegscent behavior problema (Ages 12 — 14)
Anxious/Depressed 4.18 4.50 2.82 2.65 3.36 2.93
Attention problems 2.65 3.23 1.25 1.85 3.18 2.66
Delinquent behavior 2.81 3.01 1.18 1.20 1.64 1.56
Aggressive behavior 8.54 6.15 5.56 4.13 6.85 4.86
N_ot_e_._ * p < .05

169

Table G5

Precursors and Concurrent Factors of EarlgSmoking Onset: Paternal Ratings of

Daughters

 

 

 

Smoking onset Smoking onset
by age 14 Non-smoker unknown
Variable Mean SD Mean SD Mean SD

 

Early temperament (Ages 3 — 5)

Attention span 5.21 2.88 5.52 3.18 5.40 2.99
Approach/Withdrawal 3.39 1 .97 3.57 1.74 4.07 1.82
Reactivity 3.28 1.61 3.15 1.73 2.79 1.70

Early child behavior problems (Ages 3 — 5)

Anxious/Depressed 2.88 2.43 2.23 2.39 2.08 2.09
Attention problems 2.80 2.18 2.39 2.22 2.37 1.75
Delinquent behavior 2.21 1.30 1.58 1.58 1.48 1.29
Aggressive behavior 10.31 5.66 7.44 5.63 8.23 5.35

Concurrent adolescent behavior problems (Ages 12 — 14)

 

Anxious/Depressed 2.92 2.70 1.92 2.96 3.41 2.64

Attention problems 2.93 2.68 1.39 2.19 3.28 2.54

Delinquent behavior 2.36 1.96 .84 .88 1.51 1.56

Aggressive behavior 8.42 5.58 3.53 4.33 6.78 5.41
Note. * p < .05

170

APPENDIX H

DRINKING AND OTHER DRUG USE (PARENT)
Follow-Up Information - Form B; 12/97

This questionnaire takes about 15 minutes to complete. All information will be used for
research only and will be kept strictly confidential. If you are not sure of the answer to a
question please answer the best you can. Please try to answer each item.

A. THE FOLLOWING QUESTIONS ARE ABOUT YOUR DRINKING OF
ALCOHOLIC BEVERAGES DURING THE PAST 6 MONTHS
(that is, since last to now.):

 

1. OVER THE LAST 6 MONTHS. ON THE AVERAGE, HOW MANY DAYS A

MONTH HAVE YOU HAD A DRINK?
days a month.

 

2. OVER THE LAST 6 MONTHS. ON A DAY WHEN YOU ARE DRINKING, HOW
MANY DRINKS DO YOU USUALLY HAVE IN 24 HOURS? (A DRINK IS A 12
OZ. CAN, GLASS OR BOTTLE OF BEER; A 4 OZ. GLASS OF WINE; A SINGLE

SHOT; OR A "SINGLE MIXED DRINK")

 

 

A little more than average

drinks per 24 hours.
3. OVER THE PAST 6 MONTHS. WHEN YOU GOT DRUNK, HOW BAD WAS
YOUR HANGOVER?
Never bad Pretty bad
Not bad Terrible
A little less than average Worst possible
Average Never drank enough to get a hangover

IF YOU DRANK NO ALCOHOLIC BEVERAGES AT ALL (NOT EVEN A FEW SIPS) IN
THE LAST 6 MONTHS, GO NOW TO PAGE 6, SECTION C.

ALL OTHERS CONTINUE ON THE NEXT PAGE

171

 

B.

THE FOLLOWING QUESTIONS ARE ABOUT YOUR DRINKING PATTERNS. IN
ANSWERING THE QUESTIONS, PLEASE THINK ABOUT WHAT YOU HAVE
DONE ON THE AVERAGE OVER THE LAST SD( MONTHS.

1 . WHEN DRINKING WINE:

a. HOW OFTEN DO YOU USUALLY HAVE WINE OR A PUNCH CONTAINING WWE?

3 or more times a day
2 times a day

Once a day

Nearly every day

3 or 4 times a week
once or twice a week

2 or 3 times a month
About once a month
Less than once a month,

but at least once a year
Less than once a year

NEVER [If checked, go to
question #2a]
b. THINK OF ALL THE TIMES YOU HAD WINE OR A PUNCH CONTAINING WINE
RECENTLY. WHEN YOU DRINK WINE, HOW OFTEN DO YOU HAVE 10 OR
MORE GLASSES?

 

Nearly every time: SKIP TO QUESTION #2 BELOW

More than half the time: SKIP TO QUESTION #2 BELOW
Less than half the time

Once in a while
NEVER

C.

WHEN YOU DRINK WINE OR A PUNCH CONTAINING WINE, HOW OFTEN DO
YOU HAVE AS MANY AS 7 TO 9 GLASSES?

Nearly every time: SKIP TO QUESTION #2 BELOW

More than half the time: SKIP TO QUESTION #2 BELOW
Less than half the time

Once in a while
NEVER

(1.

WHEN YOU DRINK WINE OR A PUNCH CONTAINING WINE, HOW OFTEN DO
YOU HAVE AS MANY AS 5 to 6 GLASSES?

Nearly every time: SKIP TO QUESTION #2 BELOW

More than half the time: SKIP TO QUESTION #2 BELOW
Less than half the time

Once in a while
NEVER

172

 

WHEN YOU DRDIK WNE OR A PUNCH CONTAINING WINE, HOW OFTEN DO
YOU HAVE AS MANY AS 3 to 4 GLASSES?

Nearly every time: SKIP TO QUESTION #2 BELOW

More than half the time: SKIP TO QUESTION #2 BELOW

Less than half the time
Once in a while
NEVER

WHEN YOU DRINK WINE OR A PUNCH CONTAINING WINE, HOW OFTEN DO
YOU HAVE 1 TO 2 GLASSES?

6.

Nearly every time
More than half the time
Less than half the time

Once in a while

2. WHEN DRNKING BEER

a. HOW OFTEN DO YOU USUALLY HAVE BEER?
2 or 3 times a month

3 or more times a day

2 times a day About once a month

Once a day Less than once a month,

Nearly every day but at least once a year

3 or 4 times a week Less than once a year

Once or twice a week NEVER [If checked, go to
question #3a]

b. THINK OF ALL THE TIMES YOU HAD BEER RECENTLY. WHEN YOU DRINK
BEER, HOW OFTEN DO YOU HAVE 10 OR MORE CANS, GLASSES OR

BOTTLES?

Nearly every time: SKIP TO QUESTION #3 BELOW
More than half the time: SKIP TO QUESTION #3 BELOW

Less than half the time
Once in a while
NEVER

WHEN YOU DRINK BEER, HOW OFTEN DO YOU HAVE AS MANY AS 7
TO 9 CANS, GLASSES OR BOTTLES?

Nearly every time: SKIP TO QUESTION #3 BELOW
More than half the time: SKIP TO QUESTION #3 BELOW

Less than half the time
Once in a while
NEVER

173

 

 

3.

WHEN YOU DRINK BEER, HOW OFTEN DO YOU HAVE AS MANY AS 5
TO 6 CANS, GLASSES OR BOTTLES?

Nearly every time: SKIP TO QUESTION #3 BELOW

More than half the time: SKIP TO QUESTION #3 BELOW
Less than half the time

Once in a while
NEVER

WHEN YOU DRINK BEER, HOW OFTEN DO YOU HAVE AS MANY AS 3
TO 4 CANS, GLASSES OR BOTTLES?

Nearly every time: SKIP TO QUESTION #3 BELOW

More than half the time: SKIP TO QUESTION #3 BELOW
Less than half the time

Once in a while
NEVER

WHEN YOU DRINK BEER, HOW OFTEN DO YOU HAVE 1 TO 2
CANS, GLASSES OR BOTTLES?

Nearly every time
More than half the time
Less than half the time
Once in a while
NEVER

WHEN DRNKWG WHISKEY OR LIQUOR

HOW OFTEN DO YOU USUALLY HAVE WHISKEY OR LIQUOR (SUCH AS

MARTINIS, MANHATTANS, HIGHBALLS, OR STRAIGHT DRINKS INCLUDING
SCOTCH, BOURBON, GIN, VODKA, RUM, ETC.)?

3 or more times a day

2 or 3 times a month
2 times a day About once a month
Once a day Less than once a month,
Nearly every day but at least once a year
3 or 4 times a week Less than once a year
__ Once or twice a week

NEVER [If checked, go to
question #4]

174

 

 

THINK OF ALL THE TIMES YOU HAD DRINKS CONTAINING WHISKEY OR
OTHER LIQUOR RECENTLY. WHEN YOU HAVE HAD THEM, HOW OFTEN DO
YOU HAVE 10 OR MORE DRINKS?

Nearly every time: SKIP TO QUESTION #4 BELOW

More than half the time: SKIP TO QUESTION #4 BELOW
Less than half the time

Once in a while
NEVER

WHEN YOU HAVE HAD DRINKS CONTAINING WHISKEY OR OTHER
LIQUOR, HOW OFTEN DO YOU HAVE AS MANY AS 7 TO 9 DRINKS?

Nearly every time: SKIP TO QUESTION #4 BELOW

More than half the time: SKIP TO QUESTION #4 BELOW
Less than half the time

Once in a while
NEVER

 

WHEN YOU HAVE HAD DRINKS CONTAINING WHISKEY OR OTHER
LIQUOR, HOW OFTEN DO YOU HAVE AS MANY AS 5 TO 6 DRINKS?

Nearly every time: SKIP TO QUESTION #4 BELOW

More than half the time: SKIP TO QUESTION #4 BELOW
Less than half the time

Once in a while
NEVER

e.

WHEN YOU HAVE HAD DRINKS CONTAINING WHISKEY OR LIQUOR,
HOW OFTEN DO YOU HAVE 3 TO 4 DRINKS?

Nearly every time: SKIP TO QUESTION #4 BELOW

More than half the time: SKIP TO QUESTION #4 BELOW
Less than half the time

Once in a while
NEVER

f.

WHEN YOU HAVE HAD DRINKS CONTAINING WHISKEY OR LIQUOR,
HOW OFTEN DO YOU HAVE 1 TO 2 DRINKS?

__ Nearly every time
__ More than half the time
__ Less than half the time
__ Once in a while

__ NEVER

 

175

WHEN DRINKING ANYTHING. CHECK HOW OFTEN YOU HAVE ANY DRINK
CONTAINING ALCOHOL. WHETHER IT IS WINE, BEER, WHISKEY OR ANY
OTHER DRINK. MAKE SURE THAT YOUR ANSWER IS NOT LESS FREQUENT

THAN THE FREQUENCY REPORTED ON ANY OF THE PRECEDING

QUESTIONS.
2 or 3 times a month

3 or more times a day

2 times a day About once a month

Once a day Less than once a month,

Nearly every day but at least once a year

3 or 4 times a week Less than once a year
Never

Once or twice a week

NOW, WE'D LIKE YOU TO SHIFT GEARS AND THINK ABOUT THE PERIOD

 

 

 

C.
FOR THE 2 AND A HALF YEARS BEFORE THIS YEAR

1. OVERALL DURING THAT TIME, WOULD YOU SAY YOUR DRINKING WAS
PRE I'l Y MUCH THE SAME AS IN THIS PAST 6 MONTHS, MORE THAN IN THIS
PAST 6 MONTHS, OR LESS THAN IN THIS PAST 6 MONTHS?

My drinking was:

A lot more than in this past 6 months
Somewhat more than in this past 6 months
About the same as in this past 6 months

Somewhat less than in this past 6 months

F”

A lot less than in this past 6 months
OVER THOSE TWO AND A HALF YEARS (BETWEEN 19 AND 19_), ON
THE AVERAGE, HOW MANY DAYS A MONTH DID YOU HAVE A DRINK?

days a month.
[If you did not drink at all during that time, go to section E]

 

OVER THOSE TWO AND A HALF YEARS. ON A DAY WHEN YOU WERE
DRINKING, HOW MANY DRINKS DID YOU USUALLY HAVE IN 24 HOURS? (A
DRINK IS A 12 OZ. CAN, GLASS OR BOTTLE OF BEER; A 4 OZ. GLASS OF

WINE; A SINGLE SHOT; OR A "SINGLE MDIED DRINK")

drinks per 24 hours.

 

176

 

P

”lllllllll

9’

F"

.0

P-

OVER THOSE TWO AND A HALF YEARS, WHEN YOU GOT DRUNK, HOW BAD

WAS YOUR HANGOVER?

Never bad

Not bad

A little less than average

Average

A little more than average

Pretty Bad

Terrible

Worst possible

Never drank enough to get hangover

WAS THERE ANY PERIOD IN HERE DURING WHICH YOU DID NOT DRINK AT

ALL?
YES NO

IF YES:
For how long a time did that last?
I did not drink at all for months.

 

When was this?
From / to /

 

(month) 6?)— (month) (yr)

What led you to stop when you did?

 

 

What led you to begin drinking again, if you did?

 

 

177

D

l. OVER THE LAST 3 YEARS, THH\lK OF THE 24 HOUR PERIOD WHEN YOU DID THE
MOST DRINKING. ON THAT DAY, HOW MANY DRINKS DID YOU HAVE? (A DRINK
IS A 12 OZ. CAN, GLASS OR BOTTLE OF BEER; A 4 OZ. GLASS OF WINE; A SINGLE

SHOT; OR A SINGLE MD(ED DRINK).

30 or more drinks

25 - 29 drinks
20 - 24 drinks
15 - 19 drinks
lO - l4 drinks
7 - 9 drinks
5 - 6 drinks
3 - 4 drinks
1 - 2 drinks
None

2. APPROXIMATELY WHEN DID THIS HAPPEN?

ANSWER KEY FOR QUESTIONS BELOW:

1 2

3-5 6-10 11-20 21-50 51-100

9

(month) (year)

101-250 251-500 500+ (more than 500)

NOW SOME QUESTIONS ABOUT OUTCOMES PEOPLE SOMETIMES HAVE

BECAUSE OF DRINKING. DURING THE LAST 3 YEARS. HAVE YOU HAD ANY
OF THE FOLLOWING HAPPEN BECAUSE OF YOUR DRINKING?

YES

NO

(check one)

1. Missed school or time on job

2. Thought I was drinking too
much

3. Gone on a binge of constant
drinking for 2 or more days

4. Lost friends

My spouse or others in my
family (my parents or children)
objected to my drinking

6. Felt guilty about my drinking

Divorce or separation

178

 

IN THE LAST JUST IN THE
3 YEARS: HOW LAST YEAR-
MANY TIMES Last 12 months.
(Use key)”- HOW MANY
TIMES?
(Use kcy)’

10.

ll.

12.

13.

14.

15.

l6.

17.

18.

19.

20.

YES

(check one)

Took a drink or two first
thing in morning

Restricted my drinking to
certain times of day or week
in order to control it

or cut down (like after 5PM,
or only on weekends,

or only with other people)

Been fired or laid off

Once started drinking,
kept on going till
completely intoxicated

Had a car accident when I
was driving

Kept on drinking after

I promised myself not to

Had to go to a hospital
(other than accidents)

Had to stay in a hospital
overnight

Had the shakes "the
morning after"

Heard or saw or felt things
that weren't there
(hallucinations), several
days after stopping drinking

Had blackouts (couldn't
remember later what
you'd done while drinking)

Been given a ticket for
drunk driving (DWI or DUIL)

Had jerking or fits
(convulsions) several days
afler stopping drinking

179

NO

IN THE LAST

3 YEARS: HOW

MANY TIMES
(Use key)‘+

JUST IN THE
LAST YEAR-
Last 12 months.

HOW MANY

TIMES?

(Use key)‘

21.

22.

23.

24.

25.

26.

27.

28.

29.

YES

(check one)

Been given a ticket for
public intoxication, drunk
and disorderly or other
non-driving alcohol arrest

Had the D.T.'s (delirium
tremens, shakes, sweating,
rapid heart, etc.)

within 2 - 3 days

after stopping drinking

Found that I had a strong
craving for a drink at
some time each day

Needed to drink a lot more
in order to get an effect,
or found that I no longer

could get high on the amount
I used to drink

Found that I was able to
drink a lot more than I used
to before I would get drunk

Had days where I drank
much more that I expected
to when I began

Found that I often
continued drinking for
more days in a row
than I had planned to

Found that I tended to
gulp my drinks rather
than just drink them

Been arrested for a drinking
related offense

180

NO

IN THE LAST

3 YEARS: HOW

MANY TIMES
(Use key)‘+

JUST IN THE
LAST YEAR--
Last 12 months.
HOW MANY
TIMES?
(Use key)‘

30. Been court ordered to get

31.

IN THE LAST JUST IN THE
3 YEARS: HOW LAST YEAR--
MANY TIMES Last 12 months.

(Use key)*+ HOW MANY
YES NO TIMES?
(check one) (Use key)‘

alcohol treatment

Been put on probation or
parole for a drinking related
offense.

THE LAST SECTIONS OF THIS QUESTIONNAIRE DEAL WITH YOUR USE OF
VARIOUS DRUGS OTHER THAN ALCOHOL. WE HOPE THAT YOU CAN ANSWER
ALL QUESTIONS; BUT IF YOU FIND ONE WHICH YOU FEEL YOU CANNOT
ANSWER HONESTLY, WE WOULD PREFER THAT YOU LEAVE IT BLANK.

REMEMBER THAT YOUR ANSWERS WILL BE KEPT STRICTLY CONFIDENTIAL
AND THEY ARE NEVER CONNECTED WITH YOUR NAME. THAT IS WHY THIS
QUESTIONNAIRE IS IDENTIFIED ONLY WITH A CODE NUMBER.

THE FOLLOWING QUESTIONS ARE ABOUT CIGARETTES (CHECK THE BEST
ANSWER):

1.

HAVE YOU SMOKED CIGARETTES DURING THE PAST 3 YEARS?

Never (GO TO SECTION G on page 13)

Once or twice

Occasionally but not regularly

Regularly for a while during this year, but not now.
Regularly now

HAVE YOU SMOKED CIGARETTES DURING THE PAST 12 MONTHS?

Never (GO TO QUESTION 4)

Once or twice

Occasionally but not regularly

Regularly for a while during this year, but not now.
Regularly now

181

HOW FREQUENTLY HAVE YOU SMOKED CIGARETTES DURING THE PAST 30

DAYS?

Not at all

Less than one cigarette per day

One to five cigarettes per day

About one-half pack per day

About one pack per day

About one and one-half packs per day
Two packs or more per day

(ANSWER QUESTIONS 4-9 FOR THE MOST RECENT TIME YOU HAVE BEEN
SMOKING.)

4.

How soon after you wake up do you smoke your first cigarette?

Within 5 minutes ...................................
6—30 minutes .......................................
31-60 minutes ......................................
After 60 minutes ....................................

Which cigarette would you hate most to give up?

The first one in the morning ...........................
Any others ........................................

How many cigarettes a day do you smoke?

10 or less ..........................................
11-20 .............................................
21-30 .............................................
31 or more ........................................

Do you find it difficult to refrain from smoking in places where it is
forbidden, such as in church, the library, or the theater?

Do you smoke more frequently during the first hours after waking
than during the rest of the day?

Do you smoke if you are so ill that you are in bed most of the day?

182

(Circle one answer)

............... 3
............... 2
............... l
............... 0
(Circle one answer)
............... 1
............... 0
(Circle one answer)
............... 0
............... 1
............... 2
............... 3
(Circle one answer)
YES NO
1 0
l O
1 0

))) ))) ))) )))

380 c0 80 ( ( ( 2,80 0 80 ( ( ( 280 s0 80 ( ( ( 26% s0 80 ( ( (
08:88 88000 H U 0 82088 80-000 0 U 0 82088 81000 H U U 82088 0001000 U U 0
82288 30.2: H U U 08288 000.03 0 U 0 82088 2.0.02 0 U 0 82868 03.02 0 U 0

\I \.l ) \l \I \I \I ) \I \I/ \I )

mcowmaouo 00-0? mGOMmmUUO OQIOV mcomwmooo QOIOV mGOMmmUUO GGIOV

))) ))) ))) )))

mcowmduoo QMION mzommmooo OMION mGOmmmuuo GMuON mQOmwaooO GMION

((( ((( ((( (((

\II ) ) ) ) \II \II ) )
82088 2-2 ( ( ( 220080 3-2 ( ( ( 0:20.88 2-2 ( ( ( 0:20.88 2.3 U \(1 0
0:20.88 a6 ) ) ) 22088 To \1 \I ) 0:20—88 To ) x} ) 0:20.080 m6 ) ) )

((( ((( ((( (((

0:20.080 Wm ) ) ) 0:20.88 Wm ) x) ) 008888 m-m ) ) \1 0028038 Wm ) ) )

G. THE FOLLOWING QUESTIONS ARE ALL ABOUT NON-PRESCRIPT ION USE OF
DRUGS, EITHER FOR RECREATION OR FOR SELF-MEDICATION

 

(MARK ONE SPACE FOR EACH LINE).

0:20.080 N;

mcommmuoo 0

ON HOW MANY OCCASION
(IF ANY) HAVE YOU USED

MARIJUANA (GRASS, POT)

1.

OR HASHISH (HASH, HASH

OIL)

(((

)))

(((

(((

During the last 3 years?
During the last 12 mos?
During the last 30 days?

0:20.080 m;

(IF ANY)HAVE YOU USED

ON How MANY OCCASION
LSD (ACID).

2.

)))

(((

)))

(((

During the last 12 mos?

During the last 3 years?

0:20.88 N;

PSYCHEDELICS OTHER THAN

LSD (LIKE MESCALINE,
PEYOTE, PSILOCYBIN, PCP)

(IF ANY) HAVE YOU USED

ON HOW MANY OCCASIONS

During the last 30 days?

3.

(((

)))

(((

)))

(((

During the last 30 days?

During the last 3 years?
During the last 12 mos?

320080 N;

COCAINE (COKE, CRACK, ROCK

(IF ANY) HAVE YOU USED
COCAINE, SNOW)

ON HOW MANY OCCASIONS

4.

(((

))\II

(((

)))

(((

During the last 3 years?
During the last 12 mos?
During the last 30 days?

183

4a. Which of the following is the way that most accurately describes how you use coke?
(Please circle only one answer)

All or mostly nasal (snorting).

All or mostly smoking crack.

All or mostly freebase.

Both nasal and smoking crack.
Both nasal and freebase.

Both smoking crack and freebase.

reap-9.0:!»

(MARK ONE SPACE FOR EACH LINE)

5.
AMPHETAMINES ARE SOMETIMES PRESCRIBED BY DOCTORS TO HELP PEOPLE

LOSE WEIGHT OR TO GIVE PEOPLE MORE ENERGY. THEY ARE SOMETIMES
CALLED UPPERS, UPS, SPEED, CRYSTAL, CRANK, BENNIES, DEXIES, PEP PILLS

AND DIET PILLS.

ON HOW MANY OCCASIONS
(IF ANY) HAVE YOU TAKEN
AMPHETAMIN ES

ON YOUR OWN--THAT IS,
WITHOUT A DOCTOR
TELLING YOU TO TAKE THEM.

0 occasions

1-2 occasions

3-5 occasions

6-9 occasions
10-19 occasions
20-39 occasions
40-99 occasions
100-249 occasions
250-499 occasions
500 & above

During the last 3 years?
During the last 12 mos?
During the last 30 days?

AAA
VVV
AAA
VVV
AAA
vvv
AAA
vvv
VVV
AAA
vvv
VVV
AAA
AAA

6.
ON HOW MANY OCCASIONS
(IF ANY) HAVE YOU USED
QUAALUDES (QUADS,
SOAPERS, METHAQUALONE)
ON YOUR OWN--THAT IS,
WITHOUT A DOCTOR
TELLING YOU TO TAKE THEM.

0 occasions

1-2 occasions

3-5 occasions

6-9 occasions
10-19 occasions
20-39 occasions
40-99 occasions
100-249 occasions
250-499 occasions
500 & above

During the last 3 years?
During the last 12 mos?
During the last 30 days?

AAA
vvv
AAA
vvv
AAA
vvv
AAA
vvv
AAA
vvv
vvv
AAA
vvv
AAA
VVV
AAA
vvv
AAA

184

vvv

VVV

7.
BARBITURATES ARE SOMETIMES PRESCRIBED BY DOCTORS TO HELP PEOPLE

RELAX OR GET TO SLEEP. THEY ARE SOMETIMES CALLED DOWNS, DOWNERS,
GOOFBALLS, YELLOWS, REDS, BLUES, RAINBOWS.

ON HOW MANY OCCASIONS
(IF ANY) HAVE YOU TAKEN
BARBITURATES

ON YOUR OWN--

THAT IS, WITHOUT A DOCTOR
TELLING YOU TO TAKE THEM.

1-2 occasions

3-5 occasions

6-9 occasions
10-19 occasions
20-39 occasions
40-99 occasions
100-249 occasions
250-499 occasions
500 & above

During the last 3 years?
During the last 12 mos?
During the last 30 days?

v v v 0 occasions

AAA
vvv
AAA
VVV
AAA
VVV
AAA
vvv
AAA
AAA
Vvv
AAA
VVV
AAA
vvv
AAA

(MARK ONE SPACE FOR EACH LINE)

8.
TRANQUILIZERS ARE SOMETIMES PRESCRIBED BY DOCTORS TO CALM PEOPLE

DOWN, QUITE THEIR NERVES, OR RELAX THEIR MUSCLES. LIBRIUM, VALIUM,
AND MILTOWN ARE ALL TRANQUILIZERS.

ON HOW MANY OCCASIONS

(IF ANY) HAVE YOU TAKEN g g g g g
TRANQUILIZERS m g g g .g .5 .3 g g g
ON YOUR OWN--THAT IS, .3 g g g g 8 8 E E _g
WITHOUT A DOCTOR TELLING g 8 8 8 g g g g a? .3;
YOU TO TAKE THEM. 8 c3 .3. 3 g g g g‘ a g
o ._'. «'3 )6 v— N <1- " N m
Duringthelastsyears? ()(>()()() ()()()()()
Duringthelastlzmos? ()()()()()()<)()()()
Dufingthelast30days? () () () () () () () () () ()
9. g g a .S .6
ON How MANY OCCASIONS g g g .3 .g .g § § 8
(IF ANY) HAVE YOU USED § 8, g 2% 8 8 8 8 8 g
HEROIN(SMACK, 'g :3 § § 3 E § § g g;
HORSE, SKAG. 8 ° ° '7 8: °.~ <5 8
) g 2 2 2 <2 8 8 2 8 §
Dufingthelast3years? () () () () () () () () () ()
DuringthelastlZmOS‘? () () () () () () () () () ()
Duringthelast30days? () () () () () () () () () ()

185

 

10.

THERE ARE A NUMBER OF NARCOTICS OTHER THAN HEROIN, SUCH AS

METHADONE, OPIUM, MORPHINE, CODEINE, DEMEROL, PAREGORIC, TALWIN,

AND LAUDANUM. THESE ARE SOMETIMES PRESCRIBED BY DOCTORS.

ON HOW MANY OCCASIONS
(IF ANY) HAVE YOU TAKEN
NARCOTICS OTHER

THAN HEROIN ON YOUR OWN
- THAT IS,

WITHOUT A DOCTOR
TELLING YOU TO TAKE

0 occasions

During the last 3 years?
During the last 12 mos?
During the last 30 days?

AAA
VVV

11.
ON HOW MANY OCCASION
(IF ANY) HAVE YOU SNIFFED
GLUE, OR BREATHED THE
CONTENTS OF AEROSOL
SPRAY CANS, OR INHALED
ANY OTHER GASES OR
SPRAYS IN ORDER

TO GET HIGH

0 occasions

During the last 3 years?
During the last 12 mos?
During the last 30 days?

AAA

AAA

AAA

1-2 occasions

vvv

1-2 occasions

VVV

AAA

AAA

186

3-5 occasions

VVV

3-5 occasions

vvv

AAA

AAA

6-9 occasions

6-9 occasions

vvv

AAA

AAA

10-19 occasions

10-19 occasions

vvv

vvv

AAA

AAA

20-39 occasions

20-39 occasions

vvv

vvv

40-99 occasions

AAA

40-99 occasions

AAA

vvv

vvv

100-249 occasions

AAA

100-249 occasions

AAA

VVV

VVV

250-499 occasions

250-499 occasions

AAA

vvv

vvv

500 & above

500 & above

AAA

VVV

VVV

ANSWER KEY FOR QUESTIONS BELOW:
1 2 3-5 6-10 11-20 21-50 51-100 101-250 251-500 500+(morethan500)

H. NOW SOME QUESTIONS ABOUT NONPRESCRIPTION USE OF DRUGS.
DURING THE LAST 3 YEARS, HAVE YOU HAD ANY OF THE FOLLOWING
OUTCOMES BECAUSE OF YOUR USE OF NONPRESCRIPTION DRUGS ASKED
ABOUT IN SECTION G. (The last section). (Use the answer key at the top of page for

column 3 & 4)
IN THE LAST JUST IN THE
3 YEARS: HOW LAST YEAR-last
YES NO MANY TIMES 12 months. HOW
(check one) (Use key)” MANY TIMES?
(Use key)‘

I. Missed school or time on job
Lost friends
Been divorced or separated

Been fired or laid off

.V‘PP’!”

Had a car accident when
you were driving

6. Had to go to a hospital
(other than accidents)

7. Had to stay in hospital
overnight

8. Had to see a doctor because
of drug use (unintentional
overdose) or had a doctor say
drugs had harmed your health

9. Gone through physical
withdrawal from drugs

10. Been arrested more than
once for possession or sale
of drugs other than marijuana

10a. Been arrested for any drug
related offense

10b. Been court ordered to get
substance abuse treatment

10c. Been put on probation
or parole for a drug related
offense

187

lla. DURING THE PAST 3 YEARS HAVE YOU TAKEN DRUGS INTRAVENOUSLY
(USING A NEEDLE)? DON'T COUNT SHOTS YOU WERE GIVEN BY A
DOCTOR OR NURSE OR SHOTS YOU MAY HAVE TAKEN FOR TREATMENT
OF DIABETES.

NO YES
IF YES, WHAT ABOUT IN THE LAST 12 MONTHS?
llb. NO YES

1 1c. IF YES, (to Ila or 11b) WHAT DRUGS HAVE YOU TAKEN INTRAVENOUSLY
(IV)?

SECTION I. TREATMENT PROGRAMS

These next questions ask about your experiences getting help for problems with drugs or
alcohol during the past three years.

1. During the past 3 years, have you been in a formal treatment program for alcohol or
drug problems?
(Circle One)
Yes ................................................... 1
No .................................................... 2
la. How many times? ............................ times.

 

(If none, enter 0)

lb. What were your age(s) when you were in a formal program
for alcohol or drug problems? (e. g. 34-36, 27) ...... age(s).

2. During the past 3 years have you attended a self-help group like Alcoholics
Anonymous (AA) or Narcotics Anonymous (NA) or another similar self-help group
for alcohol or drug problems?

(Circle One)
Yes .............................................. 1
No ........................................ 2 [Go to Question 3]
2a. What were your age(s) when you attended such a group? age(s).

2b. Overall, how many times have you attended such a group? (Circle One)
1-2 3-5 6-10 11~

2c. When was the most recent time? .......... ,
month year

188

2d. Have you asked for or received sponsorship in AA at any time?

(Circle One)
Yes ................................................... 1
No .................................................... 2

During the past 3 years have you been in an outpatient therapy, or other formal
treatment program for emotional or mental health problems other than drugs or
alcohol?

(Circle One)

Yes ................................................... 1

No .................................................... 2
3a. How many different times during the past 3 years? times

(If none, enter 0)
3b. What were your age(s) when you were in such therapy? .....
age(s).

3c. Overall, during the past three years, how many sessions of treatment have you

had? ...................................... sessions

 

189

DEMOGRAPHIC BACKGROUND INFO (PARENT)
MSU-UM Family Project; 5/01

Background Information
We would like to ask you a few questions about yourself. The questions ask about
your life during the time you were growing up as well as now. Please answer a_ll of

them as completely as possible. (PLEASE PRINT)

1. What is your date of birth?

 

 

 

MONTH DAY YEAR

2. Where were you born?

 

CITY/T OWN (COUNTY IF RURAL) STATE COUNTRY (IF NOT U.S.)
Number 3 intentionally left out.

4. Until you were 18, about how many times did your family move.
CIRCLE ONE 0 1 2 3 4 5 6 7 or more.

5a. Did you live together with both of your natural parents for most of the time from birth to
18? CIRCLE ONE.
YES (If Yes, go to question 6) NO (If No, go to question 5b)

5b. What was the main reason your parents did not live together with you during that time?
CIRCLE ONE

Mother died

Father died

Both parents died

Parents divorced or separated

Parents never lived together

Other (Please explain)

9999!"?

 

5c. Which adult(s) did you live with [rm of the time from birth to 18? CIRCLE ONE
Mother, but no adult male

Father, but no adult female

Mother and step-father

Father and step-mother

Other (Please explain)

MPPN?‘

 

6. Who was the main wage earner in your home while you were growing up?
CHECK ONE
a) your father
b) your mother
c) someone else What was their relationship to you

190

ABOUT YOUR NATURAL (BIOLOGICAL) FATHER

7a. Where was be born?
State Country (If not U.S.A.)

 

ABOUT THE ADULT MALE WHO LIVED WITH YOU MOST OF THE TIME UNTIL
YOU WERE 18. (This could be your natural father, or stepfather, or someone else.)

7b. What kind of work did this adult male do (the adult male who lived with you most of the
time until you were 18?) That is what was his occupation?

 

(For example: electrical engineer, machinist, stock clerk, assembly line worker, farmer)

7c. What were his most important activities or duties?

 

(For example: keep account books, filing, selling cars, operate printing press, finish
concrete)

7d. What kind of business or industry was this?

 

(For example: TV and radio mfg., retail shoe store, automobile manufacturing,
[Oldsmobile], State Labor Dept, farm work.

7e. What was the highest grade of school he completed? CIRCLE THE HIGHEST GRADE

COMPLETED
None 0
Elementary 1 2 3 4 5 6 7 8
High School 9 10 11 12
College 1 2 3 4 Degree?
Graduate School 5 6 7 8+ Degree?

7f. Would your father identify as Latino/Hispanic/Spanish? CIRCLE ONE
a) No
b) Yes: Mexican, Mexican American, Chicano
c) Yes: Puerto Rican
d) Yes: Cuban, Cuban-American
e) Yes: Central American
f) Yes: Other Latino/Hispanic/Spanish group (print group):

 

7g. Which of the following best identifies your father’s race? CIRCLE ONE
a) White
b) Black, African American

c) Native American, American Indian
d) Asian American, Pacific Islander
e) Some other race (please print):

 

191

ABOUT YOUR NATURAL (BIOLOGICAL) MOTHER

8a. Where was she born?
State Country -- If not U.S.A.

 

 

ABOUT THE ADULT FEMALE WHO LIVED WITH YOU MOST OF THE TIME UNTIL
YOU WERE 18.
(This could be your natural mother, or stepmother, or someone else.)

8b. What kind of work did this adult female do (the adult female who lived with you most of
the time until you were 18?) That is what was her occupation?

 

(For example: electrical engineer, file clerk, assembly line worker, bookkeeper, sales clerk)

8c. What were her most important activities or duties?

 

(For example: keep account books, filing, selling clothes, teach fifth graders)

8d. What kind of business or industry was that?

 

(For example: TV and radio mfg, retail shoe store, automobile manufacturing [Oldsmobile],
state labor dept.)

8e. What was the highest grade of school she completed? CIRCLE THE HIGHEST

GRADE COMPLETED
None 0 .
Elementary 1 2 3 4 5 6 7 8
High School 9 10 11 12
College 1 3 4 Degree?
Graduate School 5 6 7 8+ Degree?

AGAIN, A QUESTION ABOUT YOUR NATURAL (BIOLOGICAL) MOTHER:

8f. Would your mother identify as Latino/Hispanic/Spanish? CIRCLE ONE
a) No
b) Yes: Mexican, Mexican American, Chicano
c) Yes: Puerto Rican
d) Yes: Cuban, Cuban-American
e) Yes: Central American
0 Yes: Other Latino/Hispanic/Spanish group (print group):

 

192

 

8g. Which of the following best identifies your mother’s race? CIRCLE ONE

a) White

b) Black, African American

c) Native American, American Indian
d) Asian American, Pacific Islander
e) Some other race (please print):

 

9x. Do you identify as Latino/I-Iispanic/Spanish? CIRCLE ONE
a) No
b) Yes: Mexican, Mexican American, Chicano
c) Yes: Puerto Rican
(1) Yes: Cuban, Cuban-American
e) Yes: Central American
f) Yes: Other Latino/I-Iispanic/Spanish group (print group):

 

9y. Which of the following best identifies your race? CIRCLE ONE
a) White
b) Black, African American

 

c) Native American, American Indian
d) Asian American, Pacific Islander
e) Some other race (please print)
92. Until you were 18, what religion was practiced in your home most of the time?
CIRCLE ONE
88 None 21 Episcopalian
1 Buddhist 22 Full Gospel (Tabernacle)
2 Christian Scientist 23 Fundamentalist
3 Hindu 24 Lutheran
4 Islam 25 Methodist
5 Jehovah's Witness 26 Moravian
6 Jewish 27 Nazarene
7 Mormon (Latter Day Saints) 28 Pentecostal
8 Orthodox (Eastern, Greek, Russian, etc.) 29 Presbyterian
12 Other Eastern (e.g. Shinto, Taoism) 3l Quaker
13 Roman Catholic 32 Reformed Church
14 Assembly of God 33 Dutch Reformed Church
15 Baptist 34 Seventh Day Adventist
16 Church of Brethren 35 Unitarian
17 Church of Christ 36 United Brethren
18 Church of God 37 Wesleyan
l9 Congregational 98 Other (name)

20 Disciples of Christ

 

(Name of "other")

9c. Until you were 18, how often did you attend religious services? CIRCLE ONE

several times a week
about once a week
2-3 times a month
once a month or less
never

MPPN.“

193

102.

Nu—Iu—d—‘u—I—le—l—n—o m
ocmuomuwmch‘V‘AWN—‘oo

lOc.

10d.

11.

12a.

Please circle the denomination/type of church that best represents your religious
preference now.

None 21 Episcopalian

Buddhist 22 Full Gospel (Tabernacle)
Christian Scientist 23 Fundamentalist

Hindu 24 Lutheran

Islam 25 Methodist

Jehovah's Witness 26 Moravian

Jewish 27 Nazarene

Mormon (Latter Day Saints) 28 Pentecostal

Orthodox (Eastern, Greek, Russian, etc.) 29 Presbyterian

Other Eastern (e.g. Shinto, Taoism) 31 Quaker

Roman Catholic 32 Reformed Church
Assembly of God 33 Dutch Reformed Church
Baptist 34 Seventh Day Adventist
Church of Brethren 35 Unitarian

Church of Christ 36 United Brethren

Church of God 37 Wesleyan
Congregational 98 Other (name)

Disciples of Christ

 

(Name of ”other")

About how often did you attend religious services in the Lagvear? CIRCLE ONE
several times a week

about once a week

2-3 times a month

once a month or less

never

MPP‘NT‘

Regardless of your attendance at religious services, how religious do you consider
yourself to be?

not religious at all

not very religious

fairly religious

very religious

PENN?"

What is the highest grade of school you have completed? CIRCLE THE HIGHEST
GRADE COMPLETED.

 

None 0
Elementary l 2 3 4 5 6 7 8
High School 9 10 11 12

POST HIGH SCHOOL
College 1 2 3 4 Degree
Graduate School 5 6 7 8+ Degree
Vo-Tech School 1 2 3 4 Certificate

What kind of work are you doing now? (What is your occupation?)

 

 

(For example: Electrical engineer, machinist, stock clerk, assembly line worker, teacher,
farmer).

194

12b. What are your most important activities or duties?

 

(For example: keep account books, filing, selling cars, operate printing press, finish concrete,
teach fifth graders, answer phone).

12c. What kind of business or industry is this?

 

(For example: TV and radio manufacturing, retail shoe store, automobile manufacturing
[Oldsmobile], State Labor Department, farm work)

12d. Are you: (Check one)
An employee of a PRIVATE company, business or individual [who works]
for wages, salary or commission?

A GOVERNMENT employee (federal, state, county, or local government?
Self-employed in OWN business, professional practice, or farm?

own business not incorporated

own business incorporated

working without pay in a family business or farm

12e. Approximately what is your present annual family income?

CIRCLE ONE
1. $4,000 or under 6. $16,001 -- $20,000
2. $ 4,001 -- $ 7,000 7. $20,001 -- $30,000
3. $ 7,001 -- $10,000 8. $30,001 -- $50,000
4. $10,001 -- $13,000 9. $50,001 -- $75,000
5. $13,001 -- $16,000 10. $75,000 -- $100,000
11. Over $100,000
12e1. How often do you have problems paying for basic necessities like food, clothing and
rent?
1. Hardly ever 2. Sometimes 3. Often

Compared to other people, do you have enough money to pay for:

More than enough Just enough Not Enough
12e2a. The food you need? 1 2 3
12e2b. The clothing you need? 1 2 3
12e2c. The medical care you need? 1 2 3

12e3. How would you describe your family’s money situation while you were growing up?
1. Very poor, not enough to get by.
2. Had enough to get by, but that’s all.
3. Had more than enough to get by.

195

 

12f. How many months out of the last 3 years have you been without a regular paid job? (DO
count months you were retired, in school full-time, a home maker or too ill to work)
(Your answer may range from 0-36). months.

12g. Please list the jobs you have had in the last three years as well as the periods during
which you were not working. Start with your current iob (or if not working, your current
activities) and work backwards. We do not need to know who your employer was, but
list your approximate dates of employment and what type of work you were doing. For
each different employment, list (1) type of work/occupation and (2) most important job
duties.

DATES OF EMPLOYMENT TYPE OF WORK/OCCUPATION
month/year month/year

(a) to

 

 

 

 

 

 

(b) to
(c) to
(d) to

 

 

NOW ABOUT YOUR MARITAL STATUS

13. How many times have you been married? CIRCLE ONE
0 1 2 3 4+

13a. Which answer best fits your current marital situation? CIRCLE ONLY ONE

1. Married or living a partner
2. Divorced
3. Separated

14a. What was the date of your marriage to your (present) spouse?

 

14b. If married more than once, what was the date of your first marriage?

 

196

NOW SOME QUESTIONS ABOUT ALL YOUR CHILDREN

15a. List all biological children (children born to you) from all relationships including your
current marriage or relationship, as well as all previous ones. * "Lives with you now" means
with you all the time or most of the time. If you are primary custodial parent or share custody
equally, circle yes ("Y") for "lives with you now."

 

 

 

 

 

 

 

 

FIRST NAME ONLY BIRTH DATE SEX LIVES WITH DECEASED
mo/day/yr (circle one) YOU NOW“ (GIVE
DATE)
(circle one)

1. M / F Y / N

2. M / F Y / N

3. M / F Y / N

4. M / F Y / N

5. M / F Y / N

 

 

15b. Now circle the names of the biological children who are from your present marriage or
relationship. If afl are from your present marriage or relationship, mark a check here

15c. Now list all the other (nonbiological) children you have from another maniage or
relationship.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

FIRST NAME BIRTH DATE SEX TYPE OF RELATIONSHIP
ONLY mo/day/year (circle one) (step, adopted, foster, relative, etc.)
1. M / F

2. M / F

3. M / F

4. M / F

5. M / F

6. M / F

7. M / F

8. M / F

9. M / F

10. M / F

 

 

197

ANTISOCIAL BEHAVIOR CHECKLIST (PARENT)
MSU-UM Family Study (9/99)

Many of uS have had adventures during our lives...times that were exciting and carefree, even
though they may have been a bit impulsive or happy-go-lucky. Please read each of the following
items. Indicate (with a check) if you have ever done any of the following activities and how
often.

 

N R S O

E A O F NEVER - You have never done this

V R M T

E E E E RARELY - Once or twice in your life

R L T N

Y I SOMETIMES - Three (3) to nine (9) times in your life

M
E OFTEN - More than ten (10) times in your life
S

 

 

l. Skipped school without a legitimate excuse for more than 5 days in one

school year.

 

. Been suspended or expelled from school for fighting.

 

. Been suspended or expelled from school for reasons other than fighting.

 

. Lied to a teacher or principal.

 

. Cursed at a teacher or principal (to their face).

 

 

. Repeated a grade in school.

 

. Taken part in a gang fight.

 

2
3
4
5
6. Hit a teacher or principal.
7
8
9

. "Beaten up" another person.

 

10. Broken street lights, car windows, or car antennas just for the fun of it.

 

11. Gone for a ride in a car someone else stole.

 

12. Teased or killed an animal (like a dog or cat) just for the fun of it.

 

13. Defied your parent’s authority (to their face).

 

14. Hit your parents.

 

15. Cursed at your parents (to their face).

 

l6. Stayed out overnight without your parent’s permission.

 

 

 

 

 

 

 

198

 

17.

Run away from home for more than 24 hours.

 

l8.

Lied to your parents.

 

l9.

Snatched a women’s purse.

 

20.

Rolled drunks just for the fun of it.

 

21.

Shoplifted merchandise valued over $25.

 

22.

Shoplitted merchandise valued under $25.

 

23.

Received a speeding ticket.

 

24.

Been questioned by the police.

 

25.

Taken part in a robbery.

 

26.

Taken part in a robbery involving physical force or a weapon.

 

27.

Been arrested for a felony.

 

28.

Resisted arrest.

 

29.

Been arrested for any other non-traffic police offenses (except fighting

or a felony).

 

30.

Been convicted or any non-traffic police offense.

 

31.

Defaulted on a debt.

 

32.

Passed bad checks for the fun of it.

 

33.

Ever used an alias.

 

34.

Gone AWOL from the military.

 

35.

Received a bad conduct or undesirable discharge from the military.

 

36.

Performed sexual acts for money.

 

37.

This item was deleted.

 

38.

Had intercourse with more than one person in a Single day.

 

39.

“Fooled around” with other women/men after you were married.

 

40.

Hit your husband/wife during an argument.

 

41.

Lied to your spouse.

 

 

 

 

 

 

42.

Spent six months without any job or permanent home.

 

199

 

 

 

43. Been fired for excessive absenteeism.

 

44. Been fired for poor job performance (except absenteeism).

 

45. Changed jobs more than 3 times in one year.

 

 

 

 

 

 

46. Lied to your boss.

 

Thank you very much for your cooperation.

200

 

HEALTH HX-WOMEN (MOTHER)
Pre-natal 3/99

HEALTH HISTORY QUESTIONNAIRE - FORM W

The following questions are mainly about your medical history, health history and health habits.
At the start, there also are some questions about your child , that ask about
the pregnancy and early developmental history. Please complete each item carefully. If you
have questions about any item, ask the interviewer. Remember that all information is
confidential and will not be disclosed to anyone.

 

 

 

 

1. Pregnancy History
During your pregnancy with , did you:
1. Ever have high blood pressure? ................... YES( ) NO( )
2. Have diabetes, or have sugar in your urine? ......... YES( ) NO( )
3. Have albumin or protein in your urine? ............. YES( ) NO( )
4. Have toxemia? ................................ YES( ) NO( )
5. Have any infections? ........................... YES( ) NO( )
If yes, please
specify
6. Have German (3 day) measles? ................... YES( ) NO( )
7. Take medicines prescribed by your doctor? .......... YES( ) NO( )

If yes, what medications?
8. Did you smoke cigarettes? ....................... YES( ) NO( )
If yes, about how many cigarettes a day? per day

 

 

 

 

9. Have a venereal disease such as gonorrhea, syphilis or herpes? .......
YES( ) NO( ) DON'T KNOW( )
If yes, please specify
10. Did you drink alcoholic beverages? ................ YES( ) NO( )
If yes, about how many drinks per day
per week
11. Did you use any nonprescription drugs? ............ YES( ) NO( )

If yes, what drugs?

 

201

12. During your pregnancy with , did you threaten to miscarry or have
premature labor? ............................... YES( ) NO( )

If yes, please explain

 

 

13. Get hurt or injured? ............................ YES( ) NO( )

If yes, please explain

 

 

14. Have Rh or other blood group incompatibility?
YES( ) NO( ) DON'T KNOW( )
15. Have other problems, diseases or conditions? ........ YES( ) NO( )

If yes, please explain

 

 

 

 

16. How long was your pregnancy? months.

17. How early did you start seeing a doctor? Starting at _months.

18. What was your child's weight at birth? ...... lb. oz.
19. Was your labor longer than 12 hours? .............. YES( ) NO( )
20. Was your labor less than 2 hours? . . . .............. YES( ) NO( )
21. Did you have a difficult deliver? .................. YES( ) NO( )

If yes, please explain

 

22. Was it a breech (bottom first) deliver? .............. YES( ) NO( )
23. Was it a caesarean delivery? ..................... YES( ) NO( )
24. Did you have a multiple birth (twins or triplets)? ..... YES( ) NO( )
25. Were you given an anaesthetic for the delivery? ...... YES( ) NO( )
If so, what

 

26. Have you had premature births, miscarriages or stillbirths?
..... YES( ) NO( )

202

H. Delivery and Newborn History During

l.
2.
3.

During

 

's delivery:

 

 

 

10.
11.
12.

13.

14.

 

 

 

 

 

 

 

Was she born with the cord around her neck? ........ YES( ) NO( )
Was she injured during birth? .................... YES( ) NO( )
Was anything wrong with your child at birth? ........ YES( ) NO( )
If yes, what?
'8 newborn period (4 weeks): did she:

Have any breathing problems? .................... YES( ) NO( )
If yes, please explain
Need to receive oxygen? ........................ YES( ) NO( )
Turn blue (cyanosis)? ........................... YES( ) NO( )
Turn yellow (jaundice)? ......................... YES( ) NO( )

If yes, did she receive: blood transfusions . . . . YES( ) NO( )

phototherapy (lights) ................. YES( ) NO( )

Have any infections? ........................... YES( ) NO( )
If so, what were they?
Receive medication? ........................... YES( ) NO( )
If so, what kind?
Have seizures (fits, convulsions)? ................. YES( ) NO( )
Have feeding problems? ......................... YES( ) NO( )
Was born with any birth defects? ........ YES( ) NO( )
If so, what
Did have any other problems? .......... YES( ) NO( )
If yes, please explain
Was she born in a hospital? ...................... YES( ) NO( )
If yes, what hospital?

 

address: (city and state)

203

15.

16.

17.
18.

What kinds of doctor(s) or clinic(s) have provided your child's health care?

 

 

 

 

 

 

 

 

Up to what age was your child breast-fed?
( ) My child was not breast-fed

( ) My child was breast-fed until the age of months.
Have you had any premature births? ............... YES( ) NO( )
Have you had a_ny caesarean births? ............... YES( ) NO( )

204

DRINKING AND OTHER DRUG USE (YOUTH)
(1/99)

This questionnaire takes about 20 minutes. All information will be used for research only and
will be kept strictly confidential. If you are not sure of the answer to a question, please answer
the best you can. Please try to answer each item. These questions are to find out how you feel
about drinking, drug use, and other topics having to do with your attitudes and behavior. Please
remember that no one will see your answers except members of the resgarch sta_ff.

A. THE FOLLOWING QUESTIONS ASK ABOUT YOUR EXPERIENCE DRINKING
ALCOHOLIC BEVERAGES, (BEER, WINE, AND LIQUOR):

1. HOW OLD WERE YOU THE FIRST TIME YOU EVER TOOK A DRINK?
DO NOT COUNT THE TIMES WHEN YOU WERE GIVEN A "SIP" BY AN
ADULT.

years old.

 

 

IF YOU'VE NEVER TAKEN A DRINK AT ALL, GO TO PAGE 11
(SECTION E), QUESTION 1.

 

2. OVER THE LAST 6 MONTHS, ON THE AVERAGE, HOW MANY DAYS A
MONTH HAVE YOU HAD A DRINK days a month.

 

2a. DURING THE 6 MONTHS BEFORE THAT PERIOD. ON THE AVERAGE,
HOW MANY DAYS A MONTH DID YOU HAVE A DRINK?
days a month.

 

3. OVER THE LAST 6 MONTHS, ON A DAY WHEN YOU ARE DRINKING,
HOW MANY DRINKS DO YOU USUALLY HAVE IN 24 HOURS? (A
DRINK IS A 12 OZ. CAN, GLASS OR BOTTLE OF BEER; A 4 OZ. GLASS
OF WINE; A 12 OZ. WINE COOLER BOTTLE; A SINGLE SHOT; OR A
SWGLE "MIXED DRINK.") drinks per 24 hours.

 

3a. DURING THE 6 MONTHS BEFORE THAT PERIOD ,ON A DAY WHEN
YOU WERE DRINKING, HOW MANY DRINKS DID YOU USUALLY
HAVE IN 24 HOURS? (A DRINK IS A 12 OZ. CAN, GLASS OR BOTTLE OF
BEER; A 4 OZ. GLASS OF WINE; A 12 OZ. WINE COOLER BOTTLE; A
SINGLE SHOT; OR A SINGLE "MIXED DRINK")

drinks per 24 hours.

 

205

4. OVER THOSE 6 MONTHS. WHEN YOU GOT DRUNK, HOW BAD WAS YOUR

HANGOVER?

__ Never bad
Not bad Pretty Bad
A little less than average Terrible
Average Worst possible

A little more than average Never drank enough to get hangover

 

IF YOU DRANK NO ALCOHOLIC BEVERAGES AT ALL (NOT EVEN A FEW SIPS) IN
THE LAST 6 MONTHS, GO TO PAGE 5, QUESTION 5.

 

B. THE FOLLOWING QUESTIONS ARE ABOUT YOUR DRINKING PATTERNS. IN
ANSWERING THE QUESTIONS, PLEASE THINK ABOUT WHAT YOU HAVE
DONE ON THE AVERAGE OVER THE LAST SIX MONTHS.
1. WHEN DRINKING BEER

a. HOW OFTEN DO YOU USUALLY HAVE BEER?

3 or more times a day 2 or 3 times a month

2 times a day About once a month

Once a day Less than once a month,

Nearly every day but at least once a year

3 or 4 times a week Less than once a year

Once or twice a week NEVER [If checked, go to
question #2a]

b. THINK OF ALL THE TIMES YOU HAD BEER RECENTLY, WHEN
YOU DRINK BEER, HOW OFTEN DO YOU HAVE 10 OR MORE
CANS, GLASSES OR BOTTLES?

Nearly every time: SKIP TO QUESTION #2 BELOW

More than half the time: SKIP TO QUESTION #2 BELOW
Less than half the time

Once in a while

NEVER

0. WHEN YOU DRINK BEER, HOW OFTEN DO YOU HAVE AS
MANY AS 7 TO 9 CANS, GLASSES OR BOTTLES?

Nearly every time: SKIP TO QUESTION #2 BELOW

More than half the time: SKIP TO QUESTION #2 BELOW
Less than half the time

Once in a while

NEVER

206

d. WHEN YOU DRINK BEER, HOW OFTEN DO YOU HAVE AS
MANY AS 5 TO 6 CANS, GLASSES OR BOTTLES?

Nearly every time: SKIP TO QUESTION #2 BELOW
More than half the time: SKIP TO QUESTION #2 BELOW

Less than half the time

Once in a while
NEVER

6. WHEN YOU DRINK BEER, HOW OFTEN DO YOU HAVE AS
MANY AS 3 to 4 CANS, GLASSES OR BOTTLES?

Nearly every time: SKIP TO QUESTION #2 BELOW
More than half the time: SKIP TO QUESTION #2 BELOW

Less than half the time

Once in a while
NEVER

f. WHEN YOU DRNK BEER, HOW OFTEN DO YOU HAVE 1 TO 2

CANS, GLASSES OR BOTTLES?

Nearly every time

More than half the time
Less than half the time

Once in a while
NEVER

WHEN DRINKING WINE:

a. HOW OFTEN DO YOU USUALLY HAVE WINE OR A WINE
COOLER, OR A PUNCH CONTAINING WINE?

3 or more times a day
2 times a day

Once a day

Nearly every day

3 or 4 times a week
Once or twice a week

207

2 or 3 times a month
About once a month
Less than once a month,
but at least once a year
Less than once a year

NEVER [If checked, go to

question #3a]

b. THINK OF ALL THE TIMES YOU HAD WINE OR A WINE
COOLER OR A PUNCH CONTAINING WINE RECENTLY, HOW
OFTEN DO YOU HAVE 10 OR MORE GLASSES OR WINE
COOLERS?

Nearly every time: SKIP TO QUESTION #3 BELOW
More than half the time: SKIP TO QUESTION #3 BELOW
Less than half the time

Once in a while
NEVER

0. WHEN YOU DRINK WINE, HOW OFTEN DO YOU HAVE AS
MANY AS 7 TO 9 GLASSES OR WINE COOLERS?

Nearly every time: SKIP TO QUESTION #3 BELOW
More than half the time: SKIP TO QUESTION #3 BELOW
Less than half the time

Once in a while
NEVER

lllll

(1. WHEN YOU DRINK WINE, HOW OFTEN DO YOU HAVE AS
MANY AS 5 to 6 GLASSES OR WINE COOLERS?

Nearly every time: SKIP TO QUESTION #3 BELOW

More than half the time: SKIP TO QUESTION #3 BELOW
Less than half the time

Once in a while

NEVER

e. WHEN YOU DRINK WINE, HOW OFTEN DO YOU HAVE AS
MANY AS 3 to 4 GLASSES OR WINE COOLERS?

Nearly every time: SKIP TO QUESTION #3 BELOW

More than half the time: SKIP TO QUESTION #3 BELOW
Less than half the time

Once in a while

NEVER

f. WHEN YOU DRINK WINE, HOW OFTEN DO YOU HAVE 1 TO 2
GLASSES OR WINE COOLERS?

Nearly every time
More than half the time
Less than half the time
Once in a while
NEVER

208

WHEN DRINKING WHISKEY OR LIQUOR

DJ

3. HOW OFTEN DO YOU USUALLY HAVE WHISKEY OR
LIQUOR (SUCH AS MARTINIS, MANHATTAN S,
HIGHBALLS, OR STRAIGHT DRINKS INCLUDING
SCOTCH, BOURBON, GIN, VODKA, RUM, ETC)?

3 or more times a day 2 or 3 times a month

2 times a day About once a month

Once a day Less than once a month, but at

Nearly every day least once a year

3 or 4 times a week Less than once a year

Once or twice a week NEVER [If checked, go to
question #4]

b. THINK OF ALL THE TIMES YOU HAD DRINKS CONTAINING
WHISKEY OR OTHER LIQUOR RECENTLY, WHEN YOU HAVE
HAD THEM, HOW OFTEN DO YOU HAVE AS MANY AS 10 OR
MORE?

Nearly every time: SKIP TO QUESTION #4 BELOW

More than half the time: SKIP TO QUESTION #4 BELOW
Less than half the time

Once in a while

NEVER

c. WHEN YOU HAVE HAD DRINKS CONTAINING WHISKEY OR
OTHER LIQUOR, HOW OFTEN DO YOU HAVE AS MANY AS 7 TO
9 DRINKS?

Nearly every time: SKIP TO QUESTION #4 BELOW

More than half the time: SKIP TO QUESTION #4 BELOW
Less than half the time

Once in a while

NEVER

d. WHEN YOU HAVE HAD DRINKS CONTAINING WHISKEY OR
OTHER LIQUOR, HOW OFTEN DO YOU HAVE AS MANY AS 5 TO
6 DRINKS?

Nearly every time: SKIP TO QUESTION #4 BELOW
More than half the time: SKIP TO QUESTION #4 BELOW
Less than half the time

Once in a while
NEVER

209

e. WHEN YOU HAVE HAD DRINKS CONTAINING WHISKEY OR
LIQUOR, HOW OFTEN DO YOU HAVE 3 TO 4 DRINKS?
Nearly every time: SKIP TO QUESTION #4 BELOW
More than half the time: SKIP TO QUESTION #4 BELOW
Less than half the time
Once in a while
NEVER

f. WHEN YOU HAVE HAD DRHNIKS CONTAINING WHISKEY OR
LIQUOR, HOW OFTEN DO YOU HAVE 1 TO 2 DRINKS?
Nearly every time
More than half the time
Less than half the time
Once in a while
NEVER

3 or more times a day 2 or 3 times a month

2 times a day About once a month
Once a day Less than once a month,
Nearly every day but at least once a year
3 or 4 times a week Less than once a year
Once or twice a week NEVER

 

NOW SOME QUESTIONS ABOUT OTHER TIME PERIODS.

 

NOW ABOUT THE PAST YEAR. ABOUT HOW MANY TIMES HAVE YOU
DRUNK JUST ENOUGH TO FEEL A LITTLE HIGH OR LIGHT-HEADED?

None Once a month

1 time in the past year Twice a month

2-3 times in the past year Once a week

4-5 times in the past year Twice a week

6-10 times in the past year More than twice a week

Hill

210

 

6. DURING THE PAST YEAR. ABOUT HOW MANY TIMES HAVE YOU GOTTEN
DRUNK OR VERY, VERY HIGH?

None Once a month

1 time in the past year Twice a month

2-3 times in the past year Once a week

4-5 times in the past year Twice a week

6-10 times in the past year More than twice a week

7. Now a question about earlier in your life; HOW OLD WERE YOU THE FIRST TIME
YOU EVER DRANK ENOUGH TO GET DRUNK?

years old; if you have never been drunk, check here

 

8a. WE ARE ALSO INTERESTED IN THE OCCASIONS THAT MAY BE RARE (OR
NOT), WHEN PEOPLE DRINK A LOT MORE THAN THEY USUALLY DO. D]
THE LAST YEAR. THINK OF THE 24 HOUR PERIOD WHEN YOU DID THE
MOST DRINKING: THIS WOULD BE A DAY SOMEWHERE IN THE PERIOD
BETWEEN , AND NOW.
(month) (year)

 

On that day, how many drinks did you have? (A drink is a 12 oz. can,
bottle or glass of beer, a 4 oz. glass of wine, a 12 oz. wine cooler bottle, a
single shot, or a single mixed drink)

30 ormore drinks
25 -29 drinks
20-24 drinks

15 - 19 drinks
10- 14 drinks

7- 9drinks

5 - 6drinks

3 - 4drinks

l- 2drinks

None

8b. APPROXIMATELY WHEN DID THIS HAPPEN? ,
(month) (year)

 

211

8c. NOW ANSWER THE QUESTION FOR ANY TIME IN YOUR LIFE BEFORE THIS
LAST YEAR. IN THE 24 HOUR PERIOD WHEN YOU DID THE MOST DRINKING.
HOW MANY DRINKS DID YOU HAVE?

30 or more drinks
25 -29 drinks

20 - 24 drinks

15 - l9 drinks

10 - 14 drinks

7 - 9 drinks

5 — 6 drinks

3 - 4 drinks

1 - 2 drinks
None

8d. APPROXIMATELY WHEN DID THIS HAPPEN?

 

(month) , (year)

 

C. NOW SOME QUESTIONS ABOUT WHERE YOU DRINK:

PLEASE INDICATE HOW OFTEN YOU DRINK BEER, WINE, OR LIQUOR IN EACH OF
THE FOLLOWING SETTINGS, PLACES, OR OCCASIONS. MARK X ON ONE BLANK
LINE IN EACH ROW.

Never drink
or don't drink Most of
in this setting Sometimes Frequently the time

 

1. At parties when other kids are
drinking and your parents or
other adults are not present.

2. At a party when other kids are
drinking and when your parents
or other adults fl present.

3. At home on special occasions
such as birthdays, or holidays
such as Thanksgiving, etc.

4. At dinner at home with your
family.

5a. At places where kids hang
around when their parents or
other adults are not present. If
you answer YES here, answer
Q. 5b and Q. So.

 

212

5b. Where?

 

So. What are you usually doing?

Never drink

 

 

or don't drink Most of
in this setting Sometimes Frequently the time
6. During or after a school activity
such as a dance or football
game, when your parents or
other adults you know are not
present or can't see you.
7. Driving around or sitting in
somebody's car at night.
8. Alone-- when no one else is
around.
9. When a grownup I know offers

it to me (not a parent).

D. NOW SOME QUESTIONS ABOUT OUTCOMES PEOPLE SOMETIMES HAVE
BECAUSE OF DRINKING. HAVE YOU EVER HAD ANY OF THE FOLLOWING HAPPEN
BECAUSE OF YOUR DRINKING?

ANSWER KEY FOR QUESTIONS BELOW:

 

 

 

1. Got into trouble with my
teachers or principal
because of my drinking.

2. Got into difficulties of

any kind with my friends.

 

 

213

1 2 3-5 6-10 11-20 21-50 51-100 101—250 251+
1
HOW MANY AGE AGE
TIMES first most
YES N_Q (approx.- time recent
(check one) see key)* time

 

ANSWER KEY FOR QUESTIONS BELOW:

 

 

 

1 2 3-5 6-10 11-20 21-50 51-100 101-250 251+ J
I
.1,
HOW MANY AGE AGE
TIMES first most
YES IQ (approx.- time recent
(check one) see key)* time

3. Driven a car when I'd had
a good bit to drink.

 

4. Been criticized by some-
one I was dating because
of my drinking.

 

5. Gotten in trouble with
the police because of
my drinking.

 

6. Gotten in trouble with

my parents because of
my drinking.

 

7. Missed school (or time on
job) because of my drinking.

 

8. Thought I was drinking too much.

 

9. Gone on a binge of constant
drinking.

 

10. Lost friends because of my
drinking.

 

ll. Felt guilty about my drinking.

 

12. Took a drink or two first
thing in the morning.

 

13. Restricted my drinking to
certain times of day or week
in order to control it
or cut down (like after 5PM,
or only on weekends,
or only with other people).

 

214

ANSWER KEY FOR QUESTIONS BELOW:

 

 

 

l 2 3-5 6-10 11-20 21—50 51-100 101-250 251+
I
i
HOW MANY AGE AGE
TIMES first most
YES 19 (approx.— time recent
(check one) see key)’ time

14. Once started drinking, kept

 

on going till drunk.

15. Had a car accident when I

 

was drinking and driving.

16. Kept on drinking afier

 

I promised myself not to.

17. Had the shakes "the

 

morning after".

18. Heard or saw or felt things

 

that weren't there
(hallucinations), several
days after stopping drinking.

19. Had blackouts (couldn't

 

remember later what
you'd done while drinking).

20. Been given a ticket for

 

drunk driving (DWI or DUIL).

21. Been given a ticket for

 

public intoxication, drunk
and disorderly or other
non-driving alcohol arrest.

22. Found that I had a strong

 

need for a drink at some
time each day.

215

 

ANSWER KEY FOR QUESTIONS BELOW:

 

 

 

23.

24.

25.

26.

27.

1 2 3-5 6-10 11-20 21-50 51-100 101-250 251+
r
.1.
HOW MANY AGE AGE
TIMES first most
YES NO (approx.- time recent
(check one) see key)* time

Needed to drink a lot more
in order to get an effect,

or found that I no longer
could get high on the amount
I used to drink.

Found that I was able to
drink a lot more than I

used to before I would get
drunk.

Had days where I drank
much more that I expected
to when I began

Found that I often
continued drinking for
more days In a row
than I had planned to

Found that I tended to
gulp my drinks rather
than just drink them

* SELECT ANSWERS FROM THE KEY AT THE TOP OF THE PAGE

 

 

 

 

 

216

 

 

E. SO FAR THE QUESTIONS HAVE ASKED FOR THE FACTS ABOUT YOUR
DRINKING. IN THIS SECTION YOU WILL BE ASKED ABOUT YOUR BEHAVIOR
AND THE BEHAVIOR OF YOUR FRIENDS WHEN DRINKING; AND, MOST
HVIPORTANTLY, WHAT YOU AND YOUR FRIENDS THINK ABOUT DRINKING.

1. Have any of your friends suggested that you try drinking?

Never
Once or twice
Several Times
Often

2. Do you think that your father (stepfather, mother's partner) ever takes a drink of beer,
wine or whiskey?

Yes fairly regularly
Yes, sometimes
No

I don't know

3. Do you think that your mother, (stepmother, father's partner) ever takes a drink of
beer, wine or whiskey?

Yes fairly regularly

Yes, sometimes

No

I don't know

4. How do you think your parents (or your family) feel about boys your age drinking?
Strongly approve

Approve

Don‘t care one way or the other
Disapprove

Strongly disapprove

I don't lmow

5. How do you think your parents (or your family) feel about girls your age drinking?
Strongly approve

Approve

Don't care one way or the other
Disapprove

Strongly disapprove

I don't lmow

217

None

How do most of the kids you hang around with feel about kids your age drinking?

Strongly approve

Approve

Neither approve nor disapprove
Disapprove

Strongly disapprove

I don't know

Does not apply

Please mark the blank which indicates the answer to the question on the right side.
Give one answer for each question. Mark X on one blank line in each row.

Less than More than All of

L2 Several mar h_al_f them

a. As far as you know, about how many of the

 

kids in your school class drink alcohol at least
sometimes?
b. About how many of the kids you hang around

 

10.

with drink alcohol at least sometimes?

Can you get alcoholic beverages when you want them?

I don't ever want them (check here if no drinking in last year)
No

Sometimes

Usually

Always

Where do you most often get the alcohol you and your fiiends drink?

I don't ever get it (check here if no drinking in last year)
From my home

A fiiend gives it to me

A friend or someone else buys it for me

I buy it myself

Other (Please explain)

 

 

Does your school show films or have discussion groups or other programs to
teach students about alcohol and drinking?

Yes No

218

THE LAST SECTIONS OF THIS QUESTIONNAIRE DEAL WITH VARIOUS DRUGS
OTHER THAN ALCOHOL. THERE IS STILL A LOT OF TALK THESE DAYS
ABOUT THIS SUBJECT, BUT VERY LITTLE ACCURATE INFORMATION.

WE HOPE THAT YOU CAN ANSWER ALL QUESTIONS TRUTHFULLY; BUT IF
YOU FIND ONE WHICH YOU FEEL YOU CANNOT ANSWER HONESTLY, WE
WOULD PREFER THAT YOU LEAVE IT BLANK.

REMEMBER THAT YOUR ANSWERS WILL BE KEPT STRICTLY CONFIDENTIAL
AND THEY ARE NEVER CONNECTED WITH YOUR NAME. THAT IS WHY THIS
QUESTIONNAIRE IS IDENTIFIED ONLY WITH A CODE NUMBER.

THE FOLLOWING QUESTIONS ARE ABOUT CIGARETTES (CHECK THE BEST
ANSWER):

la.

1b.

3a.

HAVE YOU EVER SMOKED CIGARETTES?

Never

Once or twice

Occasionally but not regularly
Regularly in the past
Regularly now

HAVE YOU SMOKED CIGARETTES DURING THE PAST 12 MONTHS?

Never

Once or twice

Occasionally but not regularly

Regularly for a while during this year, but not now
Regularly now

Not at all

Less than one cigarette per day

One to five cigarettes per day

About one-half pack per day

About one pack per day

About one and one-half packs per day
Two packs or more per day

Have you ever been around anyone else who has been smoking cigarettes?

Yes No

219

3b. How many times (circle one)

1-2 3-5 6-9 10-19 20-39 40-99 100-1000 More than 1000

3c. Who was smoking? (check all that apply)

Parents Kids I know well

Other adults I know well Kids I know, but not so well
Other adults I know, but not so well Kids I didn't know

Adults I didn't know

G. THE FOLLOWING QUESTIONS ARE ALL ABOUT NON-PRESCRIPTION USE OF
DRUGS, EITHER FOR RECREATION OR FOR SELF-MEDICATION.

 

 

8

.2
la. ON HOW MANY g g a g _g _g g g
OCCASIONS g .g .3 .2 g g g g g
(IF ANY) HAVE YOU USED ‘g g a g 8 8 8 g g
MARIJUANA (GRASS, POT) ‘8’ g g g 9; 9, a T 2

I r I O o é 8 0

OR HASHISH o .— m \o —‘ N <1- .—. E
(HASH, HASH OIL)
Inyourlifetime? () () () () () () () () ()
Dun'ngthe last 12 months? () () () () () () () () ()
Dufingthelast30days? () () () () () () () () ()
1b. HOW OLD WERE YOU WHEN YOU FIRST USED MARIJUANA? years old.

lc. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?
YES NO
1d. HOW MANY TIMES? (Circle one)
1-2 3—5 6—9 10-19 20-39 40-99 100-1000 More than 1000

1e. WHO WAS USING IT? (CHECK ALL THAT APPLY)

_ Parents _ Kids I know well

__ Other adults I know well _ Kids I know, but not so well
__ Other adults I know, but not so well _ Kids I didn't know

_ Adults I didn't know

220

2a. ON HOW MANY

 

 

OCCASIONS (IF ANY)
HAVE YOU SNIFFED GLUE,
OR BREATHED THE CONTENTS
OF AEROSOL SPRAY CANS, 3;:
OR INHALED ANY OTHER 8 a 8 '8‘ o
GASES OR SPRAYS IN a 8 8 '9 '9 ‘9 8 8

8 .2 .9. .9. 28 3 E o ....
ORDER TO GET HIGH g g a g g g 8 g g
(LIKE LIGHTER FLUID, 8 8 8 o a g; 3; ~. 2
NAIL POLISH REMOVER, g 2 3 3., g g ‘5'; § §
PAINT THINNER, AND PAINT) ,
Inyourlifetime? () () () () () () () () ()
During the last 12 months? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
During the last 30 days? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )

2b. HOW OLD WERE YOU WHEN YOU FIRST USED ANY OF THESE INHALANT S?
years old.

2c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?
YES NO

2d. HOW MANY TIMES? (Circle one)

1-2 3-5 6-9 10-19 20-39 40-99 100-1000 More than 1000

2e. WHO WAS USING IT? (CHECK ALL THAT APPLY)

__ Parents __ Kids I know well

_ Other adults I know well _ Kids I know, but not so well
_ Other adults I know, but not so well _ Kids I didn't know

_ Adults I didn't know

221

 

3a. ON HOW MANY

 

OCCASIONS m
(IF ANY) HAVE YOU USED m m m E o
AMYL a a 8 -§ E -§ § 8
OR BUTYL NTIRATES g .5 .3 .g a g g g ;
(POPPERS, g g g g 8 8 8 g a
SNAPPERS, LOCKER ROOM, g. :3 g f; if ‘3; 3, g g
VAPORALE, RUSH, KICK, o - A 8 S 8 S; ‘3 a
BULLET).

Inyourlifetime? () () () () () () () () ()
During the last 12 months? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
Duringthelast30 days? () ( ) ( ) () () () ( ) ( ) ( )
3b. HOW OLD WERE YOU WHEN YOU FIRST USED AMYL? years old.

3c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?
YES NO
3d. HOW MANY TIMES? (Circle one)
1-2 3-5 6-9 10-19 20-39 40—99 100-1000 More than 1000

3e. WHO WAS USING IT? (CHECK ALL THAT APPLY)

_ Parents _ Kids I know well

_ Other adults I know well _ Kids I know, but not so well
__ Other adults I know, but not so well _ Kids I didn't know

_ Adults I didn't lcnow

222

4a. AMPHETAMINES ARE SOMETIMES PRESCRIBED BY DOCTORS TO HELP
PEOPLE LOSE WEIGHT OR TO GIVE PEOPLE MORE ENERGY. THEY ARE
SOMETIMES CALLED UPPERS, UPS, SPEED, CRYSTAL, CRANK, BENNIES,
DEXIES, PEP PILLS, GREENIES, SPLASH AND DIET PILLS.

‘8

.2
ON HOW MANY OCCASIONS a a a .5 .5 _g § §
(IF ANY) HAVE YOU TAKEN g .3 .g .3 a g a g ;
AMPHETAMINES ON YOUR g g g g 8 8 8 g g
OWN-THAT IS, WITHOUT A a (3 .53 g 9-3 33 8, g g
DOCTOR TELLING YOU 0 .2 A 8 2 S. 8 S E
Inyourlifetime? () () () () () () () () ()
During the last 12 months? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
During the last 30 days? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
4b. HOW OLD WERE YOU WHEN YOU FIRST USED AMPHETAMINES?

years old.

 

4c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?

YES NO

 

4d. HOW MANY TIMES? (Circle one)

1-2 3-5 6-9 10-19 20-39 40-99 100-1000 Morethan 1000

4e. WHO WAS USING IT? (CHECK ALL THAT APPLY)

_ Parents _ Kids I know well

_ Other adults I lmow well _ Kids I lmow, but not so well
_ Other adults I know, but not so well _ Kids I didn't know

__ Adults I didn't know

223

5a. ON HOW MANY

OCCASIONS m m m g
(IF ANY) HAVE YOU USED 3 E, E, “g g .g g g
LSD (ACID) 8 .2 .2 .2 g g g o g
8 8 8’ 8 8 8 8 § .8
§ 8 8 8 9.: a 8 '7‘ E
o as V? a: 8 8 8 8 o
O F‘ to \O H N V q— E
—
Inyourlifetime? () () () () () () () () ()
During the last 12 months? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
During the last 30 days? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
5b. HOW OLD WERE YOU WHEN YOU FIRST USED LSD? years old.

 

5c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?

YES NO

 

 

5d. HOW MANY TIMES? (Circle one)
1-2 3-5 6-9 10-19 20-39 40-99 100-1000 More than 1000

Se. WHO WAS USING IT? (CHECK ALL THAT APPLY)

__ Parents _ Kids I know well

_ Other adults I know well _ Kids I know, but not so well
__ Other adults I know, but not so well _ Kids I didn't know

_ Adults I didn't know

224

 

6a. ON HOW MANY

 

OCCASIONS ,,
(IF ANY) HAVE YOU USED m m .. ,8
PSYCHEDELICS OTHER THAN u. m u, ,8 ,8 ,8 8 §
LSD (LIKE MESCALINE, g § § g g g g :g’ ;
PEYOTE, PSILOCYBIN, PCP, g g g g 8 8 8 g g
ANGEL DUST) 8 ° ° ° 9 8 8 g 2
o (a v.) a: 8 o' 8 o o
o —. m \o —r N st .— E
Inyourlifetime? () () () () () () () () ()
During the last 12 months? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
Duringthelast30days? ( ) () () () () (l () () ()

6b. HOW OLD WERE YOU WHEN YOU FIRST USED PSYCHEDELICS OTHER THAN
LSD? years old.

6c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?
YES NO
6d. HOW MANY TIMES? (Circle one)
1-2 3-5 6-9 10-19 20-39 40—99 100-1000 More than 1000

66. WHO WAS USING IT? (CHECK ALL THAT APPLY)

_ Parents _ Kids I lcnow well

__ Other adults I know well _ Kids I know, but not so well
__ Other adults I know, but not so well _ Kids I didn't know

__ Adults I didn't know

225

 

7a. ON HOW MANY

 

OCCASIONS g
(IF ANY) HAVE YOU USED g g g g 8
CRACK. a g E g g E. ‘3 E S

g 8 8 8 8 o 8 8 5

g 8 8 8 .‘L‘ 3 a g a

8 2 Z-I 8 2' 8' 9'. 2 8
Inyourlifetime? () () () () () () () () ()
Duringthepast 12 months? () () () () () () () () ()
Duringthelast30days? () () () () () () () () ()
7b. HOW OLD WERE YOU WHEN YOU FIRST USED CRACK? __ years old.

7c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?
YES NO
7d. HOW MANY TIMES? (Circle one)
1-2 3-5 6-9 10-19 20-39 40-99 100-1000 More than 1000

7e. WHO WAS USING IT? (CHECK ALL THAT APPLY)

_ Parents _ Kids I know well

__ Other adults I know well _ Kids I know, but not so well
_ Other adults I know, but not so well _ Kids I didn't know

_ Adults I didn't lcnow

226

 

 

8a. ON How MANY g
OCCASIONS g a a g 8
(IF ANY) HAVE YOU USED 2 a 8 8 g g 8 §
COCAINE (COKE) é 8 § 8. 5‘3 g 2 °

8 8 8 8 ° ° ° § §

0 o o o 2‘. S; a "‘ o

8 2.! V: Cr 8 8 8 8 '5

o —. m \c —< N <1- .— E
Inyourlifetime? () () () () () () () () ()
DuringthePaSt 12 months? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
DuringthelaSt 30 dayS? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
8b. HOW OLD WERE YOU WHEN YOU FIRST USED COCAINE? __ years old.

8c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?
YES NO
8d. HOW MANY TIMES? (Circle one)
1-2 3-5 6-9 10—19 20-39 40-99 100-1000 More than 1000

Se. WHO WAS USING IT? (CHECK ALL THAT APPLY)

_ Parents _ Kids I know well

_ Other adults I know well __ Kids I know, but not so well
_ Other adults I know, but not so well _ Kids I didn't know

_ Adults I didn't know

227

9a. STEROIDS, OR ANABOLIC
STEROIDS, ARE SOMETIMES

 

 

USED FOR BODY BUILDING ,,
OR TO IMPROVE ATHLETIC m .. m ,8
PERFORMANCE. ON HOW E, ., ,, g g _g <3 §
MANY OCCASIONS (IF ANY) 2 2 § g g g 5 g ;
HAVE YOU USED STEROIDS g g g g 8 8 8 g 8
ON YOUR OWN-THAT IS, g :3 g g 3 83 8 g g
WITHOUT A 8 _- ..-. 8 2 8 8 2 E
DOCTOR TELLING YOU TO

TAKE THEM.

Inyourlifetime? () () () () () () () () ()
Dufing the last 12 months? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
Duringthelastwdays? () I) () I) () () I) I) I)
9b. HOW OLD WERE YOU WHEN YOU FIRST USED STEROIDS? _ years old.

9c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?
YES NO
9d. HOW MANY TIMES? (Circle one)
1-2 3-5 6-9 10-19 20-39 40-99 100-1000 More than 1000

9e. WHO WAS USING IT? (CHECK ALL THAT APPLY)

_ Parents __ Kids I know well

__ Other adults I know well _ Kids I know, but not so well
__ Other adults I know, but not so well _ Kids I didn't know

_ Adults I didn't know

228

10a. ON HOW MANY
OCCASIONS

(IF ANY) HAVE YOU USED
QUAALUDES (QUADS,
SOAPERS, METHAQUALONE)
ON YOUR OWN--THAT IS,
WITHOUT A DOCTOR

0 occasions

1-2 occasions

3-5 occasions

6-9 occasions

10-19 occasions
20-39 occasions
40—99 occasions
100-1000 occasions
more than 1000

Inyourlifetime? () () () () () () () () ()
During the last 12 months? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
Duringthelast30days? () () () () () () () () ()
10b. HOW OLD WERE YOU WHEN YOU FIRST USED QUAALUDES? years old.

10c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?

YES NO

 

10d. HOW MANY TIMES? (Circle one)
1-2 3-5 6-9 10-19 20-39 40-99 100-1000 More than 1000

10e. WHO WAS USING IT? (CHECK ALL THAT APPLY)

_ Parents _ Kids I know well

__ Other adults I know well _ Kids I know, but not so well
__ Other adults I know, but not so well __ Kids I didn't know

__ Adults I didn't know

229

11a. TRANQUILIZERS ARE SOMETIMES PRESCRIBED BY DOCTORS TO CALM
PEOPLE DOWN, QUIET THEIR NERVES, OR RELAX THEIR MUSCLES.
LIBRIUM VALIUM, EQUANIL AND MILTOWN ARE ALL TRANQUILIZERS.

 

 

8

.2
ON HOW MANY OCCASIONS a a a E ,8 ,8, 8 §
(IF ANY) HAVE YOU TAKEN g .g .9 8 8 8 8 g E
TRANQUILIZERS ON YOUR g g g g 8 8 3’ § 8
OWN--THAT Is, WITHOUT g ,3 ,3 g 9‘: 8. on 8 E
A DOCTOR TELLING YOU 0 —'~ 8 2 3 8 8 9 8
TO TAKE THEM.
Inyourlifetime? () ( ) () () ( ) () ( ) () ()
Duringthe laSt12m0nthS? ( ) ( ) () () () () () () ()
Duringthelast30day5'? () () () () () () () () ()
11b. HOW OLD WERE YOU WHEN YOU FIRST USED TRANQUILIZERS?

years old.

11c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?
YES NO
11d. HOW MANY TIMES? (Circle one)
1-2 3-5 6-9 10~19 20-39 40-99 100-1000 More than 1000

116. WHO WAS USING IT? (CHECK ALL THAT APPLY)

_ Parents __ Kids I know well

_ Other adults I know well __ Kids I know, but not so well
_ Other adults I know, but not so well __ Kids I didn't lcnow

_ Adults I didn't know

230

12a. BARBITURATES ARE SOMETIMES PRESCRIBED BY DOCTORS TO HELP PEOPLE
RELAX OR GET TO SLEEP. THEY ARE SOMETIMES CALLED DOWNS, DOWNERS,
GOOFBALLS, YELLOWS, REDS, BLUES, RAINBOWS.

 

 

ON HOW MANY OCCASIONS a a a g o
(IF ANY) HAVE YOU TAKEN g a. a, ,9, ,9 ,9 g g
BARBITURATES a .g .g .2 8 8 8 g T
ON YOUR OWN -- THAT IS, '3) § § .3 8 8 8 g g
WITHOUT A DOCTOR g :3 33 g 9,- 83 8 g g
TELLING YOU TO TAKE THEM. o 8 8 8 2 8’. S S E
Inyourlifetime? () () () () () () () () ()
Dufingthelast () () () () () () () () ()
Dufingthelast30days? () () () () () () () () ()
12b. HOW OLD WERE YOU WHEN YOU FIRST USED BARBITURATES?
years old.

12c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?
YES NO
12d. HOW MANY TIMES? (Circle one)
1-2 3-5 6-9 10-19 20-39 40-99 100-1000 More than 1000

12c. WHO WAS USING IT? (CHECK ALL THAT APPLY)

_ Parents _ Kids I know well

__ Other adults I know well _ Kids I know, but not so well
__ Other adults I know, but not so well __ Kids I didn't know

_ Adults I didn't know

231

13a. ON How MANY g
OCCASIONS ,,, U, m .g o
(IF ANY) HAVE YOU USED 3 g g § .§ g E; 8
HEROIN (SMACK, HORSE, ‘g g, a a g g g 3 g
SKAG, JUNK). ‘§ § § § 0 ° ° 8 a
8 o o o a a 8“ ‘T a
o ‘1' V9 °.~ 8 :5 a: 8 o
o .—. m \o .—. N v .— E
—
Inyour lifetime? ( ) ( ) () ( ) ( ) ( ) () () ( )
During the last 12 months? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
Duringthe last 30 days? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
13b. HOW OLD WERE YOU WHEN YOU FIRST USED HEROIN? years old.

 

l3c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?

YES NO

 

l3d. HOW MANY TIMES? (Circle one)
1-2 3-5 6-9 10-19 20-39 40-99 100-1000 More than 1000

136. WHO WAS USING IT? (CHECK ALL THAT APPLY)

_ Parents _ Kids I know well

__ Other adults I lmow well _ Kids I know, but not so well
_ Other adults I know, but not so well _ Kids I didn't know

_ Adults I didn't lmow

232

14a. THERE ARE A NUMBER OF NARCOTICS OTHER THAN HEROIN, SUCH AS
METHADONE, OPIUM, MORPHINE, CODEINE, DEMEROL, PAREGORIC,
TALWIN, AND LAUDANUM. THESE ARE SOMETIMES PRESCRIBED BY

 

 

DOCTORS.
ON HOW MANY OCCASIONS a a m .3
(IF ANY) HAVE YOU TAKEN a ‘a a _g ,9 ,S § §
NARCOTICS OTHER THAN g g a a 5 S 3 g g
HEROIN ON YOUR OWN-- ‘3; 8 8 g g 8 c8 8 5
THAT IS, WITHOUT A DOCTOR g ,3 E g —. «p c; g g

I I I O O o

TELLING YOU TO TAKE THEM o — m ,0 — N 8 ~ 5
Inyourlifetime? ( ) ( ) () ( ) () ( ) ( ) () ()
DuringthCIaSt 12 monthS? ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
DUfingthelaSt30dayS? () () () () () () () () ()

14b. HOW OLD WERE YOU WHEN YOU FIRST USED OTHER NARCOTICS?
years old.

14c. HAVE YOU EVER BEEN AROUND ANYONE ELSE WHO HAS BEEN USING IT?
YES NO
14d. HOW MANY TIMES? (Circle one)
1-2 3-5 6-9 10-19 20-39 40-99 100-1000 More than 1000

Me. WHO WAS USING IT? (CHECK ALL THAT APPLY)

_ Parents _ Kids I lmow well

__ Other adults I know well _ Kids I know, but not so well
_ Other adults I know, but not so well _ Kids I didn't know

_ Adults I didn't know

233

H. NOW SOME OTHER QUESTIONS ABOUT NONPRESCRIPTION USE OF DRUGS.
HAVE YOU EVER HAD ANY OF THE FOLLOWING THINGS HAPPEN BECAUSE OF
YOUR USE OF THE NONPRESCRIPTION DRUGS ASKED ABOUT IN SECTION G (THE
LAST SECTION)?

ANSWER KEY FOR QUESTIONS BELOW:

 

 

 

 

1 2 3-5 6-10 11-20 21-50 51-100 101-250 251+
YES N_Q HOW MANY AGE AGE
TIMES first most recent
(approx) TIME TIME
(see key)*

1. Been absent from school one
or more times because of my
use.

 

2. Had my grades in school get
worse than they were because
of my use.

 

3. Caused me to be stopped
by the police or get a
traffic citation.

 

4. Caused some physical or
medical problem (even a
minor or unimportant one).

 

5. Found it hard to concentrate
on something I wanted to
do, because of my use.

 

6. Had trouble getting along
with my parents (at least
once) because they didn't
want me to use any of the stuff.

 

7. Found myself unable to
control my moods when I
used any of the stuff.

 

234

8. Had trouble getting along
with some of my friends
because of use.

 

9. Missed school (or time on the job).

 

10. Lost friends because of use.

 

 

11. Been fired or laid off
from a job because of use

 

12. Had a car accident when
I was driving

 

13. Had to go to a hospital
(other than accidents)

 

14. Had to stay in hospital
overnight

 

15. Had to see a doctor because
of drug use (unintentional
overdose) or had a doctor
say drugs had harmed your health

 

16. Gone through physical with-
withdrawal from drugs

 

17. Been arrested for
possession of marijuana

 

 

18a. Have you ever taken drugs intravenously (using a needle)? Don't count shots you were
given by a doctor or nurse or shots you may have taken for treatment of diabetes.

NO YES

 

 

18b. IF YES, WHAT DRUGS HAVE YOU TAKEN INTRAVENOUSLY (IV)?

 

18c. AT WHAT AGE DID YOU FIRST TAKE AN IV DRUG? years old.

 

18d. AT WHAT AGE WAS THE MOST RECENT TIME? years old.

 

235

REFERENCES

236

REFERENCES

Achenbach, T. M. (1991). Manual for the Child Behavior Checlflist/4-l 8 M
1991 profile. Burlington, VT: University of Vermont department of Psychiatry.

 

American Psychiatric Association (1994). Diagnostic and Statistical Manual Of
Mental Disorders (Fourth Edition). Washington, DC: APA.

Anthony, J. C., & Echeagaray-Wagner, F. (2000). Epidemiologic analysis of
alcohol and tobacco use: Patterns of co-occurring consumption and dependence in the
United States. Alcohol Research & Health, 24(4), 201-208.

Bagley, C. (1992). Maternal smoking and deviant behavior in 16-year-Olds: A
personality hypothesis. Personality and Individual Differences, 13(3), 377-378.

Babor, T. F ., Hofrnann, M., DelBoca, F. K., Hesselbrock, V., Meyer, R. E.,
Dolinsky, Z. S., & Rounsaville, B. (1992). Types of alcoholics, 1: Evidence for an
empirically derived typology based on indicators of vulnerability and severity. Archives
of General Psychiatry, 49, 599-608.

Bates, M. E. (2000). Integrating person-centered and variable-centered
approaches in the study of developmental courses and transitions in alcohol use:
Introduction to the Special section. Alcoholism: Clinical and Experimental Research,
2_4(6), 878-881.

Beck, A. T ., Rush, A. J ., Shaw, B. F., & Emery, G. (1979). Cogpitive therapy of
depression. New York: Guilford Press.

Beck, A. T., & Steer, R. A. (1993). Beck Depression Inventory manual. San
Antonio, TX: The Psychological Corporation.

Begleiter, H., & Projesz, B. (1999). What is inherited in the predisposition
toward alcoholism? A proposed model. Alcoholism: Clinical and Experimental
Research. 23(7), 1125-1135.

Benowitz, N. L. (1996). Pharmacology of nicotine: Addiction and therapeutics.
American Review of Pharmacology and Toxicology, 36, 597-613.

Bergman, L. R., & Magnusson, D. (1997). A person-oriented approach in
research on developmental psychopathology. Development and Psvchopathologyfi,
291 -3 19.

Bierut, L. J ., Schuckit, M. A., Hesselbrock, V., & Reich, T. (2000). Co-occurring

risk factors for alcohol dependence and habitual smoking. Alcohol Research and Hea1_th_,
EM). 233-241.

237

 

Bobo, J. K., & Husten, C. (2000). Sociocultural influences on smoking and
drinking. Alcohol Research and Health 24(4), 225-232.

Bollen, K. A. (1989). Structural equations with latent variables. New York: John
Wiley & Sons.

Brennan, P. A., Grekin, E. R., & Mednick, S. A. (1999). Maternal smoking
during pregnancy and adult male criminal outcomes. Archives of General Psychiatry, 56,
21 5-219.

Breslau, N., F enn, N., & Peterson, E. L. (1993). Early smoking initiation and
nicotine dependence in a cohort of young adults. Drug and Alcohol Dependence. 33.
129-137.

Burke, J. D., Loeber, R., & Lahey, B. B. (2001). Which aspects of ADHD are
associated with tobacco use in early adolescence? J oumal of Child Psychology and
Psychiatry, 42(4), 493-502.

 

Cahalan, D., Cisin, 1., & Crossley, H. (1969). American drinking practices:
National study of drinkingbehavior and attitudes. New Brunswick, NJ: Publications
Division, Rutgers Center of Alcohol Studies.

Carpenter, J ., & Lester, D. (1980). The etiology of alcoholism: A longitudinal
study. Mimeo New Brunswick, NJ: Rutgers Center of Alcohol Studies.

Centers for Disease Control and Prevention (1994). Guidelines for school health
programs to prevent tobacco use and addiction. Morbidity and Mortality Weekly Rgport,
43, 1-18.

Centers for Disease Control and Prevention (1999). Tobacco use - United States,
1900 - 1999. Morbidity and Mortality Weekly Report. 48, 986-993.

Centers for Disease Control and Prevention (2000). Strategies for reducing
exposure to environmental tobacco smoke, increasing tobacco-use cessation, and
reducing initiation in communities and health-care system. Morbidity and Mortality
Weekly Report. 49 (RR-12), 1-11.

Centers for Disease Control and Prevention (2001). Investment in tobac__co
control-state highlights 2001. Atlanta, GA: US. Department of Health and Human
Services, Centers for Disease Control and Prevention, National Center for Chronic
Disease Prevention and Health Promotion, Office on Smoking and Health.

Chassin, L., Curran, P., Hussong, A. M., & Colder, C. R. (1996a). The relation of

parental alcoholism to adolescent substance use: A longitudinal follow-up study. Journal
of Abnormal Psychology, 105(1), 70-80.

238

Chassin, L., Presson, C. C., Montello, D., Sherman, S. J ., & McGrew, J. (1986).
Changes in peer and parent influence during adolescence: Longitudinal versus cross-
sectional perspectives on smoking initiation. Developmental Psychology, 22(3), 327-
334.

Chassin, L., Presson, C. C., Pitts, S. C., & Sherman, S. J. (2000). The natural
history of cigarette smoking from adolescence to adulthood in a midwestem community
sample: Multiple trajectories and their psychosocial correlates. Health Psychology,
QB), 223-231.

Chassin, L., Presson, C. C., Rose, J. S., & Sherman, S. J. (1996b). The natural
history of cigarette smoking fiom adolescence to adulthood: Demographic predictors of
continuity and change. Health Psychology, 15(6), 478-484.

Chassin, L., Presson, C. C., Sherman, S. J ., & Edwards, D. A. (1990). The natural
history of cigarette smoking: Predicting young-adult smoking outcomes from adolescent
smoking patterns. Health Psychology, 9(6), 701-716.

Chassin, L., Presson, C. C., Sherman, S. J ., & Edwards, D. A. (1991). Four
pathways to young-adult smoking status: Adolescent social-psychological antecedents in
a midwestem community sample. Health Psychology, 10(6), 409-418.

Chassin, L., Presson, C. C., Todd, M., Rose, J. S., & Sherman, S. J. (1998).
Maternal socialization of adolescent smoking: The intergenerational transmission of
parenting and smoking. Developmental Psychology, 34(6), 1189-1201.

Chen, K., & Kandel, D. B. (1995). The natural history of drug use from
adolescence to the rnid-thirties in a general population sample. American Journal of
Public Health, 85(1), 41-47.

Cicchetti, D., & Rogosch, F. A. (1996). Equifinality and multifinality in
developmental psychopathology. Development and Psychopathology, 8, 597-600.

Cicchetti, D., & Toth, S. L. (1998). Perspectives on research and practice in
developmental psychopathology. In W. Damon, I. E. Sigel, & K. A. Renninger (Eds),
H_andbool_( of child psychology: Child psychology in practice (5 ed., Vol. 4, pp. 479-583).
New York: John Wiley.

Cloninger, C. R. (1987). Neurogenetic adaptive mechanisms in alcoholism.
Science 236 410-416.

 

Cooper, S. (1987). The fetal alcohol syndrome. Journal of Child Psychology and
Psychiat_ry, 28, 223-227.

239

Cornelius, M. D., & Day, N. L. (2000). The effects of tobacco use during and
after pregnancy on exposed children. Alcohol Research and Health, 24(4), 242-249.

Cornelius, M. D., Leech, S. L., Goldschmidt, L., & Day, N. L. (2000). Prenatal
tobacco exposure: Is it a risk factor for early tobacco experimentation? Nicotine &
Tobacco Research. 2, 45-52.

Curran, P. J ., Stice, E., & Chassin, L. (1997). The relation between adolescent
alcohol use and peer alcohol use: A longitudinal random coefficient models. Journal of
Consulting and Clinical Psychology. 65(1), 130-140.

Day, N. L., Cornelius, M. D., & Goldschmidt, L. (1992). The effects of prenatal
tobacco and marijuana use on offspring growth from birth through age 3 years.
Neurotoxicology and Teratology, H, 407-414.

DeLucia, C., Belz, A., & Chassin, L. (2001). Do adolescent symptomatology and
family environment vary over time with fluctuations in paternal alcoholism impairment?
Developmental Psychology,3 7(2), 207-216.

Downey, K. K., Pomerleau, C. S., & Pomerleau, O. F. (1996). Personality
differences related to smoking and adult attention deficit hyperactivity disorder. Journal
of Substance Abuse, 8(1), 129-135.

D'Unger, A., Land, K., McCall, P., & Nagin, D. (1998). How many latent classes
of delinquent/criminal careers? Results from mixed Poisson regression analyses of the
London, Philadelphia, and Racine cohorts studies. American Journal of Sociology. 103.
1 593-1630.

Ellis, D. A., Zucker, R. A., & Fitzgerald, H. E. (1997). The role of family
influences in development and risk. Alcohol Health & Research World. 2_1_, 218-226.

Engels, R. C. M. E., Knibbe, R. A., & Drop, M. J. (1999). Predictability of
smoking in adolescence: Between optimism and pessimism. Addiction. 94(1), 115-124.

Fergusson, D. M., Horwood, L. J ., & Lynskey, M. T. (1993). Maternal smoking
before and afier pregnancy: Effects on behavioral outcomes in middle childhood.
Pediatrics, 92(6), 815-822.

Fitzgerald, H. E., Sullivan, L. A., Ham, H. P., Zucker, R. A., Bruckel, S., &
Schneider, A. M. (1993). Predictors Of behavior problems in three-year-old sons of
alcoholics: Early evidence for the onset of risk. Child Development. 64. 110-123.

Fitzgerald, H. E., & Zucker, R. A. (1995). Socioeconomic status and alcoholism:

Structuring developmental pathways to addition. In H. E. Fitzgerald, B. M. Lester, & B.
Zuckerrnann (Eds), Children ofpoverg (pp. 125-147). New York: Garland.

240

Fitzgerald, H. E., Zucker, R. A., Puttler, L. I., Caplan, H. M., & Mun, E. Y.
(2000). Alcohol abuse and dependence in women and girls. Alcosccnge: International
Review Of Alcoholism Management, 3(1), 6-19.

Fried, P. A., Watkinson, B., & Gray, R. (1992). A follow—up study of attentional
behavior in 6-year-old children exposed to prenatally to marijuana, cigarettes, and
alcohol. Neurotoxicology and Teratology, 14, 299-311.

Friendly, M. (1992). User's guide for MOSAICS (Tech. Rep. No. 206). York
University, Department of Psychology.

Friendly, M. (1994). Mosaic displays for multi-way contingency tables Journal
of the American Statistical Associaaion,89, 190-200.

Ge, X., Lorenz, F. O., Conger, R. D., Elder, Jr., G. H., & Simons (1994).
Trajectories of stressful life events and depressive symptoms during adolescence.
Developmental Psychology, 30(4), 467-483.

Gelman, A., Carlin, J. 8., Stem, H. S., & Rubin, D. B. (1995). Bayesian data
analfiis. New York: Chapman & Hall/CRC.

Graham, J. W., & Hofer, S. M. (2000). Multiple imputation in multivariate
research. In T. D. Little, K. U. Schnabel, & J. Baumert (Eds), Modeling longitudinal ang
multilevel Ilia (pp. 201-218). Mahwah, NJ: Lawrence Erlbaum Associates.

Grant, B. F. (1998). Age at smoking onset and its association with alcohol
consumption and DSM-IV alcohol abuse and dependence: Results from the National
longitudinal alcohol epidemiologic survey. Journal of Substance Abuse. 10. 59-73.

Griesler, P. C., Kandel, D. B., & Davies, M. (1998). Maternal smoking in
pregnancy, child behavior problems, and adolescent smoking. Journal of Research on
Adolescence. 8(2), 159-185.

Grundy, C. T., Lunnen, K. M., Lambert, M. J ., Ashton, J. E., & Tovey, D. R.
(1994). The Hamilton Rating Scale for Depression: One scale or many? Clinical
P_sycholpgy: SciencL and Practice,1(2), 197-205.

 

Hamilton, M. (1960). A rating scale for depression. Journal of Neurology,
Neurosuygerv. and Psychiatry. 23, 56-62.

Hanna, E. Z., & Grant, B. F. (1999). Parallels to early onset alcohol use in the
relationship of early onset smoking with drug use and DSM-IV drug and dependence

disorders: Findings from the National Longitudinal Epidemiologic Survey. Alcoholism:
Clinical and Experimental Research. 23(3), 513-522.

Hartigan, J. A., & Kleiner, B. (1981). Mosaics for contingency tables. In W. F.

241

Eddy (Ed.), Proceedings of the 13th smposium on the interface between computer
scienmnd statistics (pp. 268-273). New York: Springer-Verlag.

Hartigan, J. A., & Kleiner, B. (1984). A mosaic of television ratings. flip
American Statistician, 38 (1), 32-35.

Heishman, S. J ., Kozlowski, L. T., & Henningfield, J. E. (1997). Nicotine
addiction: Implications for public health policy. Journal of Social Issues. 53(1), 13-33.

Hill, K. G., White, H. R., Chung, I.-J., Hawkins, D., & Catalano, R. F. (2000).
Early adult outcomes of adolescent binge drinking: Person- and variable-centered
analyses of binge drinking trajectories. Alcoholism: Clinical and Experimental Research,
24(6), 892-901.

Hoyle, R. H., & Panter, A. T. (1995). Writing about structural equation models.
In R. H. Hoyle (Ed.), Structural equation modeling: Concepts, issues, and applications
(pp. 158-176). New York: SAGE Publications.

Hughes, J. R. (1995). Clinical implications of the association between smoking
and alcoholism. In J. B. F ertig & J. P. Allen (Eds), Alcohol and tobacco: From basic
science to clinical mactice (pp. 171-186). Bethesda, MD: National Institute on Alcohol
Abuse and Alcoholism.

Hughes, J. K., Rose, G. L., & Callas, P. W. (2000). Nicotine is more reinforcing
in smokers with a past history of alcoholism than in smokers without this history.
Alcoholism: Clinical and Experimental Research, 24(11), 1633-1638.

Ichiyama, M A., Zucker, R. A., Fitzgerald, H. E., & Bingham, C. R. (1996).
Articulating subtype differences in self and relational experience among alcoholic men
via structural analysis of social behavior. J oumflf Consultingand Clinical Psychology,
64, 1245-1254.

Jackson, C. (1998). Cognitive susceptibility to smoking and initiation of smoking
during childhood: A longitudinal study. Preventive Medicine. .2, 129-134.

J essor, R., & J essor, S. L. (1977). Problem behavior and psychosocial
development: A longitudinal study of youth. New York, NY: Academic.

Johnson, T. P., & Mott, J. A. (2001). The reliability of self-reported age of onset
of tobacco, alcohol, and illicit drug use. Addiction. 96, 1187-1198.

Johnston, L. D., Bachman, J. G., & O'Malley, P. M. (1978). Drugs and the class

of '78: Behaviors, attitudes , and recent national trends. National Institute on Drug
Abuse: Division of Research, US. Department of Health, Education, and Welfare.

242

Jones, B. L., Nagin, D. S., & Roeder, K. (2001). A SAS procedure based on
mixture models for estimating developmental trajectories. Sociological Methods and
Research, 29, 374-393.

 

JOreskog, K. G., & SOrbom, D. (2001). LISREL (version 8.51). Chicago:
Scientific Software International.

JOreskog, K. G., & SOrbom, D. (1996). LISREL 8: User's reference guide.
Chicago: Scientific Sofiware International.

JOreskog, K. G., & SOrbom, D. (1996). PRELIS 2: User's reference guide.
Chicago: Scientific Software International.

Kandel, D. B., Wu, P., & Davies, M. (1994). Maternal smoking during pregnancy
and smoking by adolescent daughters. American Journal of Public Health, 84(9), 1407-
1413.

Kassel, J. D., Shiffrnan, S., Gnys, M., Paty, J ., & Zettler-Segal, M. (1994).
Psychosocial and personality differences in chippers and regular smokers. Addictive
Behaviors, 19(5), 565-575.

Koopmans, J. R., Slutske, W. 8., Heath, A. C., Neale, M. C., & Boomsma, D. I.
(1999). The genetics of smoking initiation and quantity smoked in Dutch adolescent and
young adult twins. Behavior Genetics, 29(6), 383-393.

Lambert, N. M., & Hartsough, C. S. (1998). Prospective study of tobacco
smoking and substance dependencies among samples of ADHD and non-ADHD
participants. Journal of Learning Disabilities, 31(6), 533-544.

Lehmacher, W. (1981). A more powerful simultaneous test procedure in
configural frequency analysis. Biometrical Journal, 23(5), 429-436.

Lerman, C., Caporaso, N. E., Audrain, J ., Main, D., Bowman, E. D., Lockshin, B.,
Boyd, N. R., & Shields, P. G. (1999). Evidence suggesting the role of specific genetic
factors in cigarette smoking. Health Psychology, 18(1), 14-20.

Lerner, R. M., Palermo, M., Spiro III, A., & Nesselroade, J. R. (1982). Assessing
the dimensions of temperamental individuality across the life Span: The Dimensions of
Temperament Survey (DOTS). Child Development. 53. 149-159.

Levin, E. D., Conners, C. K., Sparrow, E., Hinton, S. C., Erhardt, D., Meck, W.
H., Rose, J. E., & March, J. (1996). Nicotine effects on adults with attention-
deficit/hyperactivity disorder. Psychopharrnacology, 123, 55-63.

Little, H. J. (2000). Behavioral mechanisms underying the link between smoking
and drinking. Alcohol Research & Health, 24(4), 215-224.

243

Little, R. (2001). Statistical analysis of longitudinal data with missing
values. A talk given in the Department of Epidemiology, Michigan State University.

Lynskey, M. T., & Hall, W. (2001). Attention deficit hyperactivity disorder and
substance use disorders: Is there a causal link? Addiction, 96, 815-822.

Madden, P. A. F., Bucholz, K. K., Martin, N. G., & Heath, A. C. (2000). Smoking
and the genetic contribution to alcohol-dependence risk. Alcohol Research & Heal_th,
2_4(4), 209-214.

Madden, P. A. F., Heath, A. C., Pedersen, N. L., Kaprio, J ., Koskenvuo, M. J ., &
Martin, N. G. (1999). The genetics of smoking persistence in men and women. Behavior
Genetics, 29(6), 423-431.

Madden, P. A. F., Heath, A. C., Starmer, G. A., Whitfield, J. B., & Martin, N. G.
(1995). Alcohol sensitivity and smoking history in men and women. Alcoholism:
Clinical and Experimental Research, 19(5), 1111-1120.

 

Magnusson, D. (1998). The logic and implications of a person-oriented approach.
In R. B. Cairns, L. R. Bergman, & J. Kagan (Eds), Methods and models for studyingahe
individual (pp. 33-62). Thousand Oaks, CA: SAGE Publications.

Magnusson, D. (1999). On the individual: A person-oriented approach to
developmental research. European Psychologt, 4(4), 205-218.

Magnusson, D. (2000). The individual as the organizing principle in
psychological inquiry: A holistic approach. In L. R. Bergman, R. B. Cairns, L. Nilsson,
& L. Nystedt (Eds), Developmental science and the holistic approach (pp. 33-48).
Mahwah, NJ: Lawrence Erlbaum Associates, Publishers.

Malarcher, A. M., Schuhnan, J ., Epstein, L. A., Thun, M. J ., Mowery, P., Pierce,
3., Escobedo, L., & Giovino, G. A. (2000). American Journal of Epidemiology, 152(6),
573-584.

MathSoft, Inc. (1998). S-Plus for Windows professional (version 4.5). Seattle,
WA.

Mendez, D., Warner, K. E., & Courant, P. N. (1998). Has smoking cessation
ceased? Expected trends in the prevalence of smoking in the United States. American
Journal of Epidemiology. 148(3), 249-258.

Milberger, S., Biederrnan, J ., & F araone, S. V. (1997). Dr. Milberger and
colleagues reply. American Journal of Psychiatry. 154(8), 1177-1178.

244

Millberger, S., Biederman, J ., Faraone, S. V., Chen, L., & Jones, J. (1996). Is
maternal smoking during pregnancy a risk factor for attention deficit hyperactivity
disorder in children? American Journal of Psychiatry, 153(9), 1138-1142.

Millberger, S., Biederman, J ., Faraone, S. V., Chen, L., & Jones, J. (1997).
ADHD is associated with early initiation of cigarette smoking in children and
adolescents. Journal of American Academy of Child and Adolescent Psychiatry, 36(1),
37-44.

Mun, E. Y., von Eye, A., Fitzgerald, H. E., & Zucker, R. A. (2001). Using mosaic
displays in Configural Frequency Analysis. Methods of chhological Research-Online,
6(3), 164—196.

Muthén, B. (2001). Second-generation structural equation modeling with a
combination of categorical and continuous latent variables. In L. M. Collins & A. G.
Sayer (Eds), New methods for the analysis of changg (pp. 179-200). Washington DC:
American Psychological Association.

Muthén, B., & Muthén, L. K. (2000). Integrating person-centered and variable-
centered analyses: Growth mixture modeling with latent trajectory classes. Alcoholism:
Clinical and Eaperimental Reseapch. 24(6), 882-891.

Nagin, D. S. (1999). Analyzing developmental trajectories: A semiparametic,
group-based approach. Psycholgical Methods, 4(2), 139-157.

Nagin, D. S., & Tremblay, R. E. (1999). Trajectories of boys' physical
aggression, opposition, and hyperactivity on the path to physically violent and nonviolent
juvenile delinquency. Child Development, 70(5), 1181-1196.

Nagin, D. S., & Tremblay, R. E. (2001). Analyzing developmental trajectories of
distinct but related behaviors: A group-based method. Psychological Methods, 6(1), 18-
34.

Noll, R.B., Zucker, R.A., Fitzgerald, H.E., & Curtis, W.J. (1992). Cognitive and
motoric functioning of sons of alcoholic fathers and controls: The early childhood years.
Developmental Psychology, 28(4), 665-675.

O'Connor, T. G., & Rutter, M. (1996). Risk mechanisms in development: Some
conceptual and methodological considerations. Developmental Psychology, 32(4), 787-
795.

Patton, G. C., Carlin, J. R, Wolfe, C. C. R., Hibbert, M., & Bowes, G. (1998).

The course of early smoking: A population-based cohort study over three years.
Addiction. 93(8), 1251-1260.

245

Picciotto, M. R. (1998). Common aspects of the action of nicotine and other
drugs of abuse. Drug and Alcohol Dependence. 51. 165-172.

Piejak, L. A., Twitchell, G. R., Loukas, A., Fitzgerald, H. E., & Zucker, R. A.
(1996). Paternal alcoholism, personality characteristics and antisociality: Influences on
child behavior problem (Abstract). Alcoholism: Clinical and Experimental Research, 20,
473A.

Piotrowski, C., Sherry, D., & Keller, J. W. (1985). Psychodiagnostic test usage:
A survey of the Society for Personality Assessment. Journal of Personality Aasessment.
fl, 1 15-1 19.

Pomerleau, C. S. (1997). Co-factors for smoking and evolutionary
psychobiology. Addiction, 92(4), 397-408.

Pomerleau, O. F. (1995). Neurobiological interactions of alcohol and nicotine. In
J. B. F ertig & J. P. Allen (Eds), Alcohol and tobacco: From basic science to clinical
practice flip. 145-158). Bethesda, MD: National Institute on Alcohol Abuse and
Alcoholism.

Pomerleau, O. F ., Collins, A. C., Shiffinan, S., & Pomerleau, C. S. (1993). Why
some people smoke and others do not: New perspectives. J oumal of Consultingand
Clinical Psychology, 61(5), 723-731.

Pomerleau, O. F., Downey, K. K., Stelson, F. W., & Pomerleau, C. S. (1995).
Cigarette smoking in adult patients diagnosed with attention deficit hyperactivity
disorder. Journal of Substance Abuse. 7. 373-37 8.

Prescott, C. A., & Kendler, K. S. (1995). Genetic and environmental influences
on alcohol and tobacco dependence among women. Alcoholjand Tobacco: From Ba_sig
Science to Clinical Practice. Bethesda, MD, National Institute on Alcohol Abuse and
Alcoholism, Research Monograph 30.

Raykov, T., Tomer, A., & Nesselroade, J. R. (1991). Reporting structural
equation modeling results in Psychology and Aging: Some proposed guidelines.

Psychology and Aging, 6(4), 499-503.

Reider, E. E. (1991). Relationships between parental psychopathology, family
conflict. and child behavior problems in young alcoholic families. Unpublished doctoral
dissertation. East Lansing, MI: Michigan State University.

Riggs, P. D., Mikulich, S. K., Whitrnore, E. A., & Crowley, T. J. (1999).
Relationship of ADI-ID, depression, and non-tobacco substance use disorders to nicotine
dependence in substance-dependent delinquents. Drugand Alcohol Dependence. 54.
195-205.

246

Robins, L., Helzer, J ., Croughan, J ., & Ratcliffe, K. S. (1980). The NIMH
Diagostic Interview Schedule: Its historyacharacteristics and glidity. St. Louis, MO:
Washington University School of Medicine.

Rothbart, M. K., & Bates, J. E. (1998). Temperament. In W. Damon & N.
Eisenberg (Eds), Handbook of child psmhology: Social, emotionalaand personality
development (5 ed., Vol. 3, pp. 105-176). New York, NY: John Wiley & Sons, Inc.

 

Rowe, D. C., Chassin, L., Presson, C., & Sherman, S. J. (1996). Parental smoking
and the "Epidemic" spread of cigarette smoking. Journal of Applied Social Psychology,
2_6_(5), 437-454.

Rowe, D. C., & Rodgers, J. L. (1991). Adolescent smoking and drinking: Are
they "epidemics"? Journal of Studies on Alcohol. 52. 110-117.

Rowe, D. C., Rodgers, J. L., & Gilson, M. (2000). Epidemics of smoking:
Modeling tobacco use among adolescents. In J. S. Rose, L. Chassin, C. C. Presson, & S.
J. Sherman (Eds), Multivariate applications in substance use research: New methods for
new questions (pp. 233-258). Mahwah, NJ: Lawrence Erlbaum Associates.

Rubin, D. B. (1987). Multiple imputation for nonrespgpse in survey. New York:
Wiley.

Sabol, S. 2., Nelson, M. L., Fisher, C., Gunzerath, L., Brody, C. L., Hu, 8., Sirota,
L. A., Marcus, S. E., Greenberg, B. D., Lucas IV, F. R., Benjamin, J ., Murphy, D. L., &
Hamer, D. H. (1999). A genetic association for cigarette smoking behavior. Health
Psychology, 18(1), 7-13.

SAS Institute Inc. (1999). The SAS system for the Windows (version 8). Cary,
NC: SAS Institute Inc.

Schafer, J. L. (1997). Analfiis of incomplete multivariate data. London:
Chapman & Hall.

Schafer, J. L. (1999). Irnputation of missing cova_riates under a multivariate linear
mixture model. Technical report, Department of Statistics, Pennsylvania State
University.

Schafer, J. L. (2000). NORM: Multiple imputation of incomplete multivariate
alga under a normal model, version 3.03. Department of Statistics, Pennsylvania State
University.

Schafer, J. L. (2001). Multiple imputation with PAN. In L. M. Collins, & A. G.
Sayer (Eds), New methods for the analfiis of change (pp. 355-378). Washington DC:
American Psychological Association.

247

 

Schuckit, M. A. (1978). Research apestionnaira Alcoholism Treatment
Program, VA Medical Center. University of California, San Diego.

Schulenberg, J ., O'Malley, P. M., Bachman, J. G., Wadsworth, K. N., & Johnston,
L. D. (1996). Getting drunk and growing up: Trajectories of frequent binge drinking
during the transition to young adulthood. Journal of Studies on Alcohol, 57, 289-304.

Selzer, R., Vinokur, A., & van Rooijen, L.(1975). A self-administered Short
Michigan Alcohol Screening Test. Journal of Studies on Alcohol. 36. 117-126.

Sher, K. J ., Gotham, H. J ., Erickson, D. J ., & Wood, P. K. (1996). A prospective,
high-risk study of the relationship between tobacco dependence and alcohol use
disorders. Alcoholism: Clinical and Experimental Research, 20(3), 485-492.

Shiffinan, S., & Balabanis, M. (1995). Associations between alcohol and tobacco.
In J. B. Fertig & J. P. Allen (Eds), Alcohol and tobacco: From basic science to clinical
practice (pp. 17-36). Bethesda, MD: National Institute on Alcohol Abuse and
Alcoholism.

Shiffrnan, S., Kassel, J ., Paty, J ., Gnys, P., & Zettler-Segal, M. (1994a). Smoking
typology profiles of chippers and regular smokers. Journal of Substance Abuse. 6. 21-35.

Shiffman, S., Paty, J. A., Kassel, J. D., Gnys, M., & Zettler-Segal, M. (1994b).
Smoking behavior and smoking history of tobacco chippers. Experimental and Clinical
Psychophaanacology, 2(2), 126-142.

SPSS Inc. (1999). SPSS for Windows (version 10). Chicago, IL: SPSS Inc.

Steer, R. A., Beck, A. T., & Garrison, B. (1986). Applications of the Beck
Depression Inventory. In N. Sartorius, & T. A. Ban (Eds), Assessment of depreasion
(pp. 121-142). Geneva, Switzerland: World Health Organization.

Stevens, G., & Featherrnan, D. L. (1981). A revised socioeconomic index of
occupational status. Social Science ResearchLIO, 364-395.

Tarter, R. E., & Vanyukov, M. (1994). Alcoholism: A developmental disorder.
Journal of Consulting and Clinical Psychology, 62(6), 1096-1107.

Tizabi, Y., Overstreet, D. H., Rezvani, A. H., Louis, V. A., Clark, Jr., E.,
J anowsky, D. S., & Kling, M. A. (1999). Antidepressant effects of nicotine in an animal
model of depression. Psychopharmacology, 142, 193-199.

Townsend, M. Z. (1998). Influences of matemtl drug involvemenL
psychopathology and parenting on preschoolersadevelopmental status. 10 Ed behavior
problems. Unpublished dissertation. Department of Psychology, Michigan State
University, East Lansing, MI.

 

248

True, W. R., Heath, A. C., Scherrer, J. F., Waterman, B., Goldberg, J ., Lin, N.,
Eisen, S. A., Lyons, M. J ., & Tsuang, M. T. (1997). Genetic and environmental
contributions to smoking. Addiction, 92(10), 1277-1287.

US. Department of Health and Human Services. (1988). The health
consequences of smaking: Nicotine addiction, 3 report of the surgeon general,
Washington, DC: US. Government Printing Office.

US. Department of Health and Human Services. (1994). Preventing tobacco use
among young people: A report of the Surgeon General. Washington, DC: US.
Government Printing Office.

US. Department of Health and Human Services. (2001). Women and smoking;
A rgport of the Surgon General. Washington, DC: US. Government Printing Office.

von Eye, A. (1990). Introduction to configural fiequencLanalysis: The search for
types and antitypes in cross-classification. New York: Cambridge University Press.

von Eye, A. (2001a). Configural Frequency Analysis - Version 2000: A program
for 32 bit Windows operating systems. Methods of Psychological Research-Online, 6(2),
129-139.

von Eye, A. (2001b). Personal communication. Department of Psychology,
Michigan State University. East Lansing, MI.

von Eye, A. (2002). Configu_ral Frequency Analfiis Mahwah, NJ: Lawrence
Erlbaum Associates, INC.

Wakschlag, L. S., Lahey, B. B., Loeber, R., Green, S. M., Gordon, R. A., &
Leventhal, B. L. (1997). Maternal smoking during pregnancy and the risk of conduct
disorder in boys Archives of General Psychiatry. 54. 670-676.

Weissman, M. M., Warner, V., Wickramaratne, P. J., & Kandel, D. B. (1999).
Maternal smoking during pregnancy and psychopathology in offspring followed to
adulthood. Journal of American Academy of Child and Adolescent Psychiatry. 38(7),
892-899.

Weitzman, M., Gortrnaker, S., & Sobol, A. (1992). Maternal smoking and
behavior problems of children. Pediatrics, 90(3), 342-349.

Whalen, C. K., Jamner, L. D., Henker, B., Delfino, R. J ., & Lozano, J. M. (2002).

The ADHD spectrum and everyday life: Experience sampling of adolescent moods,
activities, smoking, drinking. Child Development, 73(1), 209-227.

249

 

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), 215-226.

Wohlwill, J. (1973). The studyof behavioral development. San Diego, CA:
Academic Press.

Zucker, R. A. (1987). The four alcoholisms: A developmental account of the
etiologic process. In P. C. Rivers (Ed), Nebraska symposium on motivation: Alcohol
and addictive behaviors (Vol. 34, pp. 27-83). Lincoln, NE: University of Nebraska Press.

Zucker, R. A. (1994). Pathways to alcohol problems and alcoholism: A
developmental account of the evidence for multiple alcoholisms and for contextual
contributions to risk. In R. A. Zucker, J. Howard, & G. M. Boyd (Eds), m
development of alcohol problems: Eigloring the biopsychosocial matrix of risk (NIAAA
Research Monograph No. 26). (pp. 255—289, Chapt. 13). Rockville, MD: US.
Department of Health and Human Services.

Zucker, R. A., Chermack, S. T., & Curran, G. M. (2000). Alcoholism: A life Span
perspective on etiology and course. In A. J. Sameroff, M. Lewis, & S. M. Miller (Eds),
Handbook of developmental psachopathology, 2nd ed (pp. 569—5 87). New York: Kluwer
Academic/Plenum Publishers.

Zucker, R. A., Ellis, D. A., Bingham, C. R., & Fitzgerald, H. E. (1996). The
development of alcoholic subtypes. Alcohol Health & Research World. 20(1), 46-54.

Zucker, R. A., Ellis, D. A., Fitzgerald, H. E., Bimgharn, C. R., & Sanford, K.
(1996). Other evidence for at least two alcoholisms H: Life course variation in
antisociality and heterogeneity of alcoholic outcome. Development aria
Psychopathology; 83 1-848.

Zucker, R. A., & Fitzgerald, H. E. (1991). Early developmental factors and risk
for alcohol problems. Alcohol Health and Research World. 15. 18-24.

Zucker, R. A., Fitzgerald, H. E., & Moses, H. D. (1995). Emergence of alcohol
problems and the several alcoholisms: A developmental perspective on etiologic theory
and life course trajectory. In D. Cicchetti & D. J. Cohen (Eds), Developmental
psychopatholalgy: Vol. 2. Risk, disorder and adaptation (pp. 677-711). New York:
Wiley.

Zucker, R. A., Fitzgerald, H. E., & Noll, R. (1990). Drinking and drughistory
(rev. ed., version 4). Unpublished instrument, Michigan State University, East Lansing,
MI.

Zucker, R. A., Fitzgerald, H. E., Refior, S. K., Puttler, L. 1., Pallas, D. M., & Ellis,
D. A. (2000). The clinical and social ecology of childhood for children of alcoholics:

250

 

 

Description of a study and implications for a differentiated social policy. In H. E.
Fitzgerald, B. M. Lester, & B. S. Zuckerman (Eds), Children of addiction: Research,
health, and policy issues (pp. 109-142). New York: Garland Press.

 

Zucker, R. A., Noll, R. B., Ham, H., Fitzgerald, H. E., & Sullivan, L. S. (1994).
Assessiyaantisociality with the Antisocial Behavior Checklist: Reliability and validity
studies. East Lansing (Michigan State University) and Ann Arbor (University of
Michigan), MI: MSU-UM Longitudinal Study.

 

251

 

MICHIGAN STATE mam IES
l

lllllllll lllllll‘ll llllllll ll

3 1293 02320 697-