 

Leela

 

LlBFiARY
Michigan State
University

 

 

 

This is to certify that the
dissertation entitled

DO FAMILY RISK FACTORS INFLUENCE ADHD VIA
DISRUPTION OF NEUROCOGNITIVE FUNCTIONING?

presented by

Torri Wynette Miller

has been accepted towards fulfillment
of the requirements for the

 

 

 

 

 

Doctoral degree in PsycholggL
/Maj4rofe§s‘ofs$ignature
7/3/Af’
/ /
Date

MSU is an afﬁrmative-action, equal-opportunity employer

 

 

 

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 KIProj/Aoc&Pres/ClRC/DateDue.indd

 

DO FAMILY RISK FACTORS INFLUENCE ADHD VIA DISRUPTION OF
NEUROCOGNITIVE FUNCTIONING?

By

Torri Wynette Miller

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Department of Psychology

2008

ABSTRACT

DO FAMILY RISK FACTORS INFLUENCE ADHD VIA DISRUPTION OF
NEUROCOGNITIVE FUNCTIONING?

By
Torri Wynette Miller

Family adversity has long been associated with maladaptive child behavior, yet
few studies have examined mediators and moderators of family effects on Attention
Deﬁcit Hyperactivity Disorder (ADHD). The current study evaluated the inﬂuence of
family risk factors on ADI—[D symptoms via mediating cognitive processes (response
inhibition and reaction time variability). Participants, 394 children between the ages of 6-
18 years, completed rating scales and cognitive tasks. Parents and teachers also provided
ratings. The study replicated the effect of perceived interparental conﬂict on ADHD
symptoms (Counts, Nigg, Stawicki, Rappley, & von Eye, 2005). The study validated a
composite family risk construct that included perceived interparental conﬂict, maternal
depression, and socioeconomic status. Cognitive regulation partially mediated the
association between family risk and teacher-reported ADHD symptoms. The partial
mediation of cognitive regulation was present in adolescents and not in children, in
Caucasians and not in ethnic minorities. Further, ADHD symptoms partially mediated the
relationship between family risk and adaptive functioning. Overall, the study supported
the role of cognitive processes in the effect of family risk factors on ADHD, with
moderated effects of developmental stage and race. Finally, family risk is associated with
unique impairment related to ADHD symptoms. Clinical implications of these ﬁndings

are discussed.

Copyright by
TORRI WYNETTE MILLER
2008

My deepest thanks are extended to my mother, advi sor, and supporters for your
encouragement, guidance, and inspiration.

ACKNOWLEDGEMENTS

I would like to thank members of my committee for their feedback and
recommendations throughout this process: Joel Ni gg, Ph.D., Alytia Levendosky, Ph.D.,
Frederick Leong, Ph.D., and Ruben PaJTa Cardona, PhD. I would also like to extend
gratitude to the staff and research assistants at the MSU Attention Study for their hard
work and dedication to the project. This work was supported by NIH grant R01-
MH63146. In addition, I am eternally grateful to Alexander von Eye, Ph.D. for
consultation regarding statistical analyses. Finally, I am thankful for the steadfast support

of my family and ﬁ'iends.

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... viii
LIST OF FIGURES ............................................................................................................ ix
INTRODUCTION ............................................................................................................... 1

A THEORETICAL APPROACH TO HOW THE FAMILY MAY INFLUENCE

Developmental Psychopathology ............................................................................. 2
Mediating Cognitive Processes ................................................................................ 4
Response Inhibition ..................................................................................... 5
Response Variability .................................................................................... 6
Cognitive Functioning and Family Risk in ADHD ..................................... 7
IQ and ADHD .............................................................................................. 8
Summary ...................................................................................................... 8
Conceptual Complexities ......................................................................................... 9
Summary .................................................................................................... 10
SPECIFIC RISK FACTORS AT ISSUE ........................................................................... 11
Speciﬁc Family Risk Factors Related to ADHD ................................................... I 1
Adversity in the Family Environment ........................................................ l 1
Family Conﬂict .......................................................................................... 12
Parent Risk Factors ................................................................................................ 14
Parent ADHD ............................................................................................. 14
Maternal Distress/Depression .................................................................... 15
Culture as a Moderator of Family Risk and ADHD .............................................. 16
Global Adaptive Functioning ................................................................................. 19
Possible Confounds ................................................................................................ 20
Health Risks ........................................................................ 20
Hypotheses ........................................................................................................ 22
METHOD .......................................................................................................................... 23
Participants ............................................................................................................ 23
Sample and Recruitment ........................................................................................ 23
Procedure Overview ............................................................... 23
Diagnostic Measures Pertaining to the Child ......................................................... 24
Diagnostic Dimensions of ADHD in the Child ......................................... 24
Assessment of Other Axis I Disorders in the Child ................................... 25
Assessment of Parent Psychopathology ................................................................. 25
Parent ADI-II) ............................................................................................. 26

vi

Parent Depression ...................................................................................... 26

Child Cognitive Regulation ................................................................................... 27
Logan Stop Task ........................................................................................ 27
Full Scale IQ .............................................................................................. 27
Remaining Measures Related to Family Adversity ............................................... 28
Demographics ............................................................................................ 28
Family Environment Scale ......................................................................... 28
Children’s Perception of Interparental Conﬂict Scale ............................... 30
Multigroup Ethnic Identity Measure .......................................................... 3]
Developmental History .............................................................................. 32
Adaptive Functioning ................................................................................. 33
Data Analysis ......................................................................................................... 33
RESULTS .......................................................................................................................... 35
Sample Description ................................................................................................ 35
Power Analyses ...................................................................................................... 36
Hypothesis 1a: Selected family measures would statistically predict ADHD
symptoms independent of parent ADHD, one another, ODD, and CD ................. 36
Hypothesis 1b: The same family measures would predict inattentive and
hyperactive symptoms ........................................................................................... 37
Hypothesis 2a: The family measures could be consolidated to form a latent family
risk construct .......................................................................................................... 38
Hypothesis 2b: ADHD would be signiﬁcantly related to the latent family risk
construct ................................................................................................................. 39
Hypothesis 2c: Family risk would predict ADHD when items are used to create
latent factors for ADHD symptom domains ........................................................ 41
Hypothesis 3a: Cognitive regulation would partially mediate the relationship
between family risk and ADHD ............................................................................ 42
Hypothesis 3b: Child age would moderate the partial mediation described in
3a ............................................................................................................................ 42
Hypothesis 3c: Race would moderate the partial mediation described in
3a ............................................................................................................................ 43
Hypothesis 4: ADHD would partially mediate the relationship between family
risk and adaptive functioning ................................................................................. 44
DISCUSSION .................................................................................................................... 45
Preliminary Findings ....................................................................... 45
Main Findings .............................................................................. 49
Clinical Implications .............................................................................................. 53
Limitations ............................................................................................................. 54
Conclusion ............................................................................................................. 55
APPENDICES ................................................................................................................... 56
APPENDIX A: Main Tables and Figures ............................................... 57
APPENDIX B: Family Environment Scale Analyses ................................. 76
APPENDIX C: CPIC Factor Structure .................................................. 81

vii

APPENDIX D: Alternative Regression Analyses for Hypothesis Ia ............... 82
APPENDIX E: Bootstrapping Method (Hypothesis 3a) ............................... 84

REFERENCES .................................................................................................................. 87

viii

LIST OF TABLES

TABLE 1a. Description of Sample .................................................................................... 57
TABLE 1b. Description of Household Demographics Across Groups ...................... 58
TABLE 2. Regression Model: Teacher-rated ADHD Symptoms Regressed Onto

Predictors of Risk (Hypothesis la) .................................................................................... 59
Table B1. Family Environment Scale EF A 1—factor Solution, 10 Subscales ............... 76
Table BZ. Family Environment Scale EFA 2-factor Solution, 10 Subscales ............... 77
Table B3. Family Environment Scale EFA 4-factor Solution, 10 Subscales ................ 78

Table B4. Family Environment Scale EF A 2-factor Solution, 6 Subscales using FIML
Imputation ............................................................................................ 79

Table B5. Family Environment Scale EF A 2-factor Solution, 6 Subscales using Raw
Data ................................................................................................... 79

Table Cl. CPIC Factor Structure Validated by Ni gg et al. (under review) .................. 81

Table D1. Teacher-rated ADHD Symptoms Regressed Onto Predictors of Risk, With
ODD & CD in Step 2 ............................................................................... 82

Table D2. Teacher-rated ADHD Symptoms Regressed Onto Predictors of Risk, Without
ODD & CD ........................................................................................... 83

LIST OF FIGURES

FIGURE 1. CFA Model of Family Risk (Hypothesis 2a) ................................................. 60
FIGURE 2. SEM of Family Risk Regressed onto ADHD Total Symptoms and ODD
(Hypothesis 2b) ....................................................................................... 62
FIGURE 3. SEM of ADHD Total Symptoms Regressed onto Family Risk (Hypothesis
2b) .......................................................................................................... 63
FIGURE 4. SEM of CTRS Items Regressed onto Family Risk (Hypothesis 2c) .............. 64
FIGURE 5. Mediation Model using Clustering, without Bootstrap Method (Hypothesis
3a) .......................................................................................................... 66
FIGURE 6. Mediation Model using Clustering, and Multigroup Age Analyses —- Child
Group (Hypothesis 3b) .............................................................................. 68
FIGURE 7. Mediation Model using Clustering, and Multigroup Age Analyses —
Adolescent Group (Hypothesis 3b) ................................................................ 69
FIGURE 8. Mediation Model using Clustering, and Multigroup Analyses — Caucasian
Group (Hypothesis 3c) .............................................................................. 70
FIGURE 9. Mediation Model using Clustering, and Multigroup Analyses — Ethnic
Minority Group (Hypothesis 3c) .................................................................... 72
FIGURE 10. ADHD Mediation of Family Risk and Adaptive Functioning (Hypothesis

4) ................................................................................................................................ 74
Figure B1. CF A of FES 2-F actor Solution with 6 Subscales ................................. 80
Figure E1. Mediation using Bootstrap Method, without Clustering ......................... 85

Do Family Risk Factors Inﬂuence ADHD via Disruption of Neurocognitive Functioning?

Attention deﬁcit hyperactivity disorder (ADHD) is characterized by a persistent
pattern of inattentive, hyperactive, and impulsive behaviors that begin in early childhood,
often persist throughout development, and interfere with adaptive functioning (APA,
2000). Childhood ADHD affects between 3-7% of school-aged children in the United
States (APA, 2000), and is more prevalent in males (APA, 2000; Carlson & Mann, 2000;
Gaub & Carlson, 1997). Key domains of ADHD include the predominantly
hyperactive/impulsive subtype (ADHD-H), predominantly inattentive subtype (ADHD-I),
and combined type (ADHD-C) which includes both hyperactive/impulsive and
inattentive symptoms. To meet DSM—IV criteria for ADHD, a child must evidence
impairment from the symptoms, before age 7, in two or more settings (e. g. home and
school) (APA, 2000). ADI-II) overlaps with disruptive behaviors including Oppositional
Deﬁant Disorder (ODD) and Conduct Disorder (CD), such that approximately 30-45% of
children with ADHD also have ODD and approximately 20% of children with ADHD
also have CD (Acosta, Arcos-Burgos, & Muenke, 2004; Faraone & Biederman, 1998).

Whereas liability to ADHD is largely inﬂuenced by genetic factors (Faraone et
al., 2005), ADHD expression is also associated with psychosocial risk (Biederman,
Milberger, Faraone, Kiely, & et al., 1995a; Pressman et al., 2006; Scahill et al., 1999). To
date, the latter is markedly under-investigated. ADHD is characterized by a
heterogeneous behavioral proﬁle, which may be explained by different interactions and
correlations between genes and the environment. Researchers have therefore suggested
that there are multiple causal pathways associated with ADHD (N igg, Goldsmith, &

Sachek, 2004).

One pathway involves the family environment, which may confer both genetic
and environmental risk to children toward the development of ADHD. The potential
effect of family risk on processes involved in cognitive development is of particular
interest in the present investigation. Abnormalities in a range of cognitive operations
have been identiﬁed in samples with ADHD. For example, deﬁcits in executive
functioning are common in groups of children with ADHD (N igg, Willcutt, Doyle, &
Sonuga-Barke, 2005; Sonuga—Barke, 2005). A key question is whether family risk factors
operate through these problems. If family risk factors are associated with ADHD via
cognitive mechanisms, another important question is whether this relationship changes at
different stages of child development.

Historically, studies of ADHD examined Caucasian males (Barkley, Fischer,
Smallish, & Fletcher, 2004; Loney, Kramer, & Milich, I981; Mannuzza, Klein, Bessler,
Malloy, & LaPadula, 1993). Though that trend remains a common characteristic in the
literature, studies have begun to incorporate cultural variation. Culture-speciﬁc effects
have been observed in regard to childhood externalizing behavior problems (Deater-
Deckard, Dodge, Bates, & Pettit, 1996), and associated risk factors (Lynch, 2003).
However, little is known about differential cultural effects for ADHD, in particular. The
present study, as a secondary aim, examined whether the relationship between family
risk, cognitive regulation, and ADHD changed across racial groups (using race as a proxy
for cultural variation).

A Theoretical Approach to How the Family May Inﬂuence ADHD

Developmental Psychopathology

A developmental psychopathology perspective (Rutter & Sroufe, 2000) suggests
that maladaptive outcomes emanate from multiple risk and protective factors interacting
over time (Sroufe, 1997). Mash and Johnston (2005) outline several pathways that
connect ADHD to the family: biological predisposition (Faraone et al., 2005), shared
genetic risk (Epstein et al., 2000), parenting and family environmental inﬂuences on
biological/cognitive programming (O'Conner, 2003), family conﬂict due to ADHD
(Biederman, Faraone, & Monuteaux, 2002; Lange et al., 2005; Pressman et al., 2006),
family as a moderator and/or mediator of ADI-II) outcomes (Counts, Nigg, Stawicki,
Rappley, & von Eye, 2005; Miller et al., 2006), and the family as a moderator and/or
mediator of conditions that are commonly comorbid with ADHD such as conduct
problems (Thapar, van den Bree, Fowler, Langley, & Whittinger, 2006).

This framework suggests that ADHD may be inﬂuenced by family factors that
carry different levels of inﬂuence across children and over time (Johnston & Mash,
2001). Two points need to be considered in properly conceptualizing the family in
relation to a high heritability disorder such as ADHD. First, causal processes likely
involve both genetic and environmental factors via gene-environment interactions
(individuals at genetic risk may be more susceptible to environmental risks) and gene-
environment correlations (individuals at genetic risk may be more prone to exposure to
environmental risk) (Plomin & Ruter, 1998). Although ADHD has high heritability of
liability (in the range of .7 to .8 in parent ratings; Faraone et al., 2005), monozygotic twin
concordance rates do not approach 100% (Kuntsi & Stevenson, 2000). In addition to
direct environmental effects on behavior, this may be due to differences in the expression

of genes through epigenetic variation (Rutter, Mofﬁtt, & Caspi, 2006).

Epigenetic mechanisms such as methylation allow environmental inﬂuences to
affect gene expression — not by altering the gene sequence, but by affecting the
consequences of genes such that the genes are interpreted or expressed to differing
degrees (Rutter, Moffitt, & Caspi, 2006). Thus, gene expression can be considered as a
continuum such that an individual could exhibit behavioral variation due to
environmental stress, mediated through variations in gene expression. Thus, events in the
family environment could participate in epigenesis. Clarifying this possibility ﬁrst
requires additional mapping of the nature of family inﬂuences on ADHD. The present
study examined family risk factors associated with ADHD in prior research, and clariﬁed
these relationships using a larger sample and more systematically than has been done
previously.

Second, the family environment may be differentially important during critical
developmental periods. For example, it may alter brain organization through a process
called developmental programming, which occurs during sensitive periods of brain
development wherein biological systems are thought to respond sensitively to
environmental stimuli (O'Conner, 2003). It has long been recognized that experiences
inﬂuence neural structures and processes that affect behavior (Hebb, 1949). These
experiences include external as well as internal occurrences ranging from social events to
psychological trauma, and affect short-term and long-term outcomes (Cicchetti & Curtis,
2006). A key question is whether young children evidence different behavioral effects of
family risk than adolescents.

Mediating Cognitive Processes

Family experiences presumably can result in alterations in gene expression via
modiﬁcation of neuronal and synaptic patterns, which affect cognition and, by that route
among others, behavioral change (Cicchetti & Curtis, 2006). There is growing
recognition and empirical basis for the effects of parental attunement and caregiver
interchange on early cognitive development. This is especially the case for language
development (Appelbaum, 1977; LaParo, Justice, Skibbe, & Pianta, 2004), but it is also
emerging for regulatory control (Dawson, Panagiotides, Klinger, & Spieker, 1997;
Goodman & Gotlib, 1999). Thus, the argument here is that family risk factors can lead to
subtle deﬁcits in child cognition, which can in turn contribute to ADI-II). However, the
effects of experiences on cognitive processing are not well studied in relation to ADHD.
The present study examined these effects to gain better understanding of how family
context relates to cognitive effects in ADHD, which may help pinpoint neurological
mechanisms that shape ADHD behaviors. We now turn to a description of cognitive
processes of interest in this investigation.

Many cognitive functions have been implicated in ADHD (Barkley, 1997;
Klorman et al., 1999; Tannock, Martinussen, & Frijters, 2000). Two of the most
important theorized cognitive mechanisms involved in ADHD are response inhibition and
response variability (N igg, 2005). Both of these mechanisms are early emerging and
develop throughout childhood, and so would be expected to be related to the development
of regulatory control in the family context. Each construct will be discussed in turn.

Response Inhibition. Effortful response suppression is well associated with
ADHD (Barkley, 1997; Crosbie & Schachar, 2001; Nigg, 1999). Also referred to as

inhibitory control, this executive control mechanism involves restraining a planned action

based on a sudden change in the context (e. g. stopping oneself before running out into the
street because a car has just appeared) (Schachar & Logan, 1990). The ability to inhibit
follows developmental progressions of synaptic pruning and reinforcement of speciﬁc
brain networks, namely, frontal striatal circuitry (F uster, 2002). There is evidence for
age—related change in inhibitory control across development (Booth et al., 2003; Rubia et
al., 2006; Williams, Ponesse, Schachar, Logan, & Tannock, 1999). Response inhibition
can be assessed experimentally in numerous ways. One very well accepted method is the
stop-signal experimental paradigm (Logan, Schachar, & Tannock, 1997). Approximately
30 studies of ADHD have been conducted using this measure; based on that body of
work, the main effect association of slower stop signal response with ADHD is well
established (weighted effect size =.61; Willcutt et al., 2005). The present study
investigated how performance on this task related to family risk as part of a potential
process model involving family effects on neural and hence cognitive development.
Response Variability. The second cognitive mechanism targeted here, response
variability, also has been associated with ADHD in several studies (Castellanos &
Tannock, 2002; Epstein et al., 2006; Rapport, Chung, Shore, Denney, & Issacs, 2000).
Greater response variability may be interpreted as corresponding to deﬁcits in arousal
(Parasuraman, Warm, & See, 1998), which are related to higher levels of inattention. The
ability to maintain an alert state is linked to ascending noradrenergic networks in the
brain, which are involved in arousal (Coull, 1998; Posner & Petersen, 1990). This ability
develops through childhood (Akshoomoff, 2002; Booth et al., 2003; Levy, 1980), and is
related to wakeﬁJlness, which is relevant to early caregiver attunement and parent-child

interaction (Frolich & Lehmkuhl, 2004; Kim, Brody, & Murry, 2003). In the present

study, this construct was measured using response time variability on the Logan Stop
Task to understand family effects on cognitive regulation. Response inhibition and
response time variability are both key aspects of higher order regulatory operations, and
were therefore aggregated in this study to create a composite cognitive regulation score.

Cognitive Functioning and Family Risk in ADHD. Despite the importance of
family ﬁrnctioning, few studies have examined it in relation to cognitive ﬁanctioning in
ADHD. However, a handﬁrl of studies have examined the stop task and families ——
although not in the same way that the present study has. Crosbie and Schachar (2001)
found that children with poor inhibition had families with higher ADHD prevalence than
children with good inhibition or controls (N=80). However, they did not ﬁnd group
differences in psychosocial risk (which included factors such as living in overcrowded
arrangements, subsidized housing, a single-parent family, or experience of parental
separation before the age of 2) or neurobiological risk (including high blood pressure,
seizures, nausea, perinatal bleeding, use of drugs or alcohol, or infection). Thus, the
authors concluded that the increased risk of a positive family history of ADHD among
the poor inhibition group was likely due to cognitive risk rather than psychosocial or
neurobiological risk.

A small number of other kinds of cognitive risk have been examined in relation to
families. Carlin Miller et al. (2006) examined family and cognitive risk factors for
aggression in 7-11 year-old children with ADHD. A latent variable created to represent
cognitive ﬁrnctioning included full scale IQ (F SIQ) assessed by the Wechsler Interview
Schedule for Children, Revised Edition (WISC-R) or WISC, Third Edition (WISC—III),

and general math and reading achievement scores measured by the Wide Range

Achievement Test, Revised Edition (WRAT-R) or the Wechsler Individual Achievement
Test (WIAT). The family risk factors included demographic information such as number
of siblings in the home, number of caregiving adults in the home, and history of
aggression in biological parents of target child. Aggression was measured at home and
school using the Children’s Aggression Scale — Parent and Teacher Versions (CAS-P and
CAS-T) to assess verbal, provoked physical, initiated physical, and object-directed
aggression. Family risk had an effect on both home and school aggression, but cognitive
risk was only related to school aggression. This suggests that the cognitive risk factors
they assessed do not extend into the family domain to result in signiﬁcant functional
impairment observable by parents in the home setting. Alternatively, these results could
suggest that school is associated with different cognitive and behavioral demands on
children than home.

IQ and ADHD. In clinical samples, children with ADI-1]) tend to have lower IQ
scores than controls (Barkley, Karlsson, Pollard, & Murphy, 1985; McGee, Williams, &
F eehan, 1992; Moffrtt, 1990). The extent of the difference in IQ in these studies ranges
from 7 to 10 points. ADHD severity and IQ were negatively associated in community
(Hinshaw, Morrison, Carte, & Comsweet, 1987; Peterson, Pine, Cohen, & Brook, 2001)
and clinical samples (Sonuga-Barke, Lamparelli, Stevenson, Thompson, & Henry, 1994)
with correlations ranging from r= -.25 to -.35. Thus, IQ was controlled in analyses herein
to prevent a possible confound with ADHD effects.

Summary. A logical next step is to pursue family effects in relation to speciﬁc
cognitive domains (i.e. response inhibition and vigilance) that are believed to be at the

core of ADHD in how it develops. The present study examined mediational pathways to

better understand the relationship between family risk and ADHD symptom presentation.
In this model, the cognitive component serves as an index of alterations in neural
systems, which support ﬁinctions such as response inhibition and vigilance. Family
processes are hypothesized to inﬂuence ADHD symptoms. They are also hypothesized to
inﬂuence cognitive development. In turn, the effects on cognition are hypothesized to
predict ADHD symptoms. The development of self regulation, response control, and
arousal are thought to depend on caregiver interchanges early in development and to
mature throughout childhood (Kochanska & Aksan, 2006; Panzer & Viljoen, 2005;
Tucker, Luu, & Derryberry, 2005). Thus, it is plausible that these cognitive operations
mediate the relation of family distress to ADHD, potentially differently in elementary-
aged children versus adolescents.
Conceptual C omplexities

Genes and environments are keenly interrelated, and these relationships can be
used to explore pathways to disordered behavior. Several important complications have
to be considered, however. First, twin and adoption studies show that genes have an
effect on measures of family environments (Plomin, 1994). Plomin and Rutter (1998)
presented a conceptual model to illustrate that the relationship between genetic risk and
ADHD could be mediated by psychosocial risk — an effect that would be independent of
the direct relationship between genetic risk and ADHD. In this model, genetic risk was
represented using a gene that has been associated with behavior. However, the correlation
between genetic expression and variations in psychosocial environments that they
suggested has now been described empirically (Goldsmith, Gottesman, & Lemery, 1997;

Rutter, Mofﬁtt, & Caspi, 2006). This supports the model underlying the current proposal.

However, it also introduces an important confound: environmental measures may be
genetically inﬂuenced. Thus, measures of family risk factors may also carry genetic
effects.

A second complication is that ADHD likely inﬂuences the family via effects on
important subsystems (e. g. parent-child and marital relationships) (Mash & Johnston,
2005; Parke, 2004). For example, Gerdes and Hoza (2006) found that mothers of children
with ADHD were more likely than comparison mothers to attribute their child’s
maladaptive behavior to unintentional and uncontrollable factors, and reported higher
levels of negative affect and assertive parenting in response to the behavior. This suggests
that parent-child relationships are strained when the child has ADHD. In another study,
parents of children with ADHD were more distressed than parents of controls, and coping
strategies were correlated with increased satisfaction (Podolski & Nigg, 2001). In other
studies, disturbances in family boundaries (ﬁ'om a family systems perspective) as well as
high levels of family adversity and discord have been associated with subsequent ADHD
in children (Davies, Cummings, & Winter, 2004; Jacobvitz, Hazen, Curran, & Hitchens,
2004). However, direction of effects remains unclear. Genetic effects were not modeled,
and parent ADHD, which may have contributed to these effects, does not appear to have
been controlled in these studies.

Summary

In summary, it is important to note that family risk factors likely carry genetic

effects, and that ADI-II) may have a reverse effect on family adversity. Nevertheless, the

present study addressed the important need for ﬁrrther clariﬁcation of the effects of

10

family risk factors on ADHD, and the extent to which cognitive mechanisms may explain
this relationship.

Speciﬁc Risk Factors at Issue
Specrﬁc Family Risk Factors Related to A DHD

Until this point family risk has been referred to generically, but it must be
speciﬁed. When it comes to speciﬁc family correlates, a number of family risk factors
have been associated with ADI-1]) including adversity and conﬂict, in both community
and clinic-referred samples of preschool and elementary school children (reviewed by
Johnston & Mash, 2001). In addition, race has been identiﬁed as a possible moderator of
family risk and ADHD. The major ﬁndings are brieﬂy reviewed here.

Adversity in the Family Environment. It has long been established that adverse
experiences in childhood are associated with maladaptive child behavioral outcomes
(Rutter, Cox, Tupling, Berger, & Yule, 1975). However, applications to ADHD were
more recent. Biederman et al. ( 1995b) examined family adversity as it relates to ADHD
using parental psychopathology and exposure to family conﬂict (a variable that included
six subscales of the Family Environment Scale (FES; Moos & Moos, 1974), and a
severity rating based on FES scores) as indices of adversity in a sample of boys ages 6-
17. Other measures included SES, birth order, family size, and intactness of the
biological family. This method of deriving a family adversity variable began with Rutter
et al. (1975) and is the most well recognized procedure for aggregating family risk in the
ﬁeld, having been used in various forms by subsequent investigators. Results of the
Biederman et al. (1995b) study showed more adversity in the ADHD than control group.

They also found that family conﬂict was a stronger predictor of ADHD and poorer

ll

ﬁrnctioning than parental psychopathology. In a different sample, but using a similar
adversity construct, Counts, Nigg, Stawicki, Rappley, and von Eye (2005) found that
more adversity risk factors were associated with one ADHD subtype (ADHD-C) than
another (ADHD-I) or controls. Again, perceived interparental conﬂict was the best
predictor of increased inattention and hyperactivity after controlling for other risk factors.

Conversely, among a clinical sample of adolescents, Rey, Walter, Plapp, and
Denshire (2000) found that poorer family environments (i.e. less emotional care, insecure
attachment relationships, and punitive, inconsistent, and inappropriate limit setting) were
related to ODD, CD, and increased functional impairment, but not associated with
ADHD. Of most interest here is whether family adversity contributes to ADHD
independent of these common comorbid conditions. Rey and colleagues concluded that
the quality of the family environment was not related to a diagnosis of ADHD, but with
CD. ADHD precedes CD along the developmental continuum (but may also interact with
adversity to cause CD), thus their study of adolescents could indicate a differential effect
of family environment at different stages of development. The present study examined
whether family risk predicted ADHD beyond the effects of ODD and CD, and tested
whether the inﬂuence of family environment on ADHD varied by age cohort (i.e.,
adolescents may have protective factors that younger children do not). That is, age would
act as a moderator for the effects of family risk on ADHD.

Family Conﬂict. Pressman et al. (2006) found more impairment in children from
families high in conﬂict and low in achievement and organization. Their study included
220 families consisting of two biological parents and two biological children between the

ages of 5-18 who had been diagnosed with ADHD (ADHD-affected sibling pairs).

12

Family conﬂict accounted for about 40% of similarity in impairment across siblings, and
this shared variance was not directly related to parent psychiatric diagnosis (Pressman et
al., 2006). Impairment was determined using three indicators: a clinician’s rating of the
Children’s Global Assessment Scale (C—GAS) of the K-SADS-PL, and t-scores from the
internalizing and externalizing scales of the Achenbach CBCL. Family environment was
measured using 6 subscales of the FES scale completed by both parents (cohesion,
expressiveness, conﬂict, organization, control, and achievement/orientation).
Additionally, family conﬂict was the mechanism that accounted for the relationship
between parent psychiatric diagnosis and the sibling impairment score.

Pressman et al. (2006) also found apparent birth order or developmental effects
such that the oldest sibling appeared to be vulnerable to a wider set of family
environmental risk factors than younger siblings. Though family conﬂict predicted
impairment in both siblings, low levels of family cohesion were associated with
impairment only in the oldest child. This could be a birth order effect or an age effect;
similar levels of impairment may yield different consequences in children across
development.

Another form of family conﬂict, interparental conﬂict, is also characteristic of
families of children with ADHD. Parents of children with ADHD have reported more
marital conﬂict and less marital satisfaction than parents of children without ADHD
(Murphy & Barkley, 1996). Harvey (2000) found that disagreement about childrearing
among parents is related to more problematic child behavior (perhaps through a feedback
loop as described by Patterson’s (1982) model), and more marital conﬂict. Counts et al.

(2005) found that child perception of interparental conﬂict carried most of the variance in

13

a family adversity model predicting inattention and hyperactivity that included SES,
parental psychopathology, and stressful events.

Family conﬂict is likely associated with subsequent decrements in parenting skill
and effectiveness that may play a role in the maintenance of ADHD. Maternal distress
mediated the relationship between family conﬂict and child behavior problems (Kendall,
Leo, Perrin, & Hatton, 2005). Satake, Yamashita, & Yoshida (2004) found that parents of
children with ADHD and comorbid ODD/CD were more likely than parents of controls
to have severe childhood ADHD symptoms and poorer adult mental health. Harel &
Brown (2003) found that children referred for ADHD evaluations and those who were
medicated for ADHD were more likely than controls to have parents that did not
complete college, and have one stepparent. These ﬁndings suggest that parental distress is
related to ADHD in offspring. Overall, it remains unclear how the interplay between
these aspects of the family environment contribute to ADHD in children; effects on
cognitive regulation are one potential pathway.

Several family risk variables appear to be related to ADHD, but interparental or
family conﬂict may be among the most important. The next point to consider is whether
these effects are driven by parent dysfunction.

Parent Risk Factors

Parent ADHD. ADHD is familial (F araone & Biederman, 1994a, 1994b). This
suggests that parents who themselves have ADHD may be implicated in the maintenance
and/or exacerbation of ADHD symptomology in their children. That is, parent history of
ADHD can be used as a proxy for genetic risk in their offspring (Crosbie & Schachar,

2001; Milberger, Biederman, Faraone, Guite, & Tsuang, 1997; Tsuang, Faraone, &

l4

Lyons, 1993); although transmission may also occur via psychosocial mechanisms.
Maternal ADHD limits signs of improvement shown by their preschool children
following a parent training intervention (Sonuga Barke, Daley, & Thompson, 2002).
Patterson et al. (2006) found that mother’s ADHD status was more strongly related to
child impairment in younger children, which suggests a developmental need for reliance
on mothers for organization and supervision. Further, children with a parent with ADHD
had a higher prevalence of mood, anxiety, oppositional, and conduct disorders (Minde et
al,2003)

Biederman, F araone, & Monuteaux (2002) combined two case-control studies of
nuclear families. They found that exposure to parental ADHD was linked to a disruptive
family environment (i.e. increased levels of family conﬂict and decreased family
cohesion), but did not independently increase ADHD risk in children beyond the genetic
effects. Thus, this ﬁnding showed that exposure to parental ADHD did not increase the
risk for ADHD in children beyond the level of genetic risk involved in parental ADHD
(determined using the SCID for current symptoms and the KSADS-E for past symptoms).
Nonetheless, parental ADHD was related to a disruptive family environment
(operationalized using three FES dimensions: cohesion, expressiveness, and conﬂict).
Overall, it remains unclear whether ADHD in parents confers genetic as well as
psychosocial risk to their children. The mechanisms that underlie these processes remain
to be revealed.

Maternal Distress/Depression. Negative emotion in mothers during pregnancy
may be a risk factor for ADHD in children (Lee, Chang, & Lung, 2006). Also, mothers of

children with ADHD have been found to show higher rates of depression than mothers of

15

controls (Cunningham, Benness, & Siegel, 1988). West, Houghton, Douglas, Wall, &
Whiting (1999) showed that this may be speciﬁc to mothers with more than one child
with ADHD or to mothers of children with ADHD-C. This research suggests that the
more severe the behavior (i.e. multiple children with or a child with ADHD combined
type which includes both inattentive and hyperactive/impulsive symptoms), the more
likely the association with maternal distress and depression. Chi and Hinshaw (2002)
suggested that mothers’ perceptions of their parenting behaviors may be related to
emotional distress and depressive symptoms rather than their actual behaviors as parents.
Such distortions may compromise the parent—child relationship, which could introduce
risk into the family environment that could in turn affect adaptive ﬁrnctioning in children
as well as ADHD symptomology.

Kendall, Leo, Perrin, & Hatton (2005) modeled family relationships of children
with ADHD and found that maternal distress may mediate the relationship between
family conﬂict and child behavior problems. This study also showed that the age of the
child with ADHD was associated with increased family conﬂict, an effect that was not
driven by severity of child behavior problems. This ﬁnding contradicts age effects found
in other research described earlier. However, it was supported by ﬁndings of Lewis-
Abney (1993) who found that age was negatively correlated with family ﬁrnctioning in
children with ADHD. This suggests that the longstanding collective negative effects of
child behavior resulted in an increase in family conﬂict. These discrepancies in the
literature suggest that age effects remain to be clariﬁed.

Culture as a Moderator of Family Risk and ADHD

16

The relevant environment may extend beyond the family to include cultural
inﬂuences. These inﬂuences represent key developmental contexts outside the child, such
as ecological elements of culture, as well as identity characteristics of the child and
family, and should be considered in analyses of environmental risk and protection
(Bussing, Schoenberg, & Perwien, 1998; Milich, Balentine, & Lynam, 2001). As
indicated, childhood studies of ADHD often involve predominately Caucasian male
samples GBarkley, Fischer, Smallish, & Fletcher, 2004; Loney, Kramer, & Milich, 1981;
Mannuzza, Klein, Bessler, Malloy, & LaPadula, 1993); however, knowledge of ADHD
derived on these children may not capture the disorder in children of color. There are
examples of race and culture moderating family inﬂuences on child psychopathology
(Angold et al., 2002; Deater-Deckard, Dodge, Bates, & Pettit, 1996; Lynch, 2003),
however such effects are not well studied in ADHD, and less so with respect to cognitive
inﬂuences.

As noted, the literature on ethnic differences in ADHD, as well as on how ADHD
presents in different cultural contexts, is in its early development. Findings from the
small literature on symptom severity suggest that ADHD in African Americans is
associated with more symptoms (Arnold et al., 2003) and more comorbid pathology
(Samuel et al., 1998). However, more investigation is needed to determine whether this
suggests differential presentation of ADHD across race, or the presence of more complex
moderating variables. Additionally, Bussing, Schoenberg, and Perwien (1998) reported
that African American parents were less knowledgeable and had limited access to
information about ADHD compared to Caucasian parents. In another study, conducted in

Venezuela, Montiel-Nava, Montiel-Barbero, and Pefia (2005) found signiﬁcant

17

differences in cohesion, recreational activity orientation, and intellectual activity
orientation between ADHD and controls. They also found increased ADHD symptom
severity associated with greater levels of family conﬂict and lower levels of cohesion as
measured by the FES. These studies suggest that an increased understanding of cultural
effects on ADHD will help uncover potential differences in behavioral tolerance and
interpretation, hence culture as a potential moderator of family risk.

Race, ethnicity, and culture are terms that are often used interchangeably in the
literature as grouping variables based on assumed shared psychosocial characteristics that
may inﬂuence behavior of individuals within these groups (Betancourt & Lopez, 1993;
Okazaki & Sue, 1995). One of the most frequent critiques of race research in mainstream
literature is that race is often measured as a demographic variable in and of itself, but that
more measures should be used to determine the extent of the cultural contribution (Tyson,
2004). The inherent risk in focusing on such grouping variables is that diversity among
individuals within the groups is often unrecognized (Tyson, 2004). Unfortunately, much
of the literature on race and ADHD relies solely on demographic variables to measure
racial/ethnic inﬂuences. As a possible solution to this problem, the present study
incorporated measures of ethnic identity (Phinney, 1992). The Multigroup Ethnic Identity
Measure (MEIM) measures one’s sense of belonging, commitment, and active investment
in their ethnic group. This construct may help assess a process responsible for cultural
variation. Ethnic identity was covaried in race analyses, rather than utilizing the race and
ethnic identity constructs as unique contributors, to more ﬁttingly represent culture.

Overall, more clarity is needed to determine the inﬂuence of cultural constructs

such as ethnic identity, which may moderate family environmental risk effects on ADHD.

l8

An understanding of which underlying mechanisms account for observed cultural
differences in ADHD may inform more appropriate diagnostic and risk and prevention
practices that may better serve children from diverse backgrounds.

Global Adaptive Functioning

Impairment is used to describe deﬁcits in ﬁmctioning that include an inability to
carry out appropriate tasks or interactions by oneself or with friends and family in home,
school, and work domains. DSM-IV criterion D requires that children evidence
impairment in functioning in at least two domains (i.e. home and at school or other
places) in order to meet eligibility for an ADHD diagnosis (APA, 2000). Diagnostically,
this construct is very important as it conveys whether the condition limits one’s ability to
function appropriately. However, its operational deﬁnition has been elusive, as it
necessarily involves some degree of social value judgment (i.e. what is impairing to one
individual may not be considered impairing to another), and some amount of overlap with
ADHD symptoms themselves. Gordon et al. (2006) reported that the size of the ADHD
group when classiﬁed by symptom count decreased by 77% when measures of
impairment (based on criterion D) were required to be included in deﬁning the group
versus when they were not.

ADHD has been associated with high levels of impairment related to conduct,
emotion, and peer interactions (Gaub & Carlson, 1997 ; Gershon, 2002; Strine et al.,
2006), and difficulties related to impulsivity, hyperactivity, and inattention often result in
difﬁculty at school that may result in low grades and disciplinary measures for affected
children. There is also evidence that impairment can continue across domains for years

after the initial problems arise (Hinshaw, Owens, Sami, & Fargeon, 2006).

19

For present purposes, the key issue is that it remains unclear which mechanisms
are involved in the relationships between family risk (speciﬁcally developmental risk),
ADHD, and adaptive impairment. Put another way, it remains to be determined whether
adversity is a non-speciﬁc index of impairment in children, or whether it has a speciﬁc
effect on ADHD symptoms. Ultimately, if family risk variables such as those described
herein are linked to inattention, hyperactivity, and impulsivity, and these conditions affect
adaptive functioning, then ADHD (a highly heritable condition) may mediate the
relationship between family risk and functional impairment. On the other hand, family
effects related to ADHD could be explained by overall deﬁcits in family functioning, or
by comorbid conditions.

Possible C onfounds

Health Risks. Several risk factors associated with ADHD in children, in particular
low birth weight, maternal alcohol use, and maternal tobacco use during pregnancy, are
also associated with high levels of family adversity. If they are not considered, spurious
conclusions may result. Key confounds are therefore considered in this section.

With regard to low birth weight, in a longitudinal study by McGrath et al. (2005),
188 infants (39 full-term and 149 pre-term) were tested for medical and neurological
status at age 18 months and 30 months. Behavior at age 4 was rated by parents and
trained evaluators on inattention and activity level. A number of signiﬁcant associations
were found including low birth weight and gestational age, which predicted hyperactivity
and inattention at age 4. Low birth weight was associated with increased risk for ADHD
(among other conditions) in a study of a sizeable sample of children ages 0-12 years

(N=7817) (Stein, Siegel, & Bauman, 2006).

20

With regard to alcohol use and smoking during pregnancy, in a study that
examined children of monozygotic and dizygotic female twins at risk for alcohol use
disorder, Knopik et al. (2006) found evidence for a genetic relationship of increased risk
for ADHD. Children of twins with a history of alcohol use disorder and children of
mothers with no history of alcohol use disorder but‘with a co-twin with an alcohol use
disorder were more likely to show ADHD than children of controls. This study also
found signiﬁcant effects for prenatal smoking; the alcohol use disorder results remained
signiﬁcant when statistical adjustment was made for prenatal smoking. Thus, it appeared
that genetic risk for alcohol use disorder may also inﬂuence risk for ADHD.

Prenatal smoking appears to be a well-studied risk factor for ADI-II) (Button,
Thapar, & McGuffm, 2005; Rodriguez & Bohlin, 2005), and its effects appear to be
related to inﬂuences during early brain development (Ernst, Moolchan, & Robinson,
2001). In a study of response to preventive intervention, children exposed to prenatal
cigarette smoking by mothers displayed higher levels of ADHI) symptoms and did not
respond to the intervention - they were even more likely to experiment with smoking
themselves at an early age (Vuijk, van Lier, Huizink, Verhulst, & Crijnen, 2006). Further,
inattentive symptoms were higher in children whose mothers smoked over 10 cigarettes
per day versus controls (after adjusting for confounds including maternal ADI-ID, birth
weight, and prenatal alcohol use) (Schmitz et al., 2006). Linnet et al. (2003) reviewed the
literature on the risk of prenatal exposure to psychosocial stress, alcohol, nicotine, and
caffeine on ADHD behavior to that time. They concluded that prenatal cigarette exposure

conferred the clearest risk, whereas results of studies of other risk factors were

21

contradictory and inconclusive. However, studies on psychological stress during
pregnancy tended to show a small relationship to ADHD in the child.

Overall, prenatal and perinatal factors appear to be associated with increased risk
for ADHD. Yet these are likely to co—occur with family adversity, taking place in lower
income, less organized families with more parent psychopathology. Therefore, it is
appropriate to attempt to account for these effects when investigating pathways between
family risk and ADHD.

Hypotheses

The study examined the extent to which family risk factors inﬂuence ADHD via
disruption of neurocognitive functioning. Preliminary hypotheses posited that 1a)
selected family measures would statistically predict ADHD symptoms independent of
parent ADHD, one another, ODD, and CD; 1b) the same family measures would predict
inattentive and hyperactive symptoms. Next, it was hypothesized that 2a) the family
measures could be consolidated to form a latent family risk construct; 2b) ADHD would
be signiﬁcantly related to the latent family risk construct; 2c) family risk would predict
ADHD when items were used to create latent factors for ADHD symptom domains. With
the preliminary analyses completed, the focus shifted to the primary hypotheses: 3a)
cognitive regulation would partially mediate the relationship between family risk and
ADHD; and child age (3b) and race (3c) would moderate the partial mediation described
in 3a. Finally, it was hypothesized that 4) ADHD would partially mediate the relationship

between family risk and adaptive functioning.

22

METHOD

Participants

The sample included 394 girls and boys between the ages of 6-1 8 years old — 53%
of whom met DSM-IV criteria for ADHD1 and 47% of whom were typically developing
comparison children. Data were collected using the Michigan State University Attention
Study protocol for children and adolescents. Participants were recruited widely from the
local community via advertisements placed in the community newspaper, on local radio
broadcasts, and during previews at a local movie theater, and through letters and
pamphlets mailed to parents of children in several of the local school districts. Volunteers
were evaluated for study eligibility through a standard multistage screening process, as
follows.
Sample and Recruitment

Procedure overview. First, prospective participants contacted the project office
for an explanation of the study, and a brief phone screen to check for rule-outs (age 6-18,
no sensory-motor handicap, no neurological illness, medication rule outs, and native
English-speaking). Second, eligible participants were scheduled for a 3 hour diagnostic
visit wherein the visiting parent gave written informed consent (in compliance with the
MSU Internal Review Board) and completed semi-structured clinical interviews and
normative rating scales to assess ADHD and other Axis I disorders in the child. During

this visit, the child gave a written assent to participate in the study and completed

 

‘ A diagnostic team of certiﬁed mental health professionals reviewed information from the semi-structured
interview and parent, teacher, and youth-report rating scales to arrive at a categorical “best estimate”
diagnosis of ADHD present or absent (Faraone, 2000). Inter-clinician agreement was adequate (from k =
.864 to .967). This categorical diagnosis was used in Table la to describe the sample; however continuous
Conners’ Teacher Rating Scale scores were used to address study hypotheses.

23

academic and IQ screening in addition to normative rating scales and a brief clinical
interview to assess ADHD and other Axis I disorders.

After the screening visit, some families were screened out and paid for their time
whereas other families were asked to return to the laboratory to complete another 3 hour
visit during which the cognitive measures were obtained. During this visit the parent
completed rating scales regarding behavior, mood, personality, and relationships
regarding their child, family, and self. The child completed a series of cognitive tasks and
rating scales regarding their perceptions of their parents and families as well as their own
habits and preferences. Thus, by design cognitive and family measures were available on
all 394 families in the sample selected for study.

The second parent, when available, completed 30-45 minutes of questionnaires
and returned them to the laboratory by mail. The child’s current teachers (main teacher
for elementary aged children; teacher selected by family for adolescents) received a 30-
minute packet of questionnaires regarding the child’s behavior at school.

Diagnostic Measures Pertaining to the Child

Diagnostic Dimensions of ADHD in the child. To obtain normative, standardized
dimensional ratings of attention problems as well as other current symptoms, teachers
completed the Conners’ et al. Teacher Rating Scale — Revised: Short (CTRS; Conners,
Sitarenios, Parker, & Epstein, 1998). This 28-item questionnaire involved rating the
child’s behavior in the classroom and in school on a 4-point Likert scale (0 = not true at
all; 3 = very much true). Two of the subscales were of interest: hyperactivity (a = .94)
and cognitive problems (a = .85; used in this study as a proxy for inattention). High

scores on the hyperactivity scale indicated trouble sitting still or remaining at the same

24

task for very long, more restlessness and impulsivity than most individuals their age, and
a continuous need to be “on the go” (Conners et al., 1998). High scores on the cognitive
problems (or inattention) scale indicated slower learning than most individuals their age,
difficulty organizing schoolwork, trouble ﬁnishing tasks or schoolwork, and problems
concentrating on tasks that required constant mental effort. The subscales displayed
appropriate factorial validity (N=1,897, all loadings ranged between .436 and .657,
p<.05) with adequate ﬁt (GFI=.91, AGFI=.88, RMS=.062) and assessed relatively
distinct dimensions (Conners et al., 1998). Test-retest reliability coefﬁcients ranged from
.47-.86, (p<.05) indicating stability of the subscales (Conners et al., 1998).

Assessment of other Axis I disorders in the child The Kiddie Schedule for
Affective Disorders and SchiZOphrenia (K-SADS-E; Puig-Antich & Ryan, 1986) for
DSM-IV was administered by a trained master’s level clinician after extensive training in
the interview and checkout of taped interviews for validity reviewed by our staff social
worker. Diagnoses examined in the current study included ODD, CD, major depressive
disorder, dysthymic disorder, bipolar disorder, substance abuse and dependence,
psychotic disorders, obsessive-compulsive disorder, panic disorder, agoraphobia, simple
phobia, social phobia, elimination disorders, sleep disorders, and eating disorders.
Autistic disorder was screened by added symptom questions and was a rule out as noted.
The K-SADS has exhibited acceptable validity and reliability (Ambrosini, 2000). To
assess reliability of the interview procedures in our sample, 20 K-SADS interviews were
videotaped and coded by two qualiﬁed interviewers. Inter-interviewer reliability for Axis
I disorders was acceptable (k = .798).

Assessment of Parent Psychopathology

25

Parent ADHD. Parent ADHD was a key control variable — it could represent
genetic risk and/or could be a driver of family adversity. Therefore, its control was
considered crucial to understanding family ﬁndings. To assess parent ADHD, both
parents were asked to complete self-report ratings of ADHD symptoms in childhood
(ages 5-12, which precedes child ADI-II) in time sequence) and currently (past 6 months).
These included the Conners’ Adult ADHD Rating Scale: Long Version (1999), and
ADHD Rating Scale (Barkley & Murphy, 1998), which have acceptable reliability and
validity. The primary parent (usually the mother) was also asked to rate their child’s other
biological parent using the Barkley & Murphy ADHD Rating Scale. I combined this
information via a composite ADHD score algorithm that involved (1) converting the
scales into a uniform rubric (z-scores), and (2) taking the self score for each symptom
whenever both ratings were available. Higher symptom ratings indicated more problems
associated with attention, hyperactivity, and/or impulsivity. Adults with a history of
ADHD in childhood may under-endorse childhood symptoms (Barkley, Fischer,
Smallish, & Fletcher, 2002), thus parent’s current ADHD symptoms were considered in
analyses. Reliability was acceptable for the parent 1 (a = .85) and parent 2 (or = .76)
composite scores in this sample.

Parent Depression. Maternal depression was one of the variables used in the
Rutter index of family adversity. To assess it, the visiting parent (usually the mother)
completed the 13-item depression subscale of the Symptom Checklist-90-R (SCL-90-R;
Derogatis, 1994). The SCL-90-R is a 90-item measure of psychological distress during
the past week, which is rated on a 5-point Likert scale (0 = not at all, 1 = a little bit, 2 =

moderately, 3 = quite a bit, 4 = extremely). Thus, higher ratings indicated more

26

depressive symptomology. Some items include: feeling low in energy, crying easily,
feeling lonely, feeling blue, and feeling no interest in things. This scale was been found to
be statistically reliable in this sample (a = .94).

Child Cognitive Regulation

Cognitive measures were selected so as to capture two core functions thought to
be involved in ADHD: response inhibition and response variability. Full scale IQ was be
covaried.

Logan Stop Task. (Logan, 1994). The stop task is a computerized choice-reaction
time task that measures executive response inhibition. During the task, the participant
was instructed to press the “X” or the “0” key as quickly as they can when an “X” or “0”
appears on the computer screen. On 25% of trials, a warning tone sounds, indicating they
should interrupt their response. The timing of the tone was varied in a stochastic tracking
procedure to maintain a successﬁil inhibition rate of 50%. The amount of warning needed
(stop signal reaction time (SSRT)) was then computed by subtracting average stop signal
latency (warning time) from average go response time (Logan, 1994). Data cleaning
procedures were identical those described by Nigg (1999). The response time (RT)
variability score is the standard deviation of go reaction time, and an index of arousal
such that greater variability indicates deﬁcits in arousal (Parasuraman, Warm, & See,
1998). Larger values for SSRT and RT variability indicate deﬁcits in regulatory control.
Thus, a cognitive regulation composite variable was generated using the mean of the
SSRT and RT variability z-scores (r = .35, p < .001, a = .89).

Full scale IQ. Child participants completed a 3-subtest short form of the WISC-

IV (Wechsler, 2003) or 5-subtest short-form of the WISC-IH (Wechsler, 1991). The hill-

27

scale IQ (FSIQ) score was estimated based on established norms for the respective age
groups, with WISC-IV reliability = .97 and validity = .87 (Sattler & Dumont, 2004) and
WISC-III reliability = .96 and validity = .89 (Sattler, 2001). For 17 year-olds, a 5 or 3-
subtest short-form of the WAIS-III (Wechsler, 1997) was used, and F SIQ was estimated
using norms established for this age group, with reliability >93 and validity = .88
(Sattler, 2001).

Remaining Measures Related to Family Adversity

Demographics. The visiting parent was asked to provide information about each
parent’s education level, occupation, employment status, and yearly household income on
a family background questionnaire. Household income was used in analyses as a proxy
for SES in the family risk model. Parents also indicated their ethnicity and the ethnicity
of the participating child(ren). In the event that there was disagreement among reporters
of ethnicity, the primary parent’s report was used in analyses of ethnic moderation of
family effects described later.

Family Environment Scale (FES; Moos & Moos, 1974). The visiting parent
(usually but not always the mother) completed the FES. It is a 90-item questionnaire
comprised of true/false questions related to the participant’s general perceptions of their
family. Moos, Insel, & Humphrey (1974) used a sample of 8 14 family members in 240
families and found three reliable and valid dimensions that encompass 10 subscales:
Relationship (Cohesion, Expressiveness, Conﬂict), Personal Grth (Independence,
Achievement Orientation, Intellectual-Cultural Orientation, Active-Recreational

Orientation, and Moral-Religious Emphasis), and System Maintenance (Organization and

28

Control). Internal consistencies ranged between r = .64 to r = .78, which is acceptable.
Test-retest reliability was also acceptable and ranged between .68 and .86.

FES Dat_a Reduction. The F ES and several additional family factors should create

 

a latent family risk variable for use in study analyses. In order to reduce the large number
of FES subscales into a smaller and more manageable number of factors, an Exploratory
Factor Analysis (EF A) was conducted. Analyses were conducted in Mplus using Full
Information Maximum Likelihood (F IML). SPSS produced similar results (with raw
data) to those generated using FIML. This supports the utility of the F IML method in this
dataset, as the imputation process did not signiﬁcantly alter the results. Mplus was used
to test the ﬁt of the 10 FES scales to 1, 2, 3, and 4 factor models (Appendix B). The ﬁt
indices suggested an unacceptable ﬁt for 1 and 2 factor models. A 3-factor model would
not converge. A 4-factor model suggested good ﬁt, however 2 of the 4 factors contained
only 1 subscale. When the 4—factor model was examined in a Conﬁrmatory Factor
Analysis (CF A), it was under-identiﬁed and standard errors could not be computed.
Together, these ﬁndings suggest that a 2-factor model would be the most promising.

An examination of the factor loadings in the 2-factor EF A analysis revealed that
several subscales loaded similarly on both factors. The subscale with the smallest
difference between factor scores (Achievement) was removed from the model and it was
re-run to determine whether the ﬁt improved. An improvement was observed, but not
enough to support an acceptable ﬁt. The Conﬂict, Cohesion, and Active-Recreational
subscales were systematically removed from the 2-factor model (in the manner described
above), until the ﬁt indices suggested that the model ﬁt the data well (Muthén & Muthén,

1998-2007). The promax loadings were interpreted because the factors were correlated

29

(Muthén & Muthén, 1998-2007). Appendix B also shows a CF A of the 6-subscale 2-
factor solution with overall good ﬁt (x2 (5, N = 374) = 10.84, p > .05, CF I = .99, RMSEA
= .06, RMR = .03.). The correlation between the two factors was also signiﬁcant (r = .55,
p < .05), as were the variance components (p < .001 ), suggesting that each factor
accounts for an amount of variance that is signiﬁcantly different from zero. It is also
notable that a l-factor model of the 6 subscales did not result in good ﬁt, and a 3-factor
model had worse ﬁt and accounted for less variance when compared to the 2-factor
model.

The two theoretically meaningful and statistically reliable factors were labeled
Values and Expressiveness. The Values factor was comprised by the Control,
Organizational, Independence, and Moral-Religious subscales, which indicate the extent
to which the family has set rules and procedures, clear organization and structure,
independence, and ethical and religious values. The Expressiveness factor includes the
Intellectual-Cultural and Expressiveness subscales, indicating the extent to which the
family is interested in political, social, cultural, and intellectual activities, and encourages
open actions and expression of feelings.

C hildren's Perception of Interparental Conﬂict Scale (CPIC; Grych, Seid, &
F incham, 1992). The CPIC is a reliable and valid 48-item questionnaire completed by the
target child assessing their appraisals of and reactions to parental conﬂict. The items were
rated T = true, ST = sort of true, or F = false. Grych et al. initially proposed that there be
9 subscales that assess frequency, intensity, resolution, content, perceived threat, coping
efﬁcacy, self—blame, triangulation, and stability related to parental conﬂict. However,

Nigg et al. (under review) conducted the ﬁrst empirical study of the 48 items and

30

completed a cross-replicated EFA-CFA design superior to analyses used in the Grych et
al. study. The Nigg et al. study included the present study’s sample, and therefore
provided an appropriate solution for use with the present sample. The resulting 4 factors,
shown in Appendix C, were used in analyses herein: Conﬂict Quality (5 items, including
“My parents hardly ever argue”, “When my parents have an argument they yell a lot”),
Marital Unhappiness (7 items, including “My parents have arguments because they are
not happy together”, “My parents have pushed or shoved each other during an
argument”), Self Blame (9 items, including “My parents’ arguments are usually about
something I did”, “My parents blame me when they have arguments”), and Perceived
Threat (5 items, including “1 get scared when my parents argue”, “When my parents
argue I worry that they might get divorced”). The scale was reliable in the current dataset
(a = .89).

Multigroup Ethnic Identity Measure (MEIM; Phinney, 1992). This 12-item scale
measures ethnic identity search (a developmental and cognitive component) as well as
afﬁrrnation, belonging, and commitment (an affective component). Examples of ethnic
identity search items include: “I think a lot about how my life will be affected by my
ethnic group membership”, “I am active in organizations or social groups that include
mostly members of my own ethnic group.” Examples of affirmation, belonging, and
commitment items include: “I understand pretty well what my ethnic group membership
means to me”, “I have a lot of pride in my ethnic group.” Items were rated on a 4-point
Likert scale (1 = strongly disagree, 2 = disagree, 3 = agree, 4 = strongly agree). The

measure was reliable in the current dataset (a = .88). A total score was used in analyses.

31

Developmental History. To assess perinatal and health risk factors that could
account for apparent family adversity effects, the visiting parent (usually the mother)
completed a developmental history of the child across three domains: (1) pregnancy
history, (2) birth, and (3) growth and development. The pregnancy history section
included 9 items (e. g. mother use of tobacco, alcohol, mother illness or pregnancy
complications, medications during pregnancy, severe emotional distress during
pregnancy, and exposure to x-ray shortly before or during pregnancy). The birth measure
was comprised of 6 items (e. g. timing of the birth, child’s birth weight, type and
normalcy of delivery, baby medical difficulty at birth, and whether breast fed). Finally,
the growth and development section included 12 questions that referred to signiﬁcant
developmental milestones (e. g. difﬁculty during baby’s ﬁrst month at home, age at ﬁrst
step, ﬁrst single word, bed wetting, etc.) and health information (e. g. medications,
psychotherapy history, serious injury, allergies, medical illness). Information for the three
scales was aggregated by assigning a point for each endorsed risk incident, resulting in a
count measure of developmental wellbeing in each of the three domains (prenatal risk,
perinatal risk, and early delay). The yes/no coding revealed very low endorsement rates,
which limited the ability of the items to correlate, and reliability of the three individual
scales was not adequate (a < .51). Therefore, the mean of the standardized (z) scores was
taken in order to create a reliable and meaningful composite total risk score. To ensure
that the composite was sound, the inter-item correlations and scale internal reliability
were examined. The 27 items were entered into SPSS to test reliability, and items were
systematically removed based on highest alpha if item deleted score until the scale

demonstrated acceptable reliability. Six items were retained (a = .700): low birth weight

32

(below 2500g), tobacco use during pregnancy, use of alcohol during ﬁrst trimester,
second trimester, and third trimester, and number of drinks consumed per week during
pregnancy.

Adaptive Functioning. Adaptive ﬁinctioning was assessed by two measures:
clinician rated Global Assessment of Functioning (GAF; APA 2000), and parent ratings
of Total Competency on the Child Behavior Checklist (CBCL; Achenbach, 1991). The
interviewing clinician assigned GAF scores at the end of the structured clinical
interviews. This score of overall functional adjustment was an index of impairment from
0 to 100 with high scores indicating better functioning and low scores indicating more
impairment. Inter-rater reliability for GAF scores (ICC r = .714) was determined using
the same coding procedure used for K-SADS interviews (20 interviews were videotaped
and coded by two qualiﬁed interviewers). The CBCL Competence Scale is a parent rating
that measures the child’s participation and efficacy in activities in social, school, and
home domains. It is scored using T-scores ranging from 0 to 100 with high scores
indicating better functioning and low scores indicating clinical impairment. To create a
single score, the GAP and CBCL scores (r = .478, p < .001) were standardized and
averaged into a single adaptive functioning composite (u = .70).

Data Analysis

Missing data that occurred at random due to changes in the protocol,
administrative or computer error, child reﬁisal or invalid response on some measures
were imputed using ﬁill information maximum likelihood (F IML) in Mplus using the

Type = complex command (Muthén & Muthén, 1998-2007). This imputation procedure

33

follows the guidelines in McCartney, Burchinal, and Bub (2006), and is considered
superior to listwise deletion.

Data collected for this study are nested to account for non-independence of
observations, as they include data from siblings within the same family. Mplus was used
to cluster the data in order to correct for the nestedness in all analyses (unless otherwise
indicated).

A priori decision criteria for factor analyses, and structural equation models are
outlined in this section. Hu and Bentler (1999) suggest that Root Mean Square Error of
Approximation (RMSEA) values below .06 and Root Mean Square Residual (RMR)
values below .08 indicate satisfactory model ﬁt, with lower values indicating better
model ﬁt. Chi-square values are indicative of best model ﬁt when they are closest to zero
with no degrees of freedom, and the model ﬁt worsens as the chi-square value increases
(Kline, 2005). Further, a normed chi-square value can be obtained to reduce its sensitivity
to sample size, such that ledf < 2 can indicate acceptable ﬁt (Kline, 2005). Mplus
Version 4 outputs provide unstandardized coefficients (Est), standard errors (S.E.),
estimates divided by their standard errors (Est/SE), and standardized coefficients
(StdYX) based on latent and observed variables’ variances (representing the amount of
change in Y per standard deviation unit X). The Est/SE. value is a Z statistic, and tests
the null hypothesis that the population parameter estimate is zero. The Z values are
signiﬁcant at the p < .05 level when they exceed d: 1.96. The StdYX values are the
regression weights shown in the ﬁgures herein. The regressions captured between as well

as within group variation.

34

RESULTS

Sample Description

Demographic information for the sample is shown in Tables la and lb. ADHD
diagnoses were established using the Best Estimate Diagnosis, and these assignments are
further validated by ADHD symptom rating scales that show signiﬁcant elevations in the
ADHD sample. Table 1a shows a greater proportion of male participants with ADHD;
this is common in ADHD studies and may reﬂect the male preponderance of ADHD in
the population (APA, 2000; Gaub & Carlson, 1997). Each hypothesis that involved
ADHD was checked for confounding effects of gender, and none were found. In addition,
Table la shows a greater proportion of children under age 13 in the ADI-II) group. Age
was covaried in analyses that involved ADHD, and results were not signiﬁcantly
affected. The percentage of minority participants was similar across groups. The ﬁill
sample was 74% Caucasian, 12% African American, 5% Latino, 2% other, and 7%
multiracial, representative of the ethnic makeup of the local county (US Census Bureau,
2000). Consistent with other studies of ADHD, participants with ADHD were more likely
to have symptoms of ODD, CD, greater global impairment (GAF), and ADHD and
depressive symptoms in the primary parent than the non-ADHD comparison group.

Annual household income was signiﬁcantly lower in the ADHD group (M = 62k,
SD = 34k) than in the non-ADHD comparison group (M = 71k, SD = 35k; p < .05). This
is consistent with research that showed an increased prevalence of ADHD among
individuals with lower socioeconomic status (Barkley, 1998). Table 1b shows that the
average income for the ﬁll] sample was 66k (SD = 35k), which reveals socioeconomic

variability in the sample with a relatively normally distributed range from approximately

35

6k to 180k US dollars annually. Group comparisons revealed that more households in the
child (under age 13) and ethnic minority groups were below the poverty level compared
to households of target children in the adolescent and Caucasian groups. Percentages
closely mirrored the poverty rates in the local county (14.6%; US Census Bureau, 2000).
No signiﬁcant differences were found in regards to parent’s education level among the
child and adolescent groups; however, parents in the Caucasian group attained higher
levels of education than those in the ethnic minority group. Similarly, no signiﬁcant age
group differences were found related to the child’s living situation; however, children in
the Caucasian group were signiﬁcantly more likely to live with both parents than children
in the ethnic minority group.
Power Analyses

For multiple regression analyses a priori power analyses required a sample of at
least N=172 to reliably detect a medium effect size (f? = .15) with up to 10 predictors with
power (b) = .95 and alpha = .05 (Cohen, 1977, 1992; Faul & Erdfelder, 1992). All
analyses were at this power level or better, and with N = 394, there was ample power to
detect medium sized effects. In analyses with 2 predictors, there was sufficient power to
detect a small effect size (f = .02) at a power of b = .714 (a = .05) (Cohen, 1977, 1992;
Paul & Erdfelder, 1992). In the SEM analyses, there were at least 10 participants for
every 1 estimated variable, 5 participants for every parameter (Bentler, 1990; Kline,
2005)
Hypothesis 1 a. Selected family measures would statistically predict ADHD symptoms

independent of parent ADHD, one another, ODD, and CD.

36

To test this preliminary hypothesis, a hierarchical multiple regression was
conducted in SPSS2 with teacher-rated ADHD symptoms as the dependent variable3 .
Developmental risk factors (i.e. low birth weight, prenatal smoking, and prenatal alcohol
use) as recalled by the mother were entered into Step 1 of this model to account for
potential confounds. Next, parent 1 ADHD and parent 2 ADHD were entered into Step 2.
Finally, FES factors, CPIC factors, parent depression, income, ODD, and CD were
entered into Step 3. This analysis helps clarify whether family risk factors contribute to
ADHD independent of common comorbid conditions (Rey, Walter, Plapp, & Denshire,
2000). Results of this analysis are presented in Table 2. Step 3 brought a signiﬁcant gain
in r2 (A 12 = .175, p < .001), which justiﬁed pursuing the corresponding effects. The CPIC
Self Blame factor was uniquely related to ADHD apart from the other family measures
(including Developmental Risk factors), parent ADHD, ODD, and CD (B = .143, p =
.042)?

Hypothesis 1 b. The same family measures would predict inattentive and hyperactive
symptoms.

Two additional preliminary multiple regression analyses were conducted in SPSS
using the predictors and the procedures outlined in Hypothesis 1a. However, in this case
the dependent variables were (1) the number of ADHD inattentive symptoms and (2) the
number of ADHD hyperactive/impulsive symptoms deﬁned by teacher report of current

symptoms on the Conners’. In each analysis, there was a signiﬁcant gain in r2 (A r2 =

 

2 Results were similar when I ran these analyses in Mplus using FIML.

3 Ratings on the Conners’ Teacher Rating Scale were significantly correlated with parent report ratings on
the KSADS-E semi-structured interview (r = .65, p < .001), and on the Conners’ Parent Rating Scale (r =
.61, p < .001). Teacher ratings were used as the dependent variable to limit reporter bias, as several
independent variables were obtained via parent report.

4 As an additional check on the effect of covariates, ODD and CD were removed from the model and
results are shown in Appendix D. Another analysis was run with ODD and CD in Step 2, and ﬁndings were
similar to those reported in Hypothesis la. Results of this amlysis are also shown in Appendix D.

37

.196, p < .001 for inattentive symptoms, and A r2 = .187, p < .001 for
hyperactive/impulsive symptoms).

In the analysis of inattentive symptoms, the CPIC Conﬂict Quality factor (B =
.177, p = .03 7), and the CPIC Self Blame factor (B = .269, p < .001) were signiﬁcant
beyond the effects of the other family measures (including Developmental Risk factors),
apart from parent ADI-II), ODD, or CD. However, in the analysis of
hyperactive/impulsive symptoms, there was only a signiﬁcant effect of ODD symptoms
(B = .226, p = .004). This illustrates a difference in the predictability of inattentive and
hyperactive/impulsive symptoms of ADHD.

The CPIC Self Blame factor was a consistently signiﬁcant predictor of ADHD
symptoms. A supplementary regression analysis was conducted to determine whether
ODD, inattentive symptoms, or hyperactive/impulsive symptoms inﬂuence the CPIC Self
Blame factor. The regression model was signiﬁcant, F = 12.55, p < .001, R2 = .095.
However, the inattentive symptom variable was the only signiﬁcant predictor (B = .270, p
< .001). These results suggest that the signiﬁcant effect of CPIC Self Blame was not
carried by ODD or hyperactive/impulsive symptoms.

Hypothesis 2a. The family measures could be consolidated to form a latent family risk
construct.

A Conﬁrmatory Factor Analysis (CF A) was conducted to evaluate the ﬁt of our a
priori single factor model of “family risk” (comprised by FES factors, CPIC factors,
parent depression, and income). The initial model ﬁt was unacceptable (x2 (14, N = 394)
= 212.15, p < .001, Comparative Fit Index (CF I) = 0.29, Root Mean Square Error of

Approximation (RMSEA) = 0.19, and Standardized Root Mean Square Residual (SRMR)

38

= 0.10). FES Values and FES Expressiveness had poor loadings on the family risk factor
(FES Values StdYX = -.14, p > .05; FES Expressiveness StdYX = -.15, p > .05), and
were therefore eliminated from the model. The means and standard deviations of the FES
factors suggested that most families endorsed sub-clinical levels of these phenomena
(clinical cut-off = T = 65; mean T score = 47.92, mean SD = 10.84), reﬂecting similarities
in value-orientation and expressiveness across the sample.

Separate analyses were conducted to determine whether ADHD was related to
FES independent of ODD. First, FES Values and FES Expressiveness were regressed
onto CTRS ADHD symptoms and ODD. ADHD signiﬁcantly predicted FES
Expressiveness (F = 6.56, p < .01, R2 = .036), but not FES Values beyond the effects of
ODD (F = 2.46, p = .09, R2 = .014). Next, CTRS ADI-ID was regressed onto FES
Expressiveness and ODD. FES Expressiveness was signiﬁcantly related to ADHD (F =
24.16, p < .001, R2 = .121), even when ODD was in the model. Nevertheless, the
previous CFA model showed that neither FES factor contributed to the family risk
variable.

The CF A model was computed again without the FES factors, and resulted in a
signiﬁcant improvement in ﬁt, A f (5, N = 394) = 27.29, p < .01. Modiﬁcation indices
suggested that CPIC Perceived Threat be allowed to correlate with CPIC Self Blame
(modiﬁcation index = 18.69). The resulting model is shown in Figure 1. It resulted in
good overall ﬁt, )(2 (8, N= 394) = 4.97, p > .05, CFI = 1.00, RMSEA = 0.00, and SRMR
= 0.02. This family risk latent factor was used in all subsequent analyses.

Hypothesis 2b. ADHD would be signiﬁcantly related to the latent family risk construct.

39

The analyses from Hypothesis 1 were conducted again using the family risk latent
variable in an SEM analysis predicting ADHD total symptoms (measured henceforth
using the Conners’ Teacher Rating Scale) as well as teacher-reported inattentive
symptoms and teacher-reported hyperactive/impulsive symptoms (also measured using
the CTRS).

When covarying family risk with developmental risk, ODD, CD, and parent
ADHD, most of the ﬁt indices were indicative of unacceptable ﬁt: x2 (36, N = 394) =
126.27, p < .001, CFI = .80, RMSEA = .08, and SRMR = .06. In this model, family risk
was appropriately indicated by the CPIC, depression, and income factors as in Hypothesis
2a, and ADHD total symptoms were signiﬁcantly regressed onto family risk (B = .607, p
< .05). However, ODD also signiﬁcantly predicted ADHD total symptoms (B = .202, p <
.05), and ODD was also signiﬁcantly correlated to family risk (r = .279, p < .05). The
signiﬁcant effects of ODD are consistent with ﬁndings of frequent comorbidity with
ADHD.

An additional analysis revealed that the relation of ADHD to family risk was not
due to the relation of ODD to family risk. The model shown in Figure 2 provided a good
overall ﬁt: 1’ (16, N: 394) = 21.3 1, p > .05, CF] = .98, RMSEA = .03, and SRMR = .03.
When ODD was removed from the model, ADI-{D’s estimate became StdYX = .37, p <
.05. Analyses have shown that the covariates (developmental risk, parent ADHD, ODD,
and CD) do not account for the relationship between ADHD and family risk, therefore
remaining analyses focused solely on the role of family risk factors in predicting ADHD

outcomes (without covariates).

40

The model shown in Figure 3 shows the relationship between the latent family
risk variable and ADHD, and provided a good overall ﬁt: 12 (12, N = 394) = 20.47, p >
.05, CF I = .97, RMSEA = .04, and SRMR = .03. A single manifest dependent variable
was used once again to represent ADHD total symptoms (CTRS), and all path estimates
were signiﬁcants. The model for inattentive symptoms (CTRS) was similar, however all
ﬁt indices except the 12 were indicative of good ﬁt: 12(12, N = 394) = 24.04, p = .02, CFI
= .96, RMSEA = .05, and SRMR = .03. The model for hyperactive/impulsive symptoms
(CTRS) was indicative of overall good ﬁt: )(2 (12, N = 394) = 11.94, p > .05, CFI = 1.0,
RMSEA = .00, and SRMR = .03. These ﬁndings consistently show a signiﬁcant
relationship between family risk and ADHD, with Figure 3 representing the most
comprehensive best-ﬁtting model across both components of the ADHD construct
measured by teacher report on the CTRS.

Hypothesis 2c. Family risk would predict A DHD when items were used to create latent
factors for A DHD symptom domains.

An SEM analysis was conducted to determine whether the latent family risk
variable was related to ADHD as represented by two latent variables -— inattention and
hyperactivity/impulsivity - indicated by CTRS items. The purpose of this was to evaluate
speciﬁcity of effects to ADHD symptom domains, as well as to help conﬁrm that
ﬁndings were not an artifact of how ADHD was measured. The model shown in Figure 4
resulted in a good overall ﬁt: A? (123, N = 394) = 147.80, p > .05, CFI = .99, RMSEA =
.02, and SRMR = .04. Following the suggested modiﬁcation indices allowed items within
the same scale to correlate with one another. The correlated items were theoretically (as

well as empirically) related.

 

5 Similar ﬁt and signiﬁcant path loadings were obtained when age was covaried.

41

The preceding hypotheses involved preliminary analyses in order to set the stage
for the primary focus of investigation, which now follows.

Hypothesis 3a. Cognitive regulation would partially mediate the relationship between
family risk and ADHD.

Cognitive regulation was, as indicated earlier, a manifest variable comprised of
the average of standardized scores for Stop Task SSRT and response variability. F SIQ
was covaried to check for effects on results. The model, shown in Figure 5, provided a
good overall ﬁt: X’ (23, N = 394) = 27.45, p > .05, CFI = .98, RMSEA = .02, SRMR =
.02, and Bayesian Information Criterion6 (BIC) = 9079.18. The indirect effect for this
model was signiﬁcant (StdYX = .03, p < .05)7. When the mediation analysis shown in
Figure 5 was conducted using bootstrapping (rather than clusters), the ﬁndings were
similar (see Appendix E). Thus, cognitive regulation did partially mediate the
relationship between family risk and ADHD.

Hypothesis 3b. Child age would moderate the partial mediation described in 3a.

An Mplus multiple group analysis was used to test this relationship. This was
done by adding age (at assessment of outcome) as a grouping variable to the model in
Figure 5. The child group was comprised by children ages 6-12 years old (N = 166), and
the adolescent group included youth ages 13-18 years old (N =228). Although alternative
models were not formally tested, the resulting ﬁllly constrained models, shown in Figures
6 and 7, suggest that the factor loadings and the factor variance-covariance structures
were not equivalent for children and adolescents: )(2 (56, N = 394) = 121.36, p < .001, CFI

= .84, RMSEA = .08, SRMR = .07, and BIC = 8956.92. In the child group, the indirect

 

6 The model with the lowest BIC value is preferable to models with higher values (Kline. 2005).
7 The results for model ﬁt. signiﬁcant path loadings. and indirect effect were similar when age was
covaried.

42

effect was non-signiﬁcant (StdYX = .02, p > .05), whereas the indirect effect was
signiﬁcant (StdYX = .05, p < .05) in the adolescent group. The lack of an indirect effect
in the child group is due to the non-signiﬁcant path between cognitive regulation and
family risk. Qualitative differences were observed between the models in the child and
adolescent groups. For example, F SIQ was not a signiﬁcant predictor of ADHD in the
child group. Thus, family risk appears to have a signiﬁcant effect on cognitive regulation
in adolescents, though not in children. However, family risk was signiﬁcantly related to
ADHD in both groups.
Hypothesis 3c. Race would moderate the partial mediation described in 3a.

An Mplus multiple group analysis was again used to test this relationship. In this
case, race was used as a grouping variable (Caucasian N = 291, ethnic minority N = 103),
and results on the Multigroup Ethnic Identity Measure (MEIM) were covaried to prevent
sole reliance on the demographic race variable. Again, although alternative models were
not formally tested, the resulting fully constrained models, shown in Figures 8 and 9,
suggest that the factor loadings and the factor variance-covariance structures were not
equivalent for Caucasian and ethnic minority groups: )(2 (68, N = 394) = 123.33, p < .001,
CF I = .87, RMSEA = .06, SRMR = .07, BIC = 9947.08. The indirect effect was
signiﬁcant in the Caucasian group (StdYX = .04, p < .05), but was not signiﬁcant in the
ethnic minority group (StdYX = .01, p > .05). This is apparent in Figure 9, as family risk
did not predict cognitive regulation and cognitive regulation did not predict ADHD in the
ethnic minority group. FSIQ and MEIM were not signiﬁcantly related to ADHD in either
group. The effect of family risk on ADHD was not accounted for by cognitive regulation

in the ethnic minority group.

43

Given the similarity of the race effects to the age ﬁndings, an additional check
was run to ensure that results were not an artifact of the sample distribution of race and
age. There was a proportional distribution of Caucasian to ethnic minority participants in
the child and adolescent age groups: child Caucasian N = 115 (69%), child ethnic
minority N = 51 (31%), adolescent Caucasian N = 176 (77%), adolescent ethnic minority
N = 52 (23%).

Hypothesis 4. ADHD would partially mediate the relationship between family risk and
adaptive functioning.

ADHD should partially mediate the relationship between family risk and adaptive
functioning if family risk is speciﬁc to ADHD, and not merely representative of global
impairment. To test this, an Mplus mediation analysis was conducted (as in Hypothesis 3)
using a composite adaptive ﬁinctioning score that represented parent and clinician report.

The model, shown in Figure 10, resulted in good overall ﬁt: {(17, N = 394) =
29.72, p < .05, CFI = .97, RMSEA = .04, and SRMR = .04. The indirect effect was
signiﬁcant (StdYX = -. l 3, p < .05), resulting in a partial mediations. This showed that
the effect of family risk on participants’ adaptive functioning was signiﬁcantly partially

mediated by ADHD.

 

8 Results and signiﬁcant path loadings were similar when age was covaried.

44

DISCUSSION

Adversity in the family environment has been established as a risk factor for
maladaptive child behavioral outcomes (Rutter et al., 1975); however the mediating
mechanisms associated with this relationship were under-researched. The present
investigation had as its main aim to examine whether cognitive processes accounted for
the relationship between family risk and ADHD. Via preliminary analyses, the present
study replicated existing research showing that family risk factors predict ADHD and that
these measures were able to be ﬁt as valid indicators of a latent composite family risk.
construct using conﬁrmatory factor analysis. To the main hypothesis, cognitive regulation
did partially explain the relation between family risk and ADHD, although this
relationship depended upon age and (a secondary focus) race group. Finally, ADHD
partially explained the relationship between family risk and impairment. Each of these
ﬁndings is considered in sequence.
Preliminary Findings

First, it will be useful to discuss the preliminary ﬁndings that helped determine
the building blocks for use in the study’s main hypothesis regarding neurocognitive
ﬁinctioning. Consistent with Hypothesis 1a, family measures (Conﬂict Quality, Marital
Unhappiness, Self Blame, Perceived Threat, Maternal Depression, and Income) were
collectively associated with ADHD symptoms (measured using CTRS Total Symptoms)
even when covaried with one another, developmental risk factors, parent ADHD, ODD,
and CD. However, only Self Blame was uniquely related to ADHD above and beyond the
other family measures. In a related study that used a different sample, but similar family

risk measures as those used in the present study (Counts et al., 2005), summary score of

45

child perception of interparental conﬂict was the best predictor of ADHD after
controlling for other risk measures. Thus, it is important to note that the present study
replicated the Counts et al. ﬁndings.

One could speculate that the effect of Self Blame, in particular, may have been
prominent because it is the only risk factor that involves perceived culpability of the child
for the family adversity. Another explanation for fewer unique predictors is that the other
family risk measures were related to aspects of ADHD that overlap with ODD, which
was also signiﬁcant in the model. According to this supposition, Self Blame evidenced an
observably signiﬁcant effect on ADHD beyond the other covariates because it may not
overlap with ODD. A supplementary analysis conﬁrmed that Self Blame does not predict
ODD when ADHD was covaried.

In a more robust test, two analyses were conducted to determine the relationship
between the selected family risk measures (Conﬂict Quality, Marital Unhappiness, Self
Blame, Perceived Threat, Maternal Depression, and Income) and (1) ADHD inattentive
and (2) ADI-ID hyperactive symptom domains (measured using the CTRS). These
ﬁndings were consistent with Hypothesis 1b and suggest that the family risk measures
were collectively associated with ADHD beyond possible confounds, even when it was
measured in terms of its two major symptom domains.

The individual family risk measures had different effects on inattentive versus
hyperactive symptoms. In regard to inattentive symptoms, CPIC Self Blame and Conﬂict
Quality factors emerged as uniquely predictive above and beyond the other measures and
confounds. One interpretation of these results is that children experience inattentive

symptoms due to impairment associated with perceiving that interparental conﬂict is the

46

child’s own fault and the frequency and intensity of parents’ arguments. Further study is
needed to determine whether inattention mediates the relationship between Self Blame
and Conﬂict Quality, and impairment. This would show that impairment speciﬁc to
inattention is related to these risk factors. In regard to hyperactive/impulsive symptoms,
family risk measures were not unique predictors beyond confounds. It appears that ODD
contributed signiﬁcantly to the effect, which is consistent with research that showed that
hyperactivity/impulsivity is more readily associated with comorbid externalizing
problems (including ODD) than the ADHD inattentive subtype (Eiraldi et al., 1997;
Milich et al., 2001).

The present study showed that the effect of Self Blame on ADHD was not due to
its relation with hyperactivity or to co-occurring ODD, but to its relation with the
inattention domain in particular. These Self Blame ﬁndings may be consistent with
research that showed that parental disagreement about childrearing was related to more
marital conﬂict as well as more problematic child behavior (Harvey, 2000). In summary,
whereas several studies have identiﬁed family conﬂict as an important risk factor
associated with ADHD (Biederman et al., 1995b; Counts et al., 2005; Lange et al., 2005;
Murphy & Barkley, 1996; Pressman et al., 2006); the present study extended these
ﬁndings by adding information on speciﬁc interparental conﬂict factors as perceived by
the child, and their relationship to the ADI-1]) inattentive symptom domain.

After looking at the individual family risk measures, the present study validated a
latent family risk construct that included most hypothesized family risk measures
(Hypothesis 2a). Measures that were signiﬁcant indicators of family risk include four

variables related to child perception of interparental conﬂict (Conﬂict Quality, Marital

47

Unhappiness, Self Blame, and Perceived Threat), as well as maternal depression, and
income. These ﬁndings are consistent with previous research on family adversity (Counts
et al., 2005; Kendall et al., 2005; Minde et al., 2003; Pressman et al., 2006; Satake,
Yamashita, & Yoshida, 2004; West et al., 1999). Contrary to hypotheses, the FES factors
(Values and Expressiveness) did not load on the family risk latent factor; however this
may be due to ﬂoor effects in the data that were more evident on these factors than on
others. Supplementary ﬁndings conﬁrmed a relationship between FES and ADHD (when
other variables were excluded), however F ES did not lend signiﬁcantly to the family risk
construct as in previous studies (Biederman, Milberger, Faraone, Kiely, & et al., 1995b;
Pressman et al., 2006). This weak result for the parent-reported FES measure may be due
to the prominent effects of child perception of interparental conﬂict as a predictor of
ADHD in the present study. Child perceived interparental conﬂict was not included in
models used in the other studies.

As anticipated in Hypothesis 2b, the signiﬁcant relationship between the family
risk latent variable and ADHD symptoms remained independent of ODD effects on
ADHD. These ﬁndings are consistent with prior studies that included children and
adolescents, assigned diagnoses using structured clinical interviews, and derived family
adversity variables using the most well recognized procedures for aggregating family risk
in the ﬁeld (Biederman et al., 1995b; Counts et al., 2005). However, Rey et al. (2000)
found no relationship between ADHD and the quality of the family environment in
clinically referred adolescents. Several methodological differences between the Rey et al.
and present (and previous) studies may account for the dissimilar results. First, the Rey et

al. study was limited to adolescents. Second, clinicians in the Rey et al. study used

48

information from patient charts to assign ADHD diagnoses, rather than using structured
clinical interviews. Third, Rey et al. measured family factors using different constructs
the Global Family Environment Scale (GFES; Rey et al., 1997). The GFES consisted of
clinician ratings of quality of the family environment (i.e. punitive, inconsistent, and
inappropriate limit setting, insecure attachment relationships, and emotional care) based
on chart reviews reﬂecting the lowest family environment score that occurred for a 12
month duration, before the child was 12 years of age. These methodological differences
may account for differences in ﬁndings.

The present study showed that family risk predicted ADI-ID even when ADHD
was evaluated as a latent variable, and items were used to create latent factors to
represent inattentive and hyperactive/impulsive symptom domains. This was consistent
with predictions in Hypothesis 2c. The most important ﬁnding here was that family risk
was not differentially associated with either inattention or hyperactivity. This means that
the latent variable did not predict one symptom domain signiﬁcantly more strongly than
the other, which suggests that family risk is relevant across ADHD symptoms, and
permits the use of an ADHD total symptom variable in subsequent analyses. These
ﬁndings ﬁirther conﬁrm the utility of the family risk variable, as it has been shown to
predict ADHD using a range of operational measures.

Main Findings

Second, and coming to the main ﬁnding in this dissertation, consistent with
Hypothesis 3a, the relationship between family risk and ADI-II) was partially mediated by
cognitive regulation. That is, family risk factors were related to cognitive processes,

which were in turn associated with ADHD deﬁcits — all independent of IQ. One way to

49

interpret this is that it supports the possible role of family events in epigenetic processes
(i.e. effects on gene expression) (Rutter et al., 2006). The present study identiﬁed
cognitive mechanisms (response inhibition and vigilance) known to shape ADHD
behaviors (Barkley, 1997; Crosbie & Schachar, 2001; Nigg, 1999), and demonstrated
their connection to family risk. It may be speculated that, among other possibilities, the
model described by Cicchetti and Curtis (2006) wherein family effects on gene
expression can manifest behavioral change via effects on cognition — might map onto the
mediation described in Figure 5.

Another interpretation of these ﬁndings is based on established relationships
between family risk factors and cognitive and behavioral ﬁinctioning in children. The
cognitive model used by many psychotherapists theorizes that an individual’s emotions
and behaviors are inﬂuenced by their perception of events (Beck, 1964; Ellis, 1962). This
model appears to explain the relationships found in Figure 5, and is also consistent with
ﬁndings of other relevant studies. Grych and F incham (1993) found that children’s
appraisals of interparental conﬂict (i.e. their cognitive interpretations of the stressor) were
affected by its intensity, content, and cause, which thereby determined the extent of the
stressor’s effect on the child’s emotions and behaviors. Nigg, Nikolas, Kriderici, Park,
and Zucker (2007) found that strong response inhibition capability was a protective factor
against ADHD when moderate to high levels of family adversity were present. The
partial mediation found in the present study may therefore have important implications
for clinical practice in support of the cognitive model.

Third, a child’s developmental stage was hypothesized to moderate the mediation

of family risk on ADI-H) via cognitive regulation, due to the inﬂuence of experiences on

50

brain development and behavior (Cicchetti & Curtis, 2006; Hebb, 1949; O’Conner,
2003). Consistent with Hypothesis 3b, age groups showed a differential effect of risk on
cognitive regulation which suggests that risk, though it predicts ADHD in both groups,
affects cognitive regulation in adolescents and not in younger children. These ﬁndings
are consistent with Richmond and Stocker’s (2007) results which indicated that
children’s appraisals of marital conﬂict declined signiﬁcantly from childhood to
adolescence in terms of perceived threat; however, little change was observed over time
in appraisals of self-blame.

An alternative explanation for the age moderation is a possible group difference in
early caregiver interchanges that are thought to facilitate the development of self
regulation, response control, and arousal (Kochanska & Aksan, 2006; Panzer & Viljoen,
2005; Tucker et al., 2005). Yet another explanation is that inhibitory control and arousal
mechanisms that are related to family risk in the adolescent group have not fully
developed to evidence a signiﬁcant effect in the younger child group (Akshoomoff, 2002;
Booth et al., 2003; Levy, 1980; Rubia et al., 2006; Williams et al., 1999). IQ was
negatively associated with ADHD in the adolescent group (Hinshaw et al., 1987;
Peterson, et al., 2001), but this path may not have been evident in the child group due to
differences in cognitive development or good matching of groups. Further study is
needed to determine other mechanisms that account for the effect of family risk on
ADHD behavior in young children.

Fourth, as a secondary focus, race was also hypothesized to moderate the
mediated relationship between family risk and ADI-II) via cognitive regulation.

Consistent with Hypothesis 3c, differences in the mediation model were observed

51

between the Caucasian and ethnic minority groups. Family risk predicted ADHD in both
groups. However in ethnic minorities, cognitive regulation did not partially mediate the
relationship between risk and ADHD. This suggests that a different mechanism may
explain the relationship in ethnic minorities. Neuropsychological deﬁcits are known to be
associated with ADHD (Nigg, Blaskey, Huang-Pollock, & Rappley, 2002), and
controlling for acculturation has been shown to account for ethnic differences in
neuropsychological test performance (Manly et al., 1998). It can be speculated that the
relationship between family risk and ADHD may change over levels of race due to
differences in neuropsychological functioning. In addition, ethnic identity did not appear
to contribute signiﬁcantly to the model in either group, though it was signiﬁcantly
correlated with family risk in the ethnic minority group. That is, increased ethnic identity
was a protective factor for ADHD. These ﬁndings are consistent with group differences
in race salience (Sellers, Smith, Shelton, Rowley, & Chavous, 1998), and suggest the
need for future studies to clarify factors that may account for the observed race
differences such as acculturation, multidimensional racial identity, or parent explanatory
models (Bussing et al., 1998; Sellers et al., 1998; Sullivan et al., 2007).

Finally, as predicted in Hypothesis 4, ADHD partially mediated the relationship
between family risk and adaptive ﬁrnctioning. This suggests that family risk has
impairing effects that are ADHD-speciﬁc, rather than simply indicating global
impairment. Though it is important to note that impairment is a requirement for the
ADHD diagnosis, these results support research that found an association between
ADHD and high levels of impairment (Gaub & Carlson, 1997; Gershon, 2002; Strine et

al., 2006), and extends the literature to include syndrome-speciﬁc effects. Family risk,

52

while generally impairing, has particular effects on child behavior. Those effects have
been outlined herein, and are further supported by this ﬁnding.
Clinical Implications

Family risk has been shown to pinpoint key cognitive factors in the development
of a subsequent expression of ADHD symptoms. Children’s perceptions of interparental
conﬂict appear to be keenly involved in this relationship, therefore helping children
develop appropriate coping mechanisms and methods of interpretation of conﬂict events
may prevent maladaptive behavior. In particular, it would be important to determine
whether early intervention could weaken the noticeably stronger relationship observed
between family risk and ADHD behavior in children. The unique effects of self-blame
and other possible interpretations of interparental conﬂict may be areas to begin such an
intervention.

The present study highlighted a mechanism whereby the family can affect ADHD
behavioral outcomes. In particular, parent training programs have become recognized as
important components of empirically supported treatment protocols for ADHD. Such
programs help parents understand the nature of the ADHD syndrome and teach behavior
management skills to facilitate a more productive experience for the child (DuPaul,
Guevremont, & Barkley, 1994). The measurement of parent training outcomes ranges
from assessing changes in parent stress, child behavior, or family ﬁinctioning
(Anastopoulos & Farley, 2003).

Another important treatment consideration is the lack of empirical support for
cognitively based treatment of ADHD, which involves strategies to help alter maladaptive

thinking patterns that affect behavioral ﬁlnctioning. Findings of the present study may

53

help explain the phenomenon of ineffective cognitive treatments for young children, as
their level of development may render these strategies inaccessible or require access via
alternate pathways. Adolescents may fare better, as they are developmentally more
equipped to tolerate cognitive skills training. However, there is a dearth of research on
CBT treatment efficacy in teens (Chronis, Jones, & Raggi, 2006). One study found
support for parent-teen training (Barkley, Edwards, Laneri, Fletcher, & Metevia, 2001 ),
consistent with family and neurocognitive pathways investigated in the present study.
That is, the family can be used as another tool to help remediate cognitive difﬁculty in
children with ADHD.

Finally, ethnic identity appeared to be a protective factor in relation to family risk
in the ethnic minority group such that individuals with greater ethnic identity had less
family risk. Recognition of these culture-speciﬁc effects can be useful in facilitating
treatment retention and compliance.
limitations

The cross-sectional design of the present study limited the ability to look at
temporal effects (i.e. whether the interparental conﬂict was recent), which would be
available if a longitudinal design was used. Also, as mentioned, genetic effects may have
been implicit in family risk factors, and there may be a feedback loop of ADHD effects
with respect to family risk. Also, sample size and statistical power were generally
adequate-to-ample in this investigation, however they may have been lower (though still
statistically viable) in ethnicity analyses involving relatively smaller groups (ethnic

minority group N = 103 (26%), although representative of the local population). In

54

addition, within-group speciﬁcity was limited as several cultures were included in the
ethnic minority group, which may affect generalizability (Tyson, 2004).
Conclusion

Overall, the ﬁndings support the relationship between family risk factors and
ADHD via disruption of neurocognitive functioning. The present study replicated
previous investigations of factors related to family risk. Further, it validated a latent
family risk variable, which was used to examine cognitive mechanisms involved in its
effect on ADHD. Analyses were conducted in a large diverse sample that included boys
and girls across a broad age range. The data further suggest that the relation among these
factors depends upon age and race. The present study extended existing research and
contributed new ﬁndings that will inform theory and clinical practice regarding key
family risk factors associated with ADI-II), a signiﬁcantly impairing condition in children

and adolescents.

55

APPENDICES

56

APPENDIX A

Main Tables and Figures

Table la. Description of Sample

 

 

Control ADHD p Full
Sample
N (%) 184 (47%) 210 (53%) -- 394
Number (%) Male 81 (42%) 139 (63%) <.001 220 (56%)
Number (%) Caucasian 132 (45%) 159 (55%) .370 291 (74%)
Age group (%) Adolescents 125 (55%) 103 (45%) <.001 228 (58%)
(%) Children under 13 59 (36%) 107 (64%) 166 (42%)
Age in years 13.9 (2.9) 12.7 (3.1) <.001 13.3 (3.1)
Oppositional Deﬁant Disorder Sx .87 (1.6) 2.5 (2.5) <.001 1.7 (2.3)
Conduct Disorder Sx .10 (.33) .40 (.79) <.001 .25 (.64)
Parent 1 Current ADHD Sx 1.3 (2.7) 2.5 (3.9) .001 1.9 (3.5)
Parent 2 Current ADHD Sx .98 (2.4) 1.6 (2.8) .109 1.3 (2.6)
Conners’ ADHD Total Score 4.2 (5.3) 16.5 (8.9) <.001 10.8 (9.7)
# Current DSM Inattentive Sx 1.0 (1.7) 7.6 (1.4) <.001 4.5 (3.6)
# Current DSM Hyp/Imp Sx .63 (1.3) 4.6 (3.2) <.001 2.7 (3.2)
CPIC Conﬂict Quality 2.0 (.63) 2.2 (.57) .002 2.1 (.61)
CPIC Marital Unhappiness 1.3 (.40) 1.4 (.44) .010 1.3 (.43)
CPIC SelfBlame 1.2 (.28) 1.4 (.38) <.001 1.3 (.36)
CPIC Perceived Threat 1.4 (.50) 1.7 (.59) <.001 1.5 (.57)
Parent 1 Depression Score .59 (.72) .79 (.82) .020 .69 (.78)
SSRT 217.8 262.6 <.001 241.4
(76.9) (119.3) (103.9)
RT Variability 132.2 152.9 <.001 143.2
(35.0) (40.7) (39.5)
FSIQ 109.3 102.9 <.001 105.8
(14.5) (13.1) (14.1)
MEIM 2.7 (.53) 2.8 (.57) .872 2.7 (.55)
GAF 80.9 (9.5) 69.3 (8.9) <.001 74.8 (10.9)

 

Notes to Table 1a: P-value reﬂects the two group comparison (control vs ADI-1D) and is
based on an independent samples T-test for continuous variables or a chi-square test for
dichotomous variables. N’s vary slightly for some measures due to missing data for some
disorders. “Sx” refers to symptoms. DSM = Diagnostic and Statistical Manual of Mental
Disorders, fourth edition, text revision (DSM-IV-TR). CPIC = Children’s Perception of
Interparental Conﬂict Scale. SSRT = Logan Stop Task Stop Signal Reaction Time
measured in milliseconds, RT Variability = Logan Stop Task Response Time Variability.
F SIQ = ﬁill scale IQ, MEIM = Multigroup Ethnic Identity Measure, and GAF = DSM
Global Assessment of Functioning score.

57

.Emto BEE 8 .3085 .550 .953 685:2, got: mo :82Eo 39:2: 98% 5:558 2:50 2: 3233 .53

we BEEoEEQm 0:3 ﬁcmﬁﬂtam mews—05 macaw cmmmeoswo EH own he meme» 3 l 2 meme 5.220 89:05 95% 58333 2:
£223 .2 emu. cone: £c86€aa #622: macaw 250 2F Ego—c mDmo $28305 H x 823 £5355 $88: 2: 8 c2258 225
32$ .oEooE 205830: 32:? com .ozazga Be: 083 Sec 5:3 8me E 02 Race 8: be: mowﬁcoocom .8335; 302588
com 58.9 $383 Eowcoaowﬁ an no Ewan E can 83% m> 6:208 83388 9.6% 025 05 mﬁococ 253.9 a: e335 8 .882

 

 

 

 

as: a Ase m Ase o §& 0 Ass o 3:22 550
32$ 9 3°03 % axes a $3 as £63 2: Hose 25
so. exec mm Goo: am 2 N. Eat w: $8 5 §E o8 2:an eom 5;» 85 226
3.8 0 $3 a as: _ Gee o $3 _ sesooo g
2%: 2 eat a Asa S 308 E as: 3 €an ea
Gav o 305 Ne 3a a as $5 mm Ea: : {2233 Gov
es _ 0 2 Gas 3 Assoc 3 Ass 9 mm @0on E 8230 oeom Gov
oeo. cs5 mm Game S o8. Ga 3 3, Axes 8 EURO m2 Boeom gem ea Seesaw €an
80. exam: 2 est a :5. SE 2 axe: mm 3e: 3 a; as 08.8% w 2:83
So. 35 am can one moo. can use £3 use can see oeooa 2238: 325.
$8.5 $5.5 9.9.0 9.9.0 2956
a .3352 5:23.80 m Enema—34‘ EEO ==m

 

8:05 mmoco< mQEQSonoQ Eonomzom mo eoatomom A: 039—.

58

Table 2. Regression Model: Teacher-rated ADHD Symptoms Regressed Onto Predictors
of Risk (Hypothesis 1a)

 

Rr B SEB p

 

Step 1 Developmental History .004 .119 .123 .067

Step 2 Developmental History .023 .101 .124 .056
Parent 1 Current ADHD Sx .105 .082 .089
Parent 2 Current ADHD Sx .096 .073 .091

Step 3 Developmental History .198 .088 .116 .049
Parent 1 Current ADHD Sx .000 .092 .000
Parent 2 Current ADHD Sx .057 .070 .055
FES Values -.121 .082 -.l34
FES Expressiveness .074 .099 .071
CPIC Conﬂict Quality .045 .085 .045
CPIC Marital Unhappiness .032 .087 .031
CPIC Self Blame .149 .073 .143*
CPIC Perceived Threat .126 .077 .121
Maternal Depression .018 .075 .019
Income -.085 .057 -.101
ODD Sx .203 .076 .206**
CD Sx .062 .089 .053

 

Notes to Table 2: “Sx” refers to symptoms. FES = Family Environment Scale, CPIC =
Children’s Perception of Interparental Conﬂict Scale, ODD = Oppositional Deﬁant
Disorder, CD = Conduct Disorder. The dependent variable used in this analysis was
teacher-reported total ADHD symptoms on the Conners’ Teacher Rating Scale. * p < .05,
** p < .01.

59

.So u gem e5 .80 u <32: .8: u Eu .8. A a .34 u $8 .1. 2 .8 NR .8.

v & ... Baum 85:00 3:83:25 CO 85389: @5620 u 030 .3255 :03 we: on?» X>Bm on: 3 game of E E88% 8:
2 $5 5530: J E 35.: £3 32:20 OED .x mo :5. .5535: E353. co: > E owcmso we Enos; of 30% 9 83st? 33030
:SW “:8:— oﬁ o6 80:22:; of :o :83 BwEaESm 2m 82? 82C. :63: 3:03: 825 $2? X>Bm SEE U N EEK o: 382

am 2358:: :2: ram: Co 382 $5 ._ 2%:

60

3 23503: V2: haste E52 <5 ._ 2%:

*0.“ III

*_w.

*_h.

 

2:85

 

 

:ommmoaom
333:2

 

 

Bah—H USU

 

 

*wﬁ.

082m UEU

 

 

/ I

32.82 UEU

 

 

 

3:30 2::

 

 

#0.

*mh.

*ow.

*ow.

61

.8. n :35 Ea .8. n <82: .8. u :6 ,8. A m 4 08 n cam n 2 .0: ”x .8. v a .. .28m 8:“: 28$ 05550

0:: :0 8:808? Din? :20: 088-558: 33 0335? E00500: 05H .390me Eamon. 3:250:00 H QQO 8:588? u

xm 6282000 3:532 n 00D dﬁom 85:00 35580:: (:0 5:000:45 9:06:50 u 0E0 00:30:: :8: mm: 02$ X>Em 05 3
game 05 E 683% :0: ﬂ WE: 5530: J S 082.: 83 5:050 QED .520: 08.502 0:03 30:; X>Em 0.50.2 ”N 0§&Q 0: 3:02

Sm 285003: DQO 0:: manEwm 30H QIQ< 00:0 Emmewom 03¢ bimmmo 2mm .N 0.5wE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2:003 0w.
QDQ All * mo .
COO
0825 .
T “x mm
UEU
052m .
All. on
0E0 *
NW 139—;
GE: 133:2 .
. LII 3
- . 030 *
* w m
32:20 .
. II E
*8. .030 *

 

 

 

62

.8. u gm :0:: .3. u <Mm§ .3.

n E0 .8. A a dem n 90». u 2 ,m: «N 005900 00>» :: :0:3 0800 0:: 338080 0:03 0:30: :3: 003300 :0: m: 0w: 65%.: 80:30
0:: 5 .3. v m .. 08% $53: 8:003. 8:05:00 0:: :0 080685. man? :80: 088-8500: :03 05:50:, 80:05:00 0: H 0808?? n
xm £28880 3:882 u :00 .0_00m 85:00 8808808: 50 55:020.: m.:0:0=:0 n 030 00:30:: :00: m0: 0::? Egm 0:: m:
05%.: 0:: E 80:80: :0: m: 0::: :0>030: : :0 0005 83 5530 07:0 30:0: 00::000: 0:03 002:; 5:05 0:32 ”m. 0300K 0: 00:02

3N 38:89:30 035 3:80,: 8:0 00888.5: 080:9:Sm 3:0,: 9:9: :0 2mm .n 95!:—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0::00:: m0.
wmr
* \ Qum— ,A.|.|l *VO.
*mm
80:: .. .
\i All. 0
*8. 080 t
80.10 mam—MA.“ 4 0.05 of
*2. 32:5— * 082m 0 on.
B. 0E0 . *
*mo
/ 13:32 4' *wm.
. 0E0
*hv
320:0 :0.
“LN. “0th :

 

 

 

63

.3. u 0:200 0.0 .8. u <32: ,8. u :00 .8. A 0 00.9: n $3 n

Z ,3; NR .3. v Q * .000m @5010 00:000H E0500 .1. mar—l0 5:20_:0E:\>:_>m:00:00:m: QIQ< n :00»: .:0_::0::0:: Dim? M 5:05
0052000 05302 n 000 “200m 65:00 0.000985 00 8:000:00 00:06::0 M 0:00 000380 :00: m0: 020> 006% 0:: m0
0ch 0:: :0 E00000 :0: m: 02: :0>0>>0: m: :0 0000.: 003 3:000 0:00 .500: 08000: 003 $29 Kazm 00:02 ”0 250$ 0: 00:02

Gm 50:803.: :3”: £800 0::0 0030000: 080:: 300.0 00 20m .0 0.53m

64

 

 

 

 

 

 

 

 

 

 

 

 

*am.

 

Gm $005003: 0:03: 3:80": 0::0 0000003: 080:: $30 .0 20m .0 0.53::

*2...

 

 

 

 

 

 

 

 

 

 

 

0800::
*CN... QQQ

0mm

x; 02:

*Nc UHAU
*cm.

MFR—m

S. 0:::0
1.0.

.20:

“wow: 07:0

0:80

*CN. “0th

 

 

1.0.

0N0.

*om.

*cc.

65

0:080 n 0::: 0:0 ,8. n gm amo. n <m:m:>:~: 00. H :00 do. A «:0sz n Guam HZ .va NR 02060

003 :_ 5:3 080m 0:: 0:00:83 0:03 0:30: :3: 02060 :0: m: 0w0 0:30: 80:80 0:: :: 200050: :0:: w:080 :0: 80:00:08

:0:: 0:80“: 0:: w:080 8000:0800 0:0 :0008 0:: :0 00:0 :0: :0: 0:: :0 80:00:00 0: H 0:: 00:00 0 0: 00:00:08 0:8: 0:0 8:00
::00_.:::w_m-:07: .3. v a: 0. .200m @000: 0:00.: 0:00:00 0:: :0 0890800 9:3 .0:0: 00t000:-:0:000: 003 0::0::0> 800:0000

0: :1 800020 00:03:08: 0:00m :0: u 05: “08808:: n xm 80:30:02: _0::0:0:>: n 000 000320 :00: 00: 30> X>0:m 0:: 00
08%: 0:: :_ 800000 :0: m: 0::: :0>030: :: :0 0000.: :03 5:030 0:00 80:0: 00::000: 0:03 0020:» X>0:m 00:3): H 0:030 0: 00:02

:0m 30050033 00:00:): 00::m:00m 80::3: aw:::0:m:_0 :86: 0000: :0::0000: .m 0::w:,..:

66

0m $005095: 00:02 000083: 50:03 ,w:_:0:m:_0 mi? 0002 800002 .m 0.59::

 

 

 

 

 

 

 

 

 

 

 

 

05:: 0500::

‘ “ram:- *wmf

_ 7 \ \_0
*hm.

V \v—

:00:—H
0.5: 00.
mm ESE A maﬁa: *
Gin—4 *Nm.

 

 

 

v;

00m.

 

 

3:5.0

_

f
._ Ea Ts
./_ 3...: 7a.
/_ _.\

:0::0_=w0m
023530

 

 

 

67

.mo03w u 05 0:0 #0. n gm 0::. n <mm0ﬁ: 0::. u :00 :00. v 0 .029 n :03 n2 .03 «N 200000: :0:: m:080 :0: 0:900:08
:00: 3:80”: 0:: m:080 8000:0800 0:0 088 0:: 00 00:0 :00: :00 0:: :0 8000—0800 0: 0. 0:: 00:000 0 0: 000008 08:00 0:0

0:00 802.0%:0007: .3. v Q ... .0_00m m800y: :0:o00H 00800 0:: :0 0830800 9:0»: :08: 088-0500: 003 0::0::0> 800:0000
0: E 800000 00:00:08: 20$ :00 n 0:m0 n0890800 .I. xm 00:08:02: 0800:): n 000 000320 :00: m0: 00_0> §00m 0:: 00
0:00.: 0:: 8 80:0000 :0: m: 0::: :0>030: : :0 0000.: 003 3:000 0:00 80:0: 00::000: 0:03 m00_0> 00:05 00:02 ”0 0:00»: 0: 00:02

3m 0:00:80sz 000:0 00:0 I 0003050 09: 0000200,: 0:0 $800020 m8? :0002 800000: .0 0:30:

 

 

 

 

 

 

 

 

 

 

 

0:9: 0800:: _
* mm.. 0.30.-
8.- \ \«_ :0: ;
0.0m. ‘ -
\v_ 82.; l_4/
0m ESE _A aw”: "wow. ‘0 0 m _.
0:::04: *mm. .

 

 

 

 

0:.

 

 

:0::0_0w0~:
0>_:_:w00

 

 

68

00.0000 u 0:0 05.. .8. u 030:0 .00. n 00030: .00. n :00 .50. v 0 .020: u :00 n2 .000 :0: 000200: :0:: 0880 :2. 0:88:05
0:05: 0:800 0:: w:080 0:000:0800 0:0 088 0:: 00 00:0 00— :00 0:: :0 0:80—0:00 0:.w 0:: 00:000 0 0: 000008 0:80 0:0

0:00 8000800002 .8. v 0 ... .0_00m m:00y: :0:000.:. 0:00:00 0:: :0 0800800 9:90: :08: 0008:0500: 003 03000> 800:0000
0: H 800000 00:00:08: 0:00m :0”: .I. 0:9: 0800800 U xm 08008000 :0::002 n 000 000380 :00: 00: 00_0> X>00m 0:: 00
0:09: 0:: 8 80:0000 :0: 0: 0::: :0>030: : :0 0000.: 003 3:000 0:00 80:0: 000000: 0:03 000_0> X>00m 00:02 K 0:000:02 0: 00:02

Sm 00000005 000:0 80000—0000: I 0003000: 0m< 0000:0002 0:0 0800080 m800 :0002 :000002 .0 0:03:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0:9: 0800:: L
*h m... *w _ .u -
*0 : .u \ \— a—OD [—
0mm. -
w \‘— -
:00:—H 7
00 .080 r 00“me *0. , -0.
Gmn< 0. mm. _. h - r
0.00 V— enn—Em— \_b\
3?: -
0 :0
.0 _ m. - .
.1. *ON.
* m r -
4 r
:00:—03¢

0>_:_:w00

 

 

69

00.3% u 08 ,8. n 0200 .00. u

«£52m Kw. u E0 .50. v 0. £08. n $0M n 2 08 «N 200800: 80:: w:080 :0: 0:900:08 008% 0:80”— 05 w:080 0:80—0:00 0:0
.0008 0:: («0 080 $2 :00 08 :0 0080000800 0::. .08— 00500 0 3 0000068 08:8 0:0 0500 802.088.0002 .3. v A ... 0.00m w800m
00000.0 “80::00 0;: :0 08000800 DIE .08: 0003080000 003 0B0t0> 0800:0000 00H 000002 8803 0850 0:80:02 u
Emma): 0:0:000 00:0w_=0:£ 0.00m :00 n 05”— 0800080 U xm 08000500 050005 n 000 00030:: :000 m0: 00_0> X505 05 m0
0:090 05 8 880000 :0: 8 8:: :0>030: J :0 0000.0 003 3:000 050 80.0: 000L000: 0:03 m00_0> X>Em 08004 .0 0:sz 0: 00:02

0m 800500010 000:0 08000000 I 000305: 000%202 0:0 $80580 wEm0 0002 0000:0002 .m 0:03..—

70

Gm 2358?: anew $68sz I wombmc< 6592232 98 $55520 was: Eco—2 223622 .w 92%:—

 

 

 

    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00...
25.2 9% 2:85 _
4.. ....1 ‘ *wm... . *hmf ‘
\— aun _
8. 8.- 3.. -
23...; _5,
mm EBB A ham;— *3: ‘ ‘ .
. ESE N.
G=Q< *VM . . ‘
*mv ’— oEEm— T
*no. ‘ ‘
// /p_ 73.32 _A/
*wm. - I
I. * _ n
* _ P ‘
/._ 3:55 T\
55353— ‘

«ﬁg—.mou

 

 

71

.8528 u 05 .8. u gm .8. n

$532 .5. n 50 48. v a. «3.: u $3 u 2 .mg N 29.28. :2: macaw a: £8565 vim ream 2: was; 20:20:00 2“
EcoE 2: mo 3% to. Sm of co 20:58.50 2H dc: 35% a 3 ©8865 Satﬁ 8m mica Emoﬁcwaéoz .3. v a ... .2QO mega
$508k 22:80 2: no 38683 93% :38 v88-$num8 $3 03353 Eoccoqon 2; .2382 >553 065m gems—:2 u
.252 45:95 oocowEBE 28m :5 u 9mm .3558? u xm £23239 3:532 .1. non .8235 coon was 02? vaEm 2: 3
game of E 683% 8c £ 35 5530: x S 35 33 5:30 U30 .520; 8:82 203 mos—9 X>Bm 91:32 .6 mSME S @802

Gm $0quva @290 $552 225m I mowing mamas—:2 Es $55320 was: ESE :25on .a ohzwﬁ

72

    

Gm 28583: 955 5:052 2255 I 89:92 asemésE was ,wctﬁmsc wEm: .252 guano—2 .a PSwE

 

 

 

 

 

 

 

*OMV
Sim—2 O—mm
* mm.-
m _ . n ﬂ .-
gag
mm .53. A 2.55—
Gﬂn< * mm. .
1...
mo. mm.
Eta—@3—

95:.ch

 

*OME

 

 

 

 

 

 

 

 

 

 

 

73

.vo. n gm :8 ,3. u <mm2m ,3. M EU ,8. v Q ,Nndm .1. Avon n 2 K: N .8828 83 z :23» 08mm 2: 38838

203 3:52 :5 ,8828 8: 2 own 68%: 8280 2: E 2868. 8: w808m 8: 89868 82¢ 38$ 2: wacEa 258.088
8 888 2: mo 25 :2 8: 2: :0 288888 2:. 0:: 8:8: a E 882:8 82:5,: 08 2:8 88286.82 .8. v a ... 6:825
832:5 EEO 2: :0 28m 882288 :39 8:82-828 :8 208 880885 :0 808mmomm< :320 88828828?

2: mo :88 a 3 88888 208 880882.: 3:88 26888 a 83 28:? 82828: 2; .2an 88¢ 8288b .82800 2:
co m8088~$ mag :28 882.8282 M xm 80H Q22 8288va 8:882 H 8Q .8238 88 was 28> X>Bm 2: mm 6ch
2: 8 828% 8: 8 3: 8838 J 8 8x: 83 3:80 050 .820; 8:82 203 82? XZBm 332 SN EamE S 882

? 32:08wa 880885 o>88< :8 xmi 38$ (:0 8:882 9:3 .3 charm

74

3V $2888 82228 2:3? Es V58 :88 .3 8882 8:3 .8 250.;

 

883.25% _L

 

~88:—

 

non

 

 

“coup—H;

 

 

u>_Es:< _4 * mm;

 

 

2:85

 

 

.552

 

*ﬁm.

 

 

mm Eﬁ
a=a<

 

5:80

TTTT

 

 

 

oo.

75

APPENDIX B
Family Environment Scale Analyses

Table B]. Family Environment Scale EFA l-factor Solution, 10 Subscales

 

 

FES Subscale Factor 1
Cohesion .605
Expressiveness .367
Conﬂict . 120
Independence .583
Achievement .391
Intellectual-Cultural .542
Active-Recreational .5 72
Moral-Religious .507
Organizational .677
Control .554

 

Notes to Table B1: Mplus Analysis using FIML imputation, Maximum Likelihood
extraction with Promax rotation; 78 (35, N = 374) = 200.41, p < .001, RMSEA = .1 1,
RMR = .08.

76

APPENDIX B (cont’d)

Table B2. Family Environment Scale EF A 2-factor Solution, 10 Subscales

 

 

FES Subscale Factor 1 Factor 2
Cohesion .635

Expressiveness .527

Conﬂict .347
Independence .463

Achievement .268
Intellectual-Cultural .75 1
Active-Recreational .583
Moral-Religious .355

Organizational .447

Control .967

 

Notes to Table 82: Mplus Analysis using FIML imputation, Maximum Likelihood
extraction with Promax rotation; 2 factors, r = .480; x2 (26, N = 374) = 88.20, p < .001,
RMSEA = .08, RMR = .05.

77

 

APPENDIX B (cont’d)

Table B3. Family Environment Scale EF A 4-factor Solution, 10 Subscales

 

 

FES Subscale Factor 1 Factor 2 Factor 3 Factor 4
Cohesion .655
Expressiveness .3 12
Conﬂict -1.693
Independence .363

Achievement .267

Intellectual-Cultural .988

Active-Recreational .456
Moral-Religious .295

Organizational .6 1 3
Control .892

 

Notes to Table B3: Mplus Analysis using FIML imputation, Maximum Likelihood
extraction with Promax rotation; 4 factors, r < .68; x2 (1 1, N = 374) = 13.02, p = .29,
RMSEA = .02, RMR = .02.

78

APPENDIX B (cont’d)

Table B4. Family Environment Scale EF A 2-factor Solution, 6 Subscales using FIML

 

 

Imputation

FES Subscale Factor 1 Factor 2
Control .860

Organizational .580

Independence .41 5

Moral-Religious .425
Intellectual-Cultural .807
Expressiveness .541

 

Notes to Table B4: Mplus Analysis using F IML imputation, Maximum Likelihood
extraction with Promax rotation; 2 factors, r = .506; x2 (4, N = 374) = 6.02, p < .19,
RMSEA = .04, RMR = .02.

Table B5. Family Environment Scale EFA 2-factor Solution, 6 Subscales using Raw

 

 

Data

FES Subscale Factor 1 Factor 2
Control .756

Organizational .641

Independence .577

Moral-Religious .500
Intellectual-Cultural .777
Expressiveness .500

 

Notes to Table B5: SPSS Analysis using Raw Data, Maximum Likelihood extraction with

Promax rotation; 2 factors accounted for 60.25% of the variance, r = .447, a = .71; x2 (4,
N: 374) = 5.94, p < .20.

79

APPENDIX B (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

FES
33* _,. Intellectual-
Cultural \
.794:
"H —> FES ‘— 5”... Expressiveness l 0‘
'/ ) Expressivenes ' .
,7”, I ‘ FES
"' Independence
, _ k, .*
82* I ‘ FES 'M'oral .\(_
' Religious \
.43*
40* _H FES 73* 1.0:.
Organizational
AS“
3“” " FES Control

 

 

 

 

Figure Bl. CF A of FES 2-Factor Solution with 6 Subscales

Notes to Figure B]: Mplus StdYX values were reported herein. These values are
standardized based on the variances of the latent and observed variables to show the
amount of change in Y per standard deviation unit of X. FES Intellectual-Cultural and
FES Independence were ﬁxed at 1; however this is not apparent in the ﬁgure, as the
StdYX value has been provided. FES = Family Environment Scale. * p < .05. x2 (5, N =
374) = 10.84,p > .05, CFI = .99, RMSEA = .06, RMR = .03.

80

APPENDIX C
CPIC Factor Structure

Table Cl. CPIC Factor Structure Validated by Nigg et al. (under review)

 

CPIC Conﬂict Marital Self Perceived
Item # Quality Unhappiness Blame Threat
l .465
12 .765
26 .747
30 .787
35 .809
2 .587
11 .733
14 .612
29 .711
37 .456
42 .404
46 .766
3 .622
8 .347
16 .722
19 .850
25 .591
28 .719
36 .688
40 .452
47 .350
6 .663
13 .535
15 .767
23 .900
44 .625

 

 

Notes to Table C1: Maximum likelihood extraction with Promax rotation. 4 factors
accounted for 52.59% of the variance, (1 = .89. The conﬁrmatory analysis conducted by
Nigg et al. (under review) resulted in the following ﬁt indices for the 4-factor solution
outlined above: X? (293, N = 1190) = 1389, p > .05, CFI = .935, RMSEA = .056.

81

APPENDIX D

Alternative Regression Analyses for Hypothesis 1a

Table D1. Teacher-rated ADHD Symptoms Regressed Onto Predictors of Risk, With

 

 

ODD & CD in Step 2
R2 B SE3 [1

Step 1 Developmental History .004 .119 .123 .067

Step 2 Developmental History .113 .078 .118 .044
ODD Sx .227 .077 .231"
CD Sx .135 .089 .116
Parent 1 Current ADHD Sx .039 .079 .033
Parent 2 Current ADHD Sx .070 .070 .067

Step 3 Developmental Histmy .198 .088 .116 .049
ODD Sx .203 .076 .206W
CD Sx .062 .089 .053
Parent 1 Current ADHD Sx .000 .092 .000
Parent 2 Current ADHD Sx .057 .070 .055
FES Values -.121 .082 -.l34
FES Expressiveness .074 .099 .071
CPIC Conﬂict Quality .045 .085 .045
CPIC Marital Unhappiness .032 .087 .031
CPIC Self Blame .149 .073 .143*
CPIC Threat .126 .077 .121
Maternal Depression .018 .075 .019
Income -.085 .057 -.101

 

Notes to Table DI : Sx = symptoms, ODD = Oppositional Deﬁant Disorder, CD =
Conduct Disorder, CPIC = Children’s Perception of Interparental Conﬂict Scale. * p
<.05, **p < .01. Step 3 brought a signiﬁcant gain in r2 (A r2= .085, p = .010).

82

APPENDIX D (cont’d)

Table D2. Teacher-rated ADHD Symptoms Regressed Onto Predictors of Risk, Without
ODD & CD

 

 

R’ B SE B [3
Step 1 Developmental History .005 .127 .122 .071
Step 2 Developmental History .024 .108 .123 .060
Parent 1 Current ADI-ID Sx .112 .081 .095
Parent 2 Current ADHD Sx .093 .072 .089
Step 3 Developmental History .147 .111 .118 .062
Parent 1 Current ADHD Sx .035 .093 .030
Parent 2 Current ADHD Sx .078 .071 .074
FES Values -.103 .082 -.114
FES Expressiveness .047 .099 .045
CPIC Conﬂict Quality .021 .086 .021
CPIC Marital Unhappiness .051 .088 .051
CPIC SelfBlame .162 .073 .156*
CPIC Threat .158 .077 .152*
Maternal Depression .043 .076 .045
Income -.112 .058 -. 134

 

Notes to Table D2: Sx = symptoms, ODD = Oppositional Deﬁant Disorder, CD =
Conduct Disorder, CPIC = Children’s Perception of Interparental Conﬂict Scale. * p
<.05. Step 3 brought a signiﬁcant gain in r2 (A r2= .123, p < .001).

APPENDIX E
Bootstrapping Method (Hypothesis 3a)

A path analysis was conducted in Mplus using bootstrapped standard errors for
the indirect effects. The model tested whether the relationship between family risk and
ADHD was partially mediated by cognitive regulation. Bootstrapped standard errors were
computed using 1000 draws. Bootstrapping is a new method (Shrout & Bolger, 2002),
and at this time statistical software does not allow for clustering of nested data while
executing the bootstrap option, therefore clustering was not used in this model.

The model, shown below, provided a good overall ﬁt: X2 (25, N = 394) = 34.18, p
> .05, CF] = .97, RMSEA = .04, and SRMR = .05. The indirect effect for this model
(StdYX = .028) was signiﬁcant at p > .05. Thus, family risk predicted poorer regulation
on cognitive tasks, and the relationship between family risk and ADHD was partially
mediated by cognitive regulation.

The Mplus bootstrap option was used to prevent Type 2 error; however, this is of
less concern in these data given the signiﬁcant effect. Therefore, the cluster feature was
used in place of the bootstrap option due to the necessity of clustering the nested family
data.

84

.8. v m ... Baum mega 3:0on .mcoccoo 2: so 5289?? OUT—Day :39 832-5588 $3 292.?
Eon—.326 2:1 .Eotozo oocowEBE 28m 23m .1. 9mm .:o_mmo.aoQ 8an22 u non .3255 :89 mm: 029 X>Bm 2: mm ,oSma
2: E 223% 8: 2 mu: 8526: J 3 35m 33 5:25 OED .580: 3:82 20>» $2? X>Bm 332 U NM QEME S $82

@5630 302:3 60502 amhmaoom mam: 5:582 .5— 03w?—

85

magma—U 505.3 €262 ambmuoom mEm: 5:332 .5— 9.33

O—mh ; 2:85 _

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

*nmv *omr -
*m_.- \ \— a—on— _
.mm. .
V \V_ .
_
..m .83 A sum”; .8. ,

GEQ< *3. . .
*3. .7— QEEM— _
*mh. -
/ /._ .f. _

*nh.

*om.
3:20 _
5553?;— -
~33:on

 

 

3.253 m— £u=oaa<

86

REFERENCES

87

REFERENCES

Achenbach, T. (1991 ). Manual for the Child Behavior Checklist/ 4-18 and 1991 proﬁle.
Burlington, VT: University of Vermont Department of Psychiatry.

Acosta, M. T., Arcos—Burgos, M., & Muenke, M. (2004). Attention deﬁcit/hyperactivity
disorder (ADHD): Complex phenotype, simple genotype? Genetics in Medicine,
6, 1-15.

Akshoomoff, N. (2002). Selective attention and active engagement in young children.
Developmental Neuropsychologa 22(3 ), 625-642.

Ambrosini, P. J. (2000). Historical development and present status of the Schedule for
Affective Disorders and Schizophrenia for School-Age Children (K-SADS).
Journal of the American Academy of Child and Adolescent Psychiatry, 39(1), 49-
58.

Anastopoulos, A. D., & Farley, S. E. (2003). A Cognitive—Behavioral Training Program
for Parents and Children with Attention-Deﬁcit/Hyperactivity Disorder. In A. E.
Kazdin & J. R. Weisz (Eds), E vidence-based psychotherapies for children and
adolescents (pp. 187-203). New York, NY: Guilford Press.

Angold, A., Erkanli, A., Farmer, E. M., Fairbank, J. A., Burns, B. J ., Keeler, G., et al.
(2002). Psychiatric disorder, impairment, and service use in rural African
American and white youth. Archives of General Psychiatry, 59, 893-901.

APA. (2 000). Diagnostic and statistical mamral of mental disorders: DSM-IV-TR (4th ed
text revision ed). Washington, DC: American Psychiatric Association.

Appelbaum, A. S. (1977). Developmental retardation in infants as a concomitant of
physical child abuse. Journal of A bnormal Child Psychology, 5(4), 417-423.

Arnold, L. E., Elliot, M., Sachs, L., Bird, H., Kraemer, H. 0, Wells, K. C., et a]. (2003).
Effects of ethnicity on treatment attendance, stimulant response/dose, and 14-
month outcome in ADHD. Journal of Consulting and Clinical Psychology, 71(4),
713-727.

Barkley, R. A. (1997). Behavioral inhibition, sustained attention, and executive functions:
Constructing a unifying theory of ADHD. Psychological Bulletin, 121, 65-94.

Barkley, RA. (1998). Attention deﬁcit hyperactivity disorder: A handbook for diagnosis
and treatment (2nd ed). New York: Guilford.

88

Barkley, R. A. (2003). Attention-Deﬁcit/Hyperactivity Disorder. In E. J. Mash & R. A.
Barkley (Eds), Child Psychopathology (2nd ed., pp. 75-143). New York: The
Guilford Press.

Barkley, R. A., Edwards, G., Laneri, M., Fletcher, K., & Metevia, L. (2001). The efﬁcacy
of problem-solving communication training alone, behaviour management
training alone, and the combination for parent—adolescent conﬂict in teenagers
with ADHD and ODD. Journal of Consulting and Clinical Psychology, 69(6),
926—941.

Barkley, R. A., Fischer, M., Smallish, L., & Fletcher, K. (2002). The persistence of
attention-deﬁcit/hyperactivity disorder into young adulthood as a function of
reporting source and deﬁnition of disorder. Journal of A bnormal Psychology,
1 1 [(279-289)

Barkley, R. A., Fischer, M., Smallish, L., & Fletcher, K. (2004). Young adult follow—up
of hyperactive children: Antisocial activities and drug use. Journal of Child
Psychology and Psychiatry, 45(2), 195-211.

Barkley, R. A., Karlsson, J., Pollard, S., & Murphy, J. V. (1985). Developmental changes
in the mother-child interactions of hyperactive boys: Effects of two dose levels of
Ritalin. Journal of Child Psychology and Psychiatry and Allied Disciplines, 26,
705-71 5.

Beck, A. T. (1964). Thinking and depression: 11. Theory and therapy. Archives of
General Psychiatry, 10, 561-571.

Bentler, P. M. (1990). Comparative ﬁt indexes in structural models. Psychological
Bulletin, 107, 238—246.

Betancourt, H., & Lopez, S. R. (1993). The study of culture, ethnicity, and race in
American psychology. American Psychologist, 48, 629-63 7.

Biederman, J ., Faraone, S. V., Keenan, K., Benjamin, J ., Krifcher, B., Moore, C., et al.
(1992). Further evidence for family-genetic risk factors in attention deﬁcit
hyperactivity disorder. Patterns of comorbidity in probands and relatives in
psychiatrically and pediatrically referred samples. Archives of General
Psychiatry, 49, 728-738.

Biederman, J., Faraone, S. V., Keenan, K., Knee, D., & Tsuang, M. T. (1990). Family-
genetic and psychosocial risk factors in DSM-HI attention deﬁcit disorder.
Journal of the American Academy of Child and Adolescent Psychiatry, 29(4),
526-533.

89

 

Biederman, J., Faraone, S. V., & Monuteaux, M. C. (2002). Impact of exposure to
parental attention—deﬁcit hyperactivity disorder on clinical features and
dysfunction in the offspring. Psychological Medicine. Vol, 32(5), 817-827.

Biederman, J., Milberger, S., Faraone, S. V., Kiely, K., & et al. (1995a). Family-
environment risk factors for attention-deﬁcit hyperactivity disorder: A test of
Rutter's indicators of adversity. Archives of General Psychiatry. Vol, 52(6), 464-
470.

Biederman, J ., Milberger, S., Faraone, S. V., Kiely, K., & et al. (1995b). Impact of
adversity on ﬁmctioning and comorbidity in children with attention-deﬁcit
hyperactivity disorder. Journal of the American Academy of Child and Adolescent
Psychiatry. Vol, 34(11), 1495-1503.

Booth, J. R., Burman, D. D., Meyer, J. R., Lei, Z., Trommer, B. L., Davenport, N. D., et
al. (2003). Neural development of selective attention and resonse inhibition.
Neurolmage, 20, 737-751.

Bussing, R., Schoenberg, N. E., & Perwien, A. R. (1998). Knowledge and information
about ADHD: evidence of cultural differences among African-American and
white parents. Social Science and Medicine, 46(7), 919-928.

Button, T. M. M., Thapar, A., & McGufﬁn, P. (2005). Relationship between antisocial
behaviour, attention-deﬁcit hyperactivity disorder and maternal prenatal smoking.
British Journal of Psychiatry, 18 7(2), 155-160.

Carlson, C. L., & Mann, M. (2000). Attention-deﬁcit/hyperactivity disorder,
predominately inattentive subtype. Child and Adolescent Psychiatric Clinics of
North America, 9(3), 499-510.

Castellanos, F. X., & Tannock, R. (2002). Neuroscience of attention deﬁcit/hyperactivity
disorder: The search for endophenotypes. Neuroscience, 3, 617-628.

Chi, T. C., & Hinshaw, S. P. (2002). Mother-child relationships of children with ADHD:
The role of maternal depressive symptoms and depression-related distortions.
Journal of A bnormal Child Psychology, 30(4), 387-400.

Chronis, A. M., Jones, H. A., & Raggi, V. L. (2006). Evidence-based psychosocial
treatments for children and adolescents with attention-deﬁcit/hyperactivity
disorder. Clinical Psychology Review, 26, 486-502.

Cicchetti, D., & Curtis, W. J. (2006). The Developing Brain and Neural Plasticity:
Implications for Normality, Psychopathology, and Resilience. In D. Cicchetti &
D. J. Cohen (Eds), Developmental Psychopathology: Developmental
Neuroscience (Vol. 2, pp. 1-64). New Jersey: Wiley.

90

Cohen, J. (1977). Statistical power analysis for the behavioral sciences (revised edition).
New York: Academic Press.

Cohen, J. (1992). A Power Primer. Psychological Bulletin, 112(1), 155-159.

Conners, C. K., Erhardt, D., Epstein, J. N., Parker, J. D., Sitarenios, G., & Sparrow, E.
(1999). Self-ratings of ADHD symptoms in adults: 1. Factor structure and
normative data. Journal of A ttention Disorders, 3(3), 141-151.

Conners, C. K., Sitarenios, G., Parker, J. D., & Epstein, J. N. (1998). The revised
Conners’ Parent Rating Scale (CPRS-R): Factor structure, reliability, and criterion
validity. Journal of A bnormal Child Psychology, 26(4), 257-268.

Cornblatt, B. A., & Malhotra, A. K. (2001). Impaired attention as an endophenotype for
molecular genetic studies of schizophrenia. American Journal of Medical
Genetics, 105(1), 11-15.

Coull, J. T. (1998). Neural correlates of attention and arousal: Insights from
electrophysiology, functional neuroimaging and psychopharmacology. Progress
in Neurobiology, 55, 343-361.

Counts, C. A., Nigg, J. T., Stawicki, J. A., Rappley, M. D., & von Eye, A. (2005). Family
Adversity in DSM-IV ADHD Combined and Inattentive Subtypes and Associated
Disruptive Behavior Problems. Journal of the American Academy of Child and
Adolescent Psychiatry, 44(7), 690-698.

Crosbie, J ., & Schachar, R. (2001). Deﬁcient inhibition as a marker for familial ADHD.
American Journal of Psychiatry. Vol, 158(11), 1884-1890.

Cunningham, C. E., Benness, B. B., & Siegel, L. S. (1988). Family ﬁmctioning, time
allocation, and parental depression in the families of normal and ADHD children.
Journal of Clinical Child Psychology, 1 7, 169- 1 77.

Davies, P. T., Cummings, E. M., & Winter, M. A. (2004). Pathways between proﬁles of
family ﬁJnctioning, child security in the interparental subsystem, and child
psychological problems. Development and Psychopathology, 16, 525-550.

Dawson, G., Panagiotides, H., Klinger, L. G., & Spieker, S. (1997). Infants of depressed
and nondepressed mothers exhibit differences in frontal brain electrical activity
during the expression of negative emotions. Developmental Psychology, 33(4),
650-656.

Deater-Deckard, K., Dodge, K. A., Bates, J. E, & Pettit, G. S. (1996). Physical discipline

among African American and European American mothers: Links to children's
externalizing behaviors. Developmental Psychology, 32(6), 1065-1072.

91

Derogatis, L. R. (1994). Symptom checklist-90-R (SCL-90-R) administration, scoring,
and procedures manual. Minneapolis, MN: National Computer Systems, Inc.

DuPaul, G. J. (1981). Parent and teacher ratings of ADHD symptoms: psychometric
properties in a community-based sample. Journal of Clinical Child Psychology,
20, 245-253.

DuPaul, G. J ., Guevremont, D. G., & Barkley, R. A. (1994). Attention-deﬁcit
hyperactivity disorder. In L. W. Craighead, W. E. Craighead, A. E. Kazdin & M.
J. Mahoney (Eds), Cognitive and Behavioral Interventions: An Empirical
Approach to Mental Health Problems. Boston, MA: Allyn & Bacon.

Ellis, A. (1962). Reason and emotion in psychotherapy. New York: Lyle Stuart.

Epstein, J. N., Conners, C. K., Erhardt, D., Arnold, L. E., Hechtman, L., Hinshaw, S. P.,
et a1. (2000). Familial aggregation of ADHD characteristics. Journal of A bnormal
Child Psychology, 28, 585-594.

Epstein, J. N., Conners, C. K., Hervey, A. S., Tonev, S. T., Arnold, L. E., Abikotf, H. B.,
et al. (2006). Assessing medication effects in the MTA study using
neuropsychological outcomes. Journal of Child Psychology and Psychiatry.
47(5), 446-456.

Eiraldi, R. 3, Power, T. J ., & Nezu, C. M. (1997). Patterns of comorbidity associated
with subtypes of attention-deﬁcit hyperactivity disorder among 6- to 12-yr-old
children. Journal of the American Academy of Child and Adolescent Psychiatry,
36(4), 503-514.

Ernst, M., Moolchan, E. T., & Robinson, M. L. (2001). Behavioral and neural
consequences of prenatal exposure to nicotine. Journal of the American Academy

of Child and Adolescent Psychiatry, 40(6), 63 0-641.

Faraone, S. (2000). Attention deﬁcit hyperactivity disorder in adults: Implications for
theories of diagnosis. Current Directions in Psychological Science, 9(1), 33-36.

F araone, S, & Biederman, J. (1994a). Is attention deﬁcit hyperactivity disorder familial?
Harvard Review of Psychiatry, 1, 271-287.

Faraone, S., & Biederman, J. (1994b). Genetics of attention-deﬁcit hyperactivity
disorder. Child & Adolescent Psychiatry Clinics of North America, 3, 285-3 02.

Faraone, S., & Biederman, J. (1998). Neurobiology of attention-deﬁcit hyperactivity
disorder. Biological Psychiatry, 44, 951-958.

92

Faraone, S., Perlis, R. H., Doyle, A. E., Smoller, J. W., Goralnick, J. J., Holmgren, M. A.,
et al. (2005). Molecular genetics of attention-deﬁcit/hyperactivity disorder.
Biological Psychiatry, 5 7(1 1), 1313-1323.

F aul, F ., & Erdfelder, E. (1992). GPOWER: A priori-, post hoc-, and compromise power
analyses for MS-DOS [Computer Program]. Bonn, Germany: Bonn University,
Department of Psychology.

Frolich, J ., & Lehmkuhl, G. (2004). Differential diagnosis of sleep and vigilance
disorders in children. Journal of Child Psychology and Psychiatry, 53(1).

Fuster, J. M. (2002). Frontal lobe and cognitive development. Journal of Neurocytology,
31(373-385).

Gaub, M., & Carlson, C. (1997). Gender differences in ADHD: A meta-analysis and
critical review. Journal of the American Academy of Child and Adolescent
Psychiatry, 36(8), 1036-1045.

Gerdes, A. C., & Hoza, B. (2006). Maternal attributions, affect, and parenting in attention
deﬁcit hyperactivity disorder and comparison families. Journal of Clinical Child
and Adolescent Psychology, 35(3), 346-355.

Gershon, J. (2002). A meta-analytic review of gender differences in ADHD. Journal of
Attention Disorders, 5(3), 143-154.

Goldsmith, H. H., Gottesman, I. I., & Lemery, K. S. (1997). Epigenetic approaches to
developmental psychopathology. Development and Psychopathology, 9, 365-387.

Goodman, S. H, & Gotlib, I. H. (1999). Risk of psychopathology in the children of
depressed mothers, a developmental model for understanding mechanisms of
transmission. Psychological Review, 106(3), 458-490.

Gordon, M., Antshel, K., Faraone, S., Barkley, R., Lewandowski, L., Hudziak, J. J., et al.
(2006). Symptoms versus impairment: The case for respecting DSM-[V's
Criterion D. Journal of A ttention Disorders, 9(3 ), 465-475.

Grych, J. H, & Fincham, F. D. (1993). Children’s appraisals of marital conﬂict: Initial
investigations of cognitive-contextual ﬁ’amework. Child Development, 64, 215-
230.

Grych, J. H., Seid, M., & Fincham, F. D. (1992). Assessing marital conﬂict from the

child's perspective: The Children's Perception of Interparental Conﬂict Scale.
Child Development, 63(3), 558-572.

93

Halperin, J. M., Sharma, V., Greenblatt, E., & Schwartz, S. (1991). Assessment of the
continuous performance test: reliability and validity in a nonreferred sample.
Psychological Assessment, 3, 603-608.

Harel, E. H., & Brown, W. D. (2003). Attention deﬁcit hyperactivity disorder in
elementary school children in Rhode Island: Associated psychosocial factors and
medications used. Clinical Pediatrics, 42(6), 497-503.

Hebb, D. O. (1949). Organization of behavior: A neuropsychological theory. New York:
Wiley.

Hinshaw, S. R, Morrison, D. C., Carte, E. T., & Comsweet, C. (1987). Factorial
dimensions of the Revised Behavior Problem Checklist: Replication and
validation within a kindergarten sample. Journal of A bnormal Child Psychology,
15, 309-327.

Hinshaw, S. P., Owens, E. B., Sami, N., & Fargeon, S. (2006). Prospective follow-up of
girls with attention-deﬁcit/hyperactivity disorder into adolescence: Evidence for
continuing cross-domain impairment. Journal of Consulting and Clinical
Psychology, 74(3), 489-499.

Jacobvitz, D., Hazen, N., Curran, M., & Hitchens, K. (2004). Observations of early
triadic family interactions: Boundary disturbances in the family predict symptoms
of depression, anxiety, and attention-deﬁcit/hyperactivity disorder in middle
childhood. Development and Psychopathology 16(577-592).

Johnston, 0, & Mash, E. J. (2001). Families of children with attention-
deﬁcit/hyperactivity disorder: Review and recommendations for ﬁiture research.
Clinical Child and Family Psychology Review, 4(3), 183-207.

Kaplan, D., & George, R. (2005). A study of the power associated with testing factor
mean differences under violations of factorial invariance. Structural Equation
Modeling: A Multidisciplinary Journal, 2, 101-1 18.

Kendall, J ., Leo, M. C., Perrin, N., & Hatton, D. (2005). Modeling ADHD Child and
Family Relationships. Western Journal of Nursing Research, 27(4), 500-518.

Kim, S., Brody, G. H., & Murry, V. M. (2003). Longitudinal links between contextual
risks, parenting and youth outcomes in rural African American families. Journal
of Black Psychology, 29(4), 359-377.

Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.).
New York: Guilford.

Klorman, R., Hazel-Femandez, H., Shaywitz, S. E., Fletcher, J. M., Marchione, K. E,
Holahan, J. M., et al. (1999). Executive ﬁJnctioning deﬁcits in attention-

94

deﬁcit/hyperactivity disorder are independent of oppositional deﬁant or reding
disorder. Journal of the American Academy of Child and Adolescent Psychiatry,
38, 1148-1155.

Knopik, V. 8., Heath, A. C., Jacob, T., Slutske, W. S., Bucholz, K. K., Madden, P. A., et
al. (2006). Maternal alcohol use disorder and offspring ADHD: Disentangling
genetic and environmental effects using a children-of-twins design. Psychological
Medicine, 36, 1461-1471.

Kochanska, G., & Aksan, N. (2006). Children's conscience and self-regulation. Journal
of Personality, 74(6), 1587-1618.

Kuntsi, J ., & Stevenson, J. (2000). Hyperactivity in children: A focus on genetic research
and psychological theories. Clinical Child and Family Psychology Review, 3, 1-
24.

Lahey, B. B., Pelham, W. E., Schaughency, E. A., & et al., (1998). Dimensions and types
of attention deﬁcit disorder. Journal of the American Academy of Child and
Adolescent Psychiatry, 27, 330-335.

Lange, G., Sheerin, D., Carr, A., Dooley, B., Barton, V., Marshall, D., et al. (2005).
Family factors associated with attention deﬁcit hyperactivity disorder and
emotional disorders in children. Journal of Family 7herapy, 27(1), 76-96.

LaParo, K. M., Justice, L., Skibbe, L. E., & Pianta, R. C. (2004). Relations among
maternal, child, and demographic factors and the persistence of preschool
language impairment. American Journal of Speech and Language Pathology,
13(4), 291—303.

Lee, C. Y., Chang, Y. Y., & Lung, F. W. (2006). The marriage-related risk factors during
maternal pregnancy in children with attention-deﬁcit hyperactivity disorder. Child
Care Health Development, 32(2), 205-211.

Levy, F. (1980). The development of sustained attention (vigilance) in children: Some
normative data. Journal of Child Psychology and Psychiatry, 21(1), 77-84.

Lewis-Abney, K. (1993). Correlates of family ﬁanctioning when a child has attention
deﬁcit disorder. Issues in Comprehensive Pediatric Nursing, 16(3), 175-190.

Linnet, K. M., Dalsgaard, S., Obel, C., Wisborg, K., Henriksen, T. B., Rodriguez, A., et
al. (2003). Maternal lifestyle factors in pregnancy risk of attention deﬁcit
hyperactivity disorder and associated behaviors: Review of the current evidence.
American Journal of Psychiatry, 160, 1028- 1 040.

95

Logan, G. D. (1994). A user's guide to the stop signal paradigm. In D. Dagenbach & T.
Carr (Eds), Inhibition in language, memory, and attention (pp. 189-239). San
Diego: Academic Press.

Logan, G. D., Schachar, R., & Tannock, R. (1997). Impulsivity and inhibitory control.
Psychological Science, 8(60-64).

Loney, J ., Kramer, J ., & Milich, R. (1981). The hyperactive child grows up: Predictors of
symptoms, delinquency, and achievement at follow-up. In K. Gadow & J. Loney
(Eds), Psychosocial Aspects of Drug Treatment for Hyperactivity (pp. 381-415).
Boulder, Colorado: Westview Press.

Lynch, M. (2003). Consequences of children's exposure to community violence. Clinical
Child and Family Psychology Review, 6(4), 265-274.

Manly, J ., Miller, S. W., Heaton, R. K., Byrd, D., Reilly, J., Velasquez, R. J., et al.
(1998). The effect of African-American acculturation on neuropsychological test
performance in normal and HIV-positive individuals. Journal of the International
Neuropsychological Society, 4, 291—302.

Mannuzza, S., Klein, R. G., Bessler, A., Malloy, P., & LaPadula, M. (1993). Adult
outcome of hyperactive boys: Educational achievement, occupational rank, and
psychiatric status. Archives of General Psychiatry, 50, 565-576.

Mash, E. J ., & Johnston, C. (2005). Attention-Deﬁcit/Hyperactivity Disorder (ADHD)
and the Family: A Developmental Psychopathology Perspective. In J. L. Hudson
& R. M. Rapee (Eds), Psychopathology and the family. Oxford: Elsevier Ltd.

McCartney, K., Burchinal, M. R., Bub, K. L. (2006). Missing data: What to do with or
without them. In W. F. Overton & W. A. Collins (Eds), Monographs of the
society for research in child development: Vol. 71. Best practices in quantitative

methods for developmentalists. Boston, MA: Blackwell Publishing.

McGee, R., Williams, S., & Feehan, M. (1992). Behavioral and developmental
characteristics of aggressive, hyperactive, and aggressive-hyperactive boys.
Journal of the American Academy of Child Psychiatry, 23, 270-279.

McGrath, M. M. (2005). Early precursors of low attention and hyperactivity in a preterm
sample at age four. Issues in Comprehensive Pediatric Nursing, 28(1), 1-15.

Milberger, S., Biederman, J., Faraone, S., Guite, J., & Tsuang, M. T. (1997). Pregnancy,

delivery and infancy complications and attention deﬁcit hyperactivity disorder:
Issues of gene environment interaction. Biological Psychiatry, 41, 65-75.

96

Milich, R., Balentine, A. C., & Lynam, D. R. (2001). ADHD combined type and ADHD
predominantly inattentive type are distinct and unrelated disorders. Clinical
Psychology: Science and Practice, 8(4), 463-488.

Miller, C. J., Miller, S. R., Trampush, J., McKay, K. E., Newcorn, J. H, & Halperin, J.
M. (2006). Family and cognitive factors: modeling risk for aggression in children
with ADHD. Journal of the American Academy of Child and Adolescent
Psychiatry, 45(3), 355-363.

Minde, K., Eakin, L., Hechtman, L., Ochs, E., Bouffard, R., Greenﬁeld, B., et al. (2003).
The psychosocial functioning of children and spouses of adults with ADHD.
Journal of Child Psychology and Psychiatry. Vol, 44(4), 63 7—646.

Moffrtt, T. E. (1990). Junvenile delinquency and attention deﬁcit disorder: Boys’
developmental trajectories from age 3 to 15. Child Development, 61, 893-910.

Montiel Nava, C., Montiel Barbero, I., & Pena, J. A. (2005). Clima familiar en el
trastomo por deﬁcit de atencion-hiperactividad / Family environment in Attention
Deﬁcit-Hyperactivity Disorder. Psicologia Conductual Revista Intemacional de
Psicologia C linica de las Salud, 13(2), 297-310.

Moos, R. H., Insel, P. M., & Humphrey, B. (1974). Family work and group environment
scales manual. Palo Alto, CA: Consulting Psychologists Press, Inc.

Moos, R. H., & Moos, B. S. (1974). Manual for the Family Environment Scale. Palo
Alto, CA: Consulting Psychologists Press.

Murphy, K. R., & Barkley, R. (1996). Parents of children with attention
deﬁcit/hyperactivity disorders: Psychological and attentional impairment.
American Journal of Orthopsychiatry, 66, 93- 102.

Muthén, L.K. and Muthén, B.O. (1998-2007). Mplus User’s Guide. F iﬁh Edition. Los
Angeles, CA: Muthén & Muthén.

Nigg, J. T. (1999). The ADHD response—inhibition deﬁcit as measured by the stop task:
Replication with DSM-IV combined type, extension, and qualiﬁcation. Journal of
Abnormal Child Psychology, 27(5), 393-402.

Nigg, J. T. (2005). Neuropsychologic theory and ﬁndings in attention-
deﬁcit/hyperactivity disorder: the state of the ﬁeld and salient challenges for the
coming decade. Biological Psychiatry, 5 7(1 1), 1424-1435.

Nigg, J. T., Blaskey, L. G., Huang-Pollock, C. L., Rappley, M. D. (2002).

Neuropsychological executive ﬁinctions and DSM-IV ADHD subtypes. Journal
of the American Academy of Child and Adolescent Psychiatry, 41, 59-66.

97

Nigg, J. T., Goldsmith, H. H., & Sachek, J. (2004). Temperament and attention deﬁcit
hyperactivity disorder: The development of a multiple pathway model. Journal of
Clinical Child and Adolescent Psychology, 33(1), 42-53.

Nigg, J ., Nikolas, M., F riderici, K., Park, L., Zucker, R. (2007). Genotype and
neuropsychological response inhibition as resilience promoters for attention-
deﬁcit/hyperactivity disorder, oppositional deﬁant disorder, and conduct disorder
under conditions of psychosocial adversity. Development and Psychopathology,
19, 767-786.

Nigg, J ., Nikolas, M., Miller, T., Burt, S. A., Klump, K. L., von Eye, A. (under review).
Factor structure of the child perception of interparental conﬂict scale for studies
of youth with externalizing behavior problems. Psychological Assessment.

Nigg, J. T., Willcutt, E. G., Doyle, A. E., & Sonuga-Barke, E. J. (2005). Causal
heterogeneity in attention-deﬁcit/hyperactivity disorder: Do we need
neuropsychologically impaired subtypes? Biological Psychiatry, 57, 1224-1230.

O'Conner, T. G. (2003). Early experiences and psychological development: Conceptual
questions, empirical illustrations, and implications for intervention. Development
and Psychopathology, 15, 671-690.

Okazaki, S., & Sue, S. (1995). Methodological issues in assessment research with ethnic
minorities. Psychological Assessment, 7(3), 367-375.

Panzer, A., & Viljoen, M. (2005). Supportive neurodevelopmental evidence for ADHD
as a developmental disorder. Medical Hypotheses, 64(4), 755-758.

Parke, R. D. (2004). Development in the family. Annual Review of Psychology, 55, 365-
399.

Patterson, G. R. (1982). Coercive family process. Eugene, OR: Castalia.
Peterson, B. S., Pine, D. S., Cohen, R, & Brook, J. S. (2001). Prospective, longitudinal
study of tic, obsessive-compulsive, and attention-deﬁcit/hyperactivity disorders in

an epidemiological sample. Journal of the American Academy of Child and
Adolescent Psychiatry, 40, 685-695.

Phinney, J. S. (1992). The Multigroup Ethnic Identity Measure: A new scale for use with
diverse groups. Journal of Adolescent Research, 7(2), 156-176.

Plomin, R. (1994). Genetics and experience: The developmental interplay between nature
and nurture. Newbury Park, CA: Sage.

Plomin, R., & Ruter, M. (1998). Child development, molecular genetics, and what to do
with genes once they are found. Child Development, 69(4), 1223-1242.

98

Podolski, C. L., & Nigg, J. T. (2001). Parent stress and coping in relation to child ADI—[D

severity and associated child disruptive behavior problems. Journal of Clinical
Child Psychology, 30(4), 503-513.

Posner, M. I., & Petersen, S. E. (1990). The attention system of the human brain. Annual
Review of Neuroscience, 13, 24-42.

Preacher, K. J ., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect
effects in simple mediation models. Behavior Research Methods, Instruments, &.
Computers, 36(4), 717-731.

Pressman, L. J ., Loo, S. K., Carpenter, E. M., Asarnow, J. R, Lynn, D., McCracken, J.
T., et al. (2006). Relationship of Family Environment and Parental Psychiatric
Diagnosis to Impairment in ADHD. Journal of the American Academy of Child
and Adolescent Psychiatry, 45(3), 346-354.

Pribram, K. H., & McGuinness, D. (1975). Arousal, activation, and effort in the control
of attention. Psychological Review, 88(2), 116-149.

Puig-Antich, J ., & Ryan, N. (1986). The Schedule for Affective Disorders and
Schizophrenia for School-Age Children (Kiddie-SADS). Pittsburgh, Pennsylvania:
Western Psychiatric Institute and Clinic.

Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for research in the
general population. Applied Psychological Measurement, 1, 385—401.

Rapport, M. D., Chung, K. M., Shore, G., Denney, C. B., & Issacs, P. (2000). Upgrading
the science and technology of assessment and diagnosis: Laboratory and clinic-
based assessment of children with ADI-H). Journal of Clinical Child Psychology,
29(4), 555-568.

Rescorla, L. A., & Achenbach, T. (2004). The Achenbach System of Empirically Based
Assessment (ASEBA ) for Ages 18-90 Years. Mahwah, NJ: Lawrence Erlbaum
Associates Publishers.

Rey, J. M., Singh, M., Hung, S. F., et al. (1997). A global scale to measure the quality of
the family environment. Archives of General Psychiatry, 54, 817-822.

Rey, J. M., Walter, G., Plapp, J. M., & Denshire, E. (2000). Family environment in
attention deﬁcit hyperactivity, oppositional deﬁant and conduct disorders.
Australian and New Zealand Journal of Psychiatry. Vol, 34(3), 453-457.

Richmond, M. K., & Stocker, C. M. (2007). Changes in children’s appraisals of marital

discord from childhood through adolescence. Journal of Family Psychology, 21,
416-425.

99

Rodriguez, A., & Bohlin, G. (2005). Are maternal smoking and stress during pregnancy

related to ADHD symptoms in children? Journal of Child Psychology and
Psychiatry, 46(3), 246-254.

Rubia, K., Smith, A. B., Woolley, J., Nosarti, C., Heyman, 1., Taylor, E., et al. (2006).
Progressive increase of frontostriatal brain activation from childhood to adulthood
during event-related tasks of cognitive control. Human Brain Mapping, 27, 973-
993.

Rutter, M., Cox, A., Tupling, C., Berger, M., & Yule, W. (1975). Attainment and
adjustment in two geographical areas. I--The prevalence of psychiatric disorder.
British Journal of Psychiatry, 126, 493-509.

Rutter, M., Mofﬁtt, T. E., & Caspi, A. (2006). Gene-environment interplay and
psychopathology: multiple varieties but real effects. Journal of Child Psychology
and Psychiatry, 47, 226-261.

Rutter, M., & Sroufe, A. L. (2000). Developmental psychopathology: Concepts and
challenges. Development and Psychopathology, 12, 265-296.

Samuel, V. J., Biederman, J ., Faraone, S. V., George, P., Mick, E., Thornell, A., et al.
(1998). Clinical characteristics of attention deﬁcit hyperactivity disorder in
African American children. American Journal of Psychiatry, 155(5), 696-698.

Sargeant, J ., Oosterlaan, J ., & van der Meere, J. (1999). Information processing and
energetic factors in Attention Deﬁcit/Hyperactivity Disorder. In H. Quay & A.
Hogan (Eds), Handbook of disruptive behavior disorders. New York: Kluwer
Academic.

Satake, H., Yamashita, H., & Yoshida, K. (2004). The Family Psychosocial
Characteristics of Children with Attention-Deﬁcit Hyperactivity Disorder With or
Without Oppositional or Conduct Problems in Japan. Child Psychiatry and
Human Development, 34(3), 219-235.

Sattler, J. M. (2001). Assessment of children: Cognitive applications (4th ed.). La Mesa,
CA: Jerome M. Sattler, Publisher, Inc.

Sattler, J. M., & Dumont, R. (2004). Assessment of children: WISC-I V and WPPSI-III
supplement. La Mesa, CA: Jerome M. Sattler, Publisher, Inc.

Scahill, L., Schwab Stone, M., Merikangas, K. R., Leckman, J. F ., Zhang, H., & Kasl, S.
(1999). Psychosocial and clinical correlates of ADHD in a community sample of
school-age children. Journal of the American Academy of Child and Adolescent
Psychiatry. Vol, 38(8), 976-984.

100

Schachar, R., & Logan, G. D. (1990). Impulsivity and inhibitory control in normal
development and childhood psychopathology. Developmental Psychology, 26,
710-720.

Schafer, J. L., & Graham, J. W. (2002). Missing data: our view of the state of the art.
Psychological Methods, 7, 147-177.

Schmitz, M., Denardin, D., Silva, T. L., Pianca, T., Hutz, M. H., Faraone, S., et al.
(2006). Smoking during pregnancy and attention-deﬁcit/hyperactivity disorder,

predominantly inattentive type: A case-control study. Journal of the American
Academy of Child and Adolescent Psychiatry, 45 (1 1), 1338-1345.

Sellers, R. M., Smith, M.A., Shelton, J. N., Rowley, S. A., Chavous, T. M. (1998).
Multidimensional model of racial identity: A reconceptualization of African
American racial identity. Personality and Social Psychology Review, 2(1), 18-3 9.

Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental
studies: New procedures and recommendations. Psychological Methods, 7, 422—
445.

Sonuga-Barke, E. J. (2005). Causal models of attention-deﬁcit/hyperactivity disorder:
From common simple deﬁcits to multiple developmental pathways. Biological
Psychiatry, 57, 1231-1238.

Sonuga-Barke, E. J., Lamparelli, M., Stevenson, J ., Thompson, M., & Henry, A. (1994).
Behaviour problems and pre-school intellectual attainment: The associations of
hyperactivity and conduct problems. Journal of Child Psychology and Psychiatry,
35, 949-960.

Sonuga Barke, E. J. S., Daley, D., & Thompson, M. (2002). Does maternal ADHD
reduce the effectiveness of parent training for preschool children's ADHD?
Journal of the American Academy of Child and Adolescent Psychiatry. Vol, 41(6),
696-702.

Sroufe, A. L. (1997). Psychopathology as an outcome of development. Development and
Psychopathology, 9, 251-268.

Stawicki, J. A., Nigg, J. T., & von Eye, A. (2006). Family psychiatric history evidence on
the nosological relations of DSM-IV ADHD combined and inattentive subtypes:
new data and meta-analysis. Journal of Child Psychology and Psychiatry, 47(9),
935-945.

Stein, R. E., Siegel, M. J., & Bauman, L. J. (2006). Are children of moderately low birth

weight at increased risk for poor health? A new look at an old question.
Pediatrics, 118(1), 217-223.

101

Stephenson, M. (2000). Development and validation of the Stephenson Multigroup
Acculturation Scale (SMAS). Psychological Assessment, 12(1), 77-88.

Strine, T. W., Lesesne, C. A., Okoro, C. A., McGuire, L. C., Chapman, D. P., Balluz, L.
S., et a1. (2006). Emotional and behavioral difﬁculties and impairments in
everyday functioning among children with a history of attention-
deﬁcit/ hyperactivity disorder. Preventing Chronic Disease: Public health
research, practice, and policy, 3(2), 1-10.

Sullivan, 8., Schwartz, S. J., Prado, G., Huang, S., Pantin, H., & Szapocznik, J. (2007). A
bidimensional model of acculturation for examining differences in family
functioning and behavior problems in Hispanic immigrant adolescents. Journal of
Early Adolescence, 27, 405-430.

Tannock, R., Martinussen, R., & Frijters, J. (2000). Naming speed performance and
stimulant effects indicate effortful, semantic processing deﬁcits in attention-
deﬁcit/ hyperactivity disorder. Journal of A bnormal Child Psychology, 28, 23 7-
252.

Thapar, A., van den Bree, M., Fowler, T., Langley, K., & Whittinger, N. (2006).
Predictors of antisocial behaviour in children with attention deﬁcit hyperactivity
disorder. European Child and Adolescent Psychiatry, 15(2), 1 18-125.

Tsuang, M. T., Faraone, S., & Lyons, M. J. (1993). Identiﬁcation of the phenotype in
psychiatric genetics. European Archives of Psychiatry and Clinical Neuroscience,
243,131-142.

Tucker, D. M., Luu, P., & Denyberry, D. (2005). Love hurts: the evolution of empathic

concern through the encephalization of nociceptive capacity. Development and
Psychopathology, I 7(3 ), 699-713.

Tyson, E. H. (2004). Ethnic Differences Using Behavior Rating Scales to Assess the
Mental Health of Children: A Conceptual and Psychometric Critique. Child
Psychiatry and Human Development, 34(3), 167-201.

United States Census Bureau. (2000). State & County Quick Facts for Ingham County,

Michigan. Retrieved August 9, 2008, from http://quickfacts.censusgov/
qfd/states/26/26065.html

Vuijk, P., van Lier, P. A. C., Huizink, A. C., Verhulst, F. C., & Crijnen, A. A. M. (2006).
Prenatal smoking predicts non-responsiveness to an intervention targeting
attention-deﬁcit/hyperactivity symptoms in elementary school children. Journal
of Child Psychology and Psychiatry, 4 7(9), 891-901.

Wechsler, D. (1991). Wechsler Intelligence Scalefor Children (3rd ed). San Antonio,
TX: Psychological Corporation.

102

Wechsler, D. (1997). Wechsler Adult Intelligence Scale (3 rd ed.) San Antonio, TX:
Psychological Corporation.

Wechsler, D. (2003). Wechsler Intelligence Scale for Children (4th ed). San Antonio,
TX: Psychological Corporation.

West, J., Houghton, 8., Douglas, G., Wall, M., & Whiting, K. (1999). Levels of self-
reported depression among mothers of children wiht attention-
deﬁcit/hyperactivity disorder. Journal of Attention Disorders, 3, 135-140.

Willcutt, E. G., Doyle, A. E., Nigg, J. T., Faraone, S., & Pennington, B. F. (2005).
Validity of the executive function theory of attention-deﬁcit/hyperactivity
disorder: A meta-analytic review. Biological Psychiatry, 5 7, 1336-1346.

Williams, B. R., Ponesse, J. S., Schachar, R., Logan, G. D., & Tannock, R. (1999).

Development of inhibitory control across the life span. Developmental
Psychology, 35(1), 205-213.

103

3 1293 03063 404

H
H
H
H
H
N “|
u
T "1|
"1|