«mam. “mm... 3w.

{lirJZ-x

#4:. n
y fr.

.5... I.
it”!

1 a... 1.
is???

.7..

 

 

my,

THESIS
1

1003 -' LIBRARY
#5 vet-811 6 Michigan State
University

 

 

 

This is to certify that the

thesis entitled

A Family Psychiatric Study of
ADHD DSM—IV Subtypes

presented by

JulieAnn Stawicki

has been accepted towards fulfillment
of the requirements for

M.A. degree in Psycholggy

 

 

- /
mtg/064

Date—MLJQQZ

0-7639 MS U is an Wank: Action/Equal Opportunity Institution

 

 

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

 

MAY 2 0 gain;

 

*Vq

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDuepes-pAS

 

A Family Psychiatric Study of ADHD DSM-IV Subtypes
By

JulieAnn Stawicki

A THESIS

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

MASTER OF ARTS
Department of Psychology

2002

Abstract
A FAMILY PSYCHIATRIC STUDY OF ADHD DSM-IV SUBTYPES
By

JulieAnn Stawicki

The present study is the first to examine parent psychiatric diagnoses identified by
structured interview of each parent in relation to prospectively defined DSM-IV ADHD
subtypes in children. Information about family psychiatric history in relation to the
DSM-IV subtypes of ADHD is relevant to theories of contextual and genetic pathways
and to the issue of the validity of the subtype distinctions. Parents of children with (a)
ADHD-combined type, (b) ADHD-inattentive type, and (c) control children completed a
structured diagnostic interview (Diagnostic Interview Schedule for DSM—IV) and were
compared on rates of ADHD and other non-ADHD disorders. Results replicated past
findings in that parents of ADHD children were more likely to have ADHD themselves
(21.4%) than were parents of control children (4.7%), even when child comorbid
disorders were controlled. Also, this study found that DSM-IV ADHD subtypes in
parents were not correlated with DSM-IV ADHD subtypes in children, as would be
predicted if the subtypes were distinct disorders. Breaking new ground, gender
differences in parent prevalence of ADI-II) were also found. Specifically, girls with
ADHD were more likely to have a mother with ADHD (20.8%) compared with control
girls (0%) whereas boys with ADHD were more likely to have a father with ADHD
(22.6%) compared with control boys (0%). In addition, results suggest possible gender-

specific transmission of ADHD.

Acknowledgements
I would like to thank my advisor, Joel Nigg, for all his time and effort spent
helping me with this thesis. I would also like to thank the staff of the Child Attention
Study for all their help and encouragement in the past 3 years. Also, I thank my
committee for their comments on my thesis. Finally, I would like to thank my family and

fiiends for their support throughout my education.

iii

Table of Contents

LIST OF TABLES .................................................................... v
INTRODUCTION .................................................................... 1
Overview of ADHD ......................................................... 1
Family psychopathology studies of ADHD ............................. 13
Summary ..................................................................... 25
Current Study ............................................................... 27
METHODS ........................................................................... 31
RESULTS ............................................................................. 44
DISCUSSION ........................................................................ 57
APPENDICES ..... I .................................................................. 68
REFERENCES ....................................................................... 71

iv

LIST OF TABLES

Table 1
Family Psychiatric Studies of ADHD Since 1970 ............................... 77

Table 2
Prevalence of ADHD in parents and Relatives of Controls
Compared to Parents of Children with ADHD, with ODD/CD ................ 79

Table 3
Rates of Psychiatric Disorders in Parents of Controls
Compared to Parents of Children with ADHD and

Parents of Children with ADHD+CD .............................................. 80
Table 4

Demographics ......................................................................... 81
Table 5

Hypothesis 1: Comparing ADHD (any type) with Controls .................... 82
Table 6 .

Hypothesis 1: Comparing ADHD (any type) with Controls

(Girls and Boys) ...................................................................... 83
Table 7

Prevalence of Parent Non-ADHD Disorder in Children with

ADHD Compared with Controls ................................................... 84
Table 8 p

Hypothesis 2: Comparing ADHD with ADHD+ODD/CD ..................... 85
Table 9

Hypothesis 3: Comparing DSM-IV ADHD Subtypes ........................... 86
Table 10 .

Hypothesis 3: Comparing ADHD Subtypes (Girls and Boys) ................. 87
Table 11

Prevalence of Parent non-ADHD Disorder by Child Subtype ................. 88

INTRODUCTION
W222

Characteristics often associated with childhood Attention Deficit Hyperactivity
Disorder (ADHD) include constantly moving from task-to-task, lacking concentration for
necessary activities such as homework, inability to modify behavior to appropriate
context, and interrupting others. The syndrome has a prevalence among school-aged
children estimated to be as high as 6.3% in North America (Szatmari, 1992). Children
with ADHD often have substantial fimctional difficulties, including significant problems
in school achievement and family and peer relationships, in addition to increased
likelihood of other psychiatric disorders.

In a review, Hinshaw (1992) noted that 20% of children with attention disorder
showed significant underachievement in school compared to same-aged peers. He noted
that they often score below normal IQ, and often have language and speech difficulties.
When their academic achievement is related to their performance on IQ tests, nearly half
of ADHD children have lower-than-expected achievement in reading, arithmetic, and
spelling (Barkley, DuPaul, & McMurray, 1990).

ADHD children also have marked difficulties in social conduct and adjustment.
Their impulsive and hyperactive behavior interferes with interactions with parents,
siblings, teachers, and peers (Barkley, 1998). Mothers of hyperactive children rate them
as more non-compliant and are observed to be more negative in their interactions with
their children compared to mothers of controls (Gomez & Sanson, 1993). Hyperactive
children have more daily serious quarrels with their siblings compared to controls
(Taylor, Sandberg, Thorley, & Giles, 1991). Teachers rate hyperactive children as

having severe problems mixing with same-aged peers (Taylor et al., 1991). They are

more often described as having problems in co-operating with others, following in the
rules of games and, subsequently, are more often the victims of teasing by other children
(Taylor et al., 1991).

ADHD ofien co-occurs with other disorders, including depression, anxiety, and
learning disabilities (Biederman, Newcorn, & Sprich, 1991). Most salient is that many
children with ADHD also exhibit other disruptive behavior disorders, specifically,
Oppositional defiant disorder (ODD) and Conduct Disorder (CD) (APA, 1994). Both of
these syndromes are characterized by disobedience and opposition to authority figures,
and angry and resentful behaviors. ODD is often a precursor to CD. Almost all children
with CD previously met criteria for ODD (Lahey & Loeber, 1996). Children with CD
also demonstrate a pattern of behaviors that violate the basic rights of others or violate
age-appropriate nouns. Some features of CD include aggression to people and animals,
destruction of property, and serious violations of rules (e. g., truancy and running away
from home overnight) (APA, 1994). Unsurprisingly, children with ADHD who have
comorbid ODD or CD experience even greater problems in social relationships with both
family (Gomez & Sanson, 1993) and peers (Taylor et al., 1992) than children with
ADHD alone.

A majority of children with ADHD often experience persistent impairing
problems into adolescence and adulthood, with poor long-term outcomes. Approximately
50-80% of children continue to display impairing inattentive or hyperactive symptoms
into adolescence (Barkley, 1998). These adolescents are more likely to be in more
automobile accidents and to be injured and killed in accidents than same-aged peers

(Barkley, et al. 1993). They are also more likely to be suspended or expelled from school

(Barkley, 1998).

Due to the significant social, academic, and family problems associated with
ADHD, its chronicity, and frequent poor long-term outcomes, ADHD has garnered much
recent attention in the medical and academic communities, as well as in the popular
medias. In fact, ADHD has become one of the most extensively studied childhood
psychiatric disorders (Cantwell, 1997). In short, it is a syndrome of substantial social and
scientific concern. Whereas causal pathways to ADHD are incompletely understood, the
field is particularly in need of information about the family environment of children with
ADHD. Moreover, the validity of its current definition requires firrther investigation.

The present study attempts to shed light on the validity of the current definition of
ADHD along with taking a first step in providing a more complete understanding of the
casual pathways of ADHD, particularly in the area of family context. This study
examines the prevalence of psychopathology in the parents of children with ADHD.
Family psychiatric studies are a first step in examining the environment of children with
ADHD and contribute in establishing the validity of a disorder and resolving
heterogeneity. Heterogeneity and changes in definition within ADHD remain major
factors impeding interpretation of past studies and necessitating new studies of family
environment and family psychiatric history, therefore, important background issues in the
definition of ADHD must first be considered.

W

Heterogeneity is an issue from several perspectives: historical differences in

definition of ADHD; subtypes defined in DSM-IV; gender; family history; and

psychiatric comorbidity. Each is briefly explained.

Recent History of Definitions of ADHD. Past studies of family psychiatric history
have used varying definitions of ADHD depending on the edition of the QS_M current at
the time of the study. The earliest editions, D_SM—_I (APA, 1952) and M (APA,
1967) featured a prototypical formulation of Hyperactive Child Syndrome. The diagnosis
was based on evidence of hyperactivity and distractibility. However, in the earliest
editions of the DSM, hyperactive child syndrome was plagued with the same poor
reliability as nearly all the other prototypes. This provoked a return to the Kraeplinian
medical model in the DSM-III (Appendix A), which sought to offer a purely operational
approach and utilized a polythetic symptom list to operationalize diagnostic criteria. The
new criteria were intended to reflect directly observable behaviors. That modification in
DSM-III (APA, 1980) greatly improved the reliability of the diagnoses, albeit with
unknown effect on validity. The syndrome was renamed Attention Deficit Disorder and
three symptom domains were identified: inattention, hyperactivity, and impulsivity. Two
subtypes were defined for the first time. A diagnosis of attention deficit disorder [a] with
or [b] without hyperactivity depended on the presence or absence of hyperactivity.
However it is notable that children had to be both inattentive and impulsive to be
diagnosed with either of the two subtypes.

Early behavioral studies of Attention Deficit disorder subtypes based on DSM-III

 

criteria therefore examined children with inattentive and impulsive symptoms and then
divided the groups based on the presence or absence of a specified number of hyperactive
symptoms. These studies found characteristic behavioral differences between children
with hyperactivity and those without hyperactivity, lending support to separating these

two subtypes of ADHD (or ADD as it was then called). For example, children with

Attention Deficit with hyperactivity were more disinhibited, impulsive, disruptive, and
more socially inappropriate, whereas Attention Deficit children without hyperactivity
were more unmotivated, “day-dreamy,” and apathetic (Barkley, DuPaul, & McMurray,
1990). Children with hyperactivity and children without hyperactivity also differed with
regard to the pattern of comorbid problems exhibited. Children with attention problems
and hyperactivity, as defined by the DSM-III criteria, had more instances of comorbid
externalizing behavior disorders such as Conduct Disorder and Oppositional Defiant
Disorder compared to purely inattentive children (Barkley, DuPaul, & McMurray, 1990).
However, although family studies were conducted during that era, as reviewed later, no
family studies compared the DSM-III subtypes in relation to frequency of parent
psychiatric disorders.

In the DSM-III-Revised (APA, 1987), a diagnosis of attention deficit disorder was
based on the total number of symptoms of hyperactivity, impulsivity, and inattention,
regardless of which symptom area the child demonstrated most. Subtypes were not
operationally defined. As a result, research on the subtypes slowed down substantially.
Yet new data continued to support the validity of subtypes. They were therefore

reinstated in revised form in the fourth edition of the DSM (APA, 1994).

ADI-II) Subtypes in the DSM-IV. Supported by extensive field trial data utilizing
factor analytic and cluster analytic methods, the most recent version, DSM-IV (APA,
1994), defined three ADHD subtypes, based on two symptom domains: (1) inattention-
disorganization and (2) hyperactivity-impulsivity. Children are classified into one of
three Subtypes according to the cluster of symptoms they exhibit. Thus, children can be

classified as Primarily Inattentive subtype, Primarily Hyperactive subtype, or Combined

subtype, the latter meaning they exceeded thresholds for symptoms of both inattention
and hyperactivity. Notably, the inattentive type children must be below threshold on
hyperactivity/impulsivity and vice versa. Although early studies of children using DSM-
III criteria suggested fundamental differences between hyperactive and non-hyperactive
inattentive children, few studies have examined the validity of ADI-II) subtypes using the
newer and revised DSM-IV criteria. None have done so with family psychiatric history of
prospectively defined DSM-IV child subtypes, as detailed later. There is considerable but
far fi'om perfect agreement between DSM-1]] and DSM-IV criteria overall; approximately
79% of children diagnosed as ADHD with DSM-IV criteria would have been diagnosed
as ADHD by DSM-1H criteria (Lahey et al., 1994). However, the DSM-III and DSM-IV
sum do not correspond as highly. There is 73% agreement between the diagnosis of
DSM-III ADHD with hyperactivity and DSM-IV-Combined subtype, but only a 43%
agreement between DSM-III ADI-II) without hyperactivity and DSM-IV-Inattentive
subtype (Lahey et al., 1994). The majority of children diagnosed with DSM-III subtypes
are also diagnosed with the corresponding DSM-IV subtype, but the DSM-IV Primarily
Inattentive type identifies 56% more children than the DSM-III without hyperactive
subtype identifies (Lahey et al., 1994). This underscores the need for further validation
studies to evaluate the relation between the DSM-IV subtypes.

However, initial behavioral studies of children with the DSM-IV subtypes
confirmed the expected behavioral differences between them. These included an earlier
age-at-onset for the Combined subtype; gender differences in prevalence rates, with more
males being diagnosed as Combined type than Inattentive type; and different rates of

comorbid disorders associated with each subtype (reviewed by Milich et al., 2001).

Thus, behavioral studies have generally supported the validity of the DSM-IV subtype
distinctions.

However, studies examining neuropsychological features of ADHD subtypes
have yielded mixed results. These studies have compared the Combined type and the
Primarily Inattentive type‘. F araone et al. (1998) found no differences between subtypes
on IQ and academic measures: Houghton et al. (1999) concluded that the combined type
had larger deficits on several neuropsychological tests. Nigg et al. (2001) found that on
measures of response speed, planning, and interference control, both subtypes of ADHD
children demonstrated significant deficits compared to the controls, but did not differ
from one another. On a measure of behavior inhibition, differences between subtypes
were dependent on gender effects: boys with ADHD-combined type performed worse
than boys with ADHD-inattentive type, while there were no differences between girls
with ADHD-combined type and girls with ADHD-inattentive type. These mixed findings
among and within studies highlight the ambiguity of the validity of the current distinction
between the subtypes, and underscore the need for additional data to shed light on the
extent to which these ADHD subtypes are closely related or distinct.

Thus, although some behavioral differences between subtypes are accepted, it
remains unknown whether these differences reflect distinct etiologies or whether they
reflect different expressions of the same underlying processes or disorder. Family studies
can inform that question. Yet the newest definitions of ADHD subtypes have not
eliminated the debate over the heterogeneity of children diagnosed with ADHD
(Hinshaw, 2001; Milich, Balentine, & Lynam, 2001), because several other types of

heterogeneity remain to be evaluated.

Famin histogz suMs. Another view of heterogeneity concerns differences in
whether children come from families in which relatives, particularly parents, have ADHD
as well. Originally, August and Stewart (1983) defined a different set of ADHD subtypes
based on such family history of ADHD: Family History-positive (FH+) meant that one
biological parent had a history of ADHD, and Family History-negative (FH-) meant that
neither biological parent had ADHD. They found that the FH+ children, as well as their
siblings, had more conduct problems. The FH- group was more likely to have academic
problems and learning disorders.

More recently Seidman and colleagues (1995) examined neuropsychological
performance in children with ADHD FH+ versus children with ADHD FH-. Both ADHD
groups were impaired on tests of neuropsychological functions compared to control
children, but the FH+ group was more impaired than the FH- group. They also found,
however, that child psychiatric comorbidity influenced neuropsychological functioning in
addition to family history. Therefore, comorbidity is a source of heterogeneity for both
the family history subtypes, as well as the DSM-IV subtypes.

Comorbidgy' disorders in children with ADHD. In studies of ADHD using DSM-
IH defined subtypes, different patterns of comorbidity were found depending on the
presence or absence of child hyperactivity. Barkley, DuPaul, and McMurray (1990)
found that children with Attention Deficit without Hyperactivity were more likely to have
Major Depressive Disorder compared not only to controls but also compared to those
children with Attention Deficit with Hyperactivity. Similar patterns of associated I
comorbidity were also found with DSM-III subtypes in relation to learning disorder (LD).

As noted earlier, learning disorders are also commonly diagnosed in children with

ADHD. In a study using the DSM-III definitions of ADHD, children with Attention
Deficit Disorder without Hyperactivity were more likely to be diagnosed with LD
(Barkley et al. 1990). However more recent studies of differences in LD comorbidity
rates for the DSM-IV subtypes have found mixed results. Some studies using the DSM-
IV definitions have found similar rates of LD for each ADHD subtype (see Barkley, 1998
for a review).

However, as noted earlier, the most common comorbid disorders diagnosed with
ADHD are other externalizing behavior disorders; rates for these comorbid disorders
differ according to ADHD subtype. Using the DSM-HI definitions of ADHD, children
with Attention Deficit with Hyperactivity were more likely to also be diagnosed with CD
or ODD compared to those children with Attention Deficit without Hyperactivity
(Barkley, DuPaul, & McMurray, 1990).

This pattern has held up in the DSM-IV ADHD subtypes. For example, Lalonde,
Turgay, and Hudson (1998) investigated the comorbidity patterns of the three ADHD
subtypes with both CD and ODD. The majority of the sample (N = 100) consisted of
children with ADHD combined type, (78%); inattentive type accounted for 15%; and
hyperactive-impulsive type accounted for the remaining 7%. Of the children with
combined type, 85% also had comorbid ODD and 8% had comorbid CD. Of those with
the primarily inattentive type, 33% also were diagnosed with ODD, but none had CD. Of
those children with the primarily hyperactive impulsive type, 100% had comorbid ODD
and 57% also had comorbid CD. Across subtypes, 78% of the sample had ODD, 10%

had CD. Such results suggest that controlling for such child comorbidity may be

essential when interpreting family history data. Unfortunately, very few family studies of
ADHD have controlled for such comorbidity.

_G_grder Differences, Another source of heterogeneity that has been almost
completely overlooked in the area of family prevalence studies is potential gender
differences in subtype expression and in ADHD expression generally. Although boys are
more frequently diagnosed with ADHD, with gender ratios of 6:1 (Barkley, 1998), many
girls are affected. In addition, boys and girls apparently differ in the expression of the
disorder. Girls with ADHD typically are more socially withdrawn and have more
internalizing symptoms than boys with ADHD (Breen, 1989). Boys and girls have
different ADHI) subtype prevalence rates as well as different symptom manifestation;
boys are more likely to be diagnosed with ADHD-Combined subtype, while girls are
more likely to be diagnosed with ADHD-Inattentive subtype (Lahey et al., 1994). Also,
boys are up to nine times more likely to be referred to a clinic than are girls (Barkley,
1998). Therefore, the use of clinic samples, as is the case for most studies of ADHD,
may be misrepresenting differences between boys and girls. Accordingly, applying
results found with primarily male samples to female samples may be overlooking
important gender effects. As reviewed later, virtually no family psychiatric studies of
ADHD have looked at results for girls.

Summgy. Problems with heterogeneity, along with the evolving definition of
ADHD, compound the difficulties in establishing an etiological pathway for ADHD. It is
important that these issues be considered in family psychiatric studies. The present study
therefore focused on DSM-IV ADHD subtypes, and included analyses of psychiatric

comorbidity, family history subtypes, and gender effects in children and parents.

10

Consideration of developmental context in terms of family history can address possible
causal factors as well as inform the validity of the current definition of ADHD subtypes.
Why a Farm Prevalence Study?

Despite the extensive literature on ADHD, family studies of ADHD are relatively
few (Cantwell, 1997). Yet family studies, especially family psychiatric studies, are an
important first step in examining the environment of children with ADHD. In addition to
providing clues to the developmental environment, family psychiatric studies also
become relevant in their contribution to establishing the validity of a disorder and
resolving heterogeneity.

The family environment is undoubtedly one of the most important factors in the
development of a child. A substantial literature establishes the important role of family
environment on children’s development and psychopathology. For example, problems in
the family ecology, such as low socioeconomic status (SES), marital discord, and family
conflict are linked with psychopathology in children (Biederman et al., 1995). Frick
(1994) reviewed dysfunction in families in relation to childhood ADHD. He concluded
that ADHD was more prevalent in families where there was high marital conflict, one
parent had antisocial or aggressive behavior, and one or both parents had a history of
psychiatric disorder. Biederman et al. (1995) found that parental mental disorder was the
most predictive of child problems in ADHD families.

Parental psychopathology in particular has ramifications throughout the family
environment. For example, parental psychiatric problems may degrade parenting ability,
rendering parents less consistent, warm, or attentive to children (Barkley, 1998). SES

may be lowered by parent inability to perform at work, a fiequent problem for adults with

11

ADHD (Barkley, 1998). Marital relationships may also be disrupted by parent psychiatric
disorder. As these multiple examples suggest, risk factors may tend to cluster directly in
families. This clustering, in turn, may be rooted in parent psychopathology in many
cases. Thus, clarifying whether ADHI) subtypes in DSM-IV differ in rates of parent
psychiatric problems is an important first step in evaluating possible contextual
differences in their development.

At the same time, transmission of psychiatric problems in families can provide
clues to genetic transmission of risk for a disorder such as ADHD. Whereas the
heritability of ADHD as an undifferentiated syndrome has been established in family,
twin, and adoption studies (Sherman, Iacono, & McGue, 1997) whether DSM-IV
subtypes transmit differently in families is largely unknown.

Overall parent psychopathological profiles can provide clues to possible
etiological pathways and therefore may inform validity of subtype distinctions in ADHD.
In psychiatric syndromes, construct validity reflects the extent to which the DSM-IV
criteria capture etiological or developmental pathways to disturbance in adjustment.
Family psychiatric history is a frequently recognized tool for obtaining clues to such
pathways. Many different approaches have been proposed for illuminating the validity of
medical as well as psychiatric syndromes (e. g., Cantwell & Baker, 1988; F eighner et al.,
1972; for a further discussion, see Appendix A), but family psychiatric history is one
reasonable component of all such efforts.

Thusa family psychiatric study can clarify whether particular disorders tend to
cluster together or whether the disorder under study is common in relatives. Such data

can provide clues as to possible shared etiologies. Despite all these advantages, adequate

12

family psychiatric studies of ADHD subtypes are in short supply. The few family studies
of ADHD nearly all used earlier definitions of ADHD, and most are Complicated by the
failure to control for other disorders that frequently co-occur with ADHD (Tannock,
1998). Taken together, all of the omissions in past research (old definitions of ADHD,
failure to control of comorbid disorders, and inadequate representation of girls) paint an
incomplete picture of the families of children with ADHD. Nevertheless, a carefirl
review of that literature is in order.

WW

Research in the area of family psychiatric history of ADHD children is relatively
new; the first studies were done in the early 1970’s, just after reliability was established
for adult diagnoses. Since that time, only about a dozen studies have examined
psychopathology in families of children with ADHD. Table I summarizes the studies to
be discussed. As the table illustrates, nearly all studies used pre-DSM-IV definitions,
focused more on boys, and failed to consider subtypes of ADHD. These studies are next
reviewed in more detail.

ADHD in families. Early studies mainly established a pattern of family
transmission of ADHD in families of children with ADHD. Cantwell (1972) found
higher rate of hyperactivity in the parents of children with Hyperactive Child Syndrome
than in the parents of controls. Cantwell also found higher prevalence rates of
hyperactivity in second-degree relatives of hyperactive children (12 % of all male
relatives, 6.3% of all relatives) compared to the relatives of controls (0.6%). In a review,
Cantwell (1975) concluded hyperactivity was more prevalent in families of children with

the disorder than controls; this was true for both first-degree relatives and extended

l3

family members. Later studies of ADHD in families have replicated these findings.

Other disorders in ADHD families. In addition to increased rates of hyperactivity

 

in the families of children with hyperactivity, early family studies also examined the rates
of other disorders in the parents of children with hyperactivity. Morrison and Stewart
(1971) compared the family psychiatric histories of children admitted to a hospital for
behavioral problems with children who were admitted for medical procedures, such as an
appendectomy. Parents of the hyperactive children had more diagnoses of alcoholism,
sociopathy, and hysteria than the parents of controls. They also were more likely to
report hyperactive symptoms during their own childhood, replicating Cantwell (1972).
The results suggested a correlation between the presence of other psychiatric diagnoses in
first-degree relatives and the presence of hyperactivity in children.

Later studies of familial transmission of ADHD and other psychiatric diagnoses
have supported Morrison and Stewart’s (1971) initial findings that parental
psychopathology was associated with child ADHD. Biederman et al. (1995) found that
chronic conflict, decreased family cohesion, and exposure to parental psychopathology,
particularly maternal psychopathology, were more common in ADHD families than in
control families. Children were diagnosed with ADHD using Achenbach’s Child
Behavior Checklist. A maternal self-report of their own history of psychiatric disorders
was correlated with externalizing and internalizing behaviors in their children. Maternal
psychiatric disorders were not reported in terms of individual disorders; rather only the
total number of diagnoses was analyzed. This total was greater in the parents of children
with ADHD than the parents of normal controls. The authors concluded that exposure to

family conflict and maternal psychiatric symptoms may index important environmental

14

adversity indicators in ADHD families. Although that study, like earlier studies,
supported a link between parental psychopathology and child ADHD using a more recent
description of ADHD, it shared key weaknesses with earlier studies: comorbid disorders
in children were not controlled, and ADHD subtypes were not defined.

Thus a critical question that arose from these studies concerned whether the
elevated parental hyperactivity and other psychiatric diagnoses in families of ADHD
children were specifically linked with child hyperactivity, or rather were linked with
general maladjustment in children. To address that issue, Stewart et al. (1980) examined
the rates of parental psychiatric diagnoses in families of children with hyperactivity
compared to the families of children with other psychiatric diagnoses. Children who were
admitted to psychiatric clinics with unsocialized, aggressive behavior, including
hyperactivity and conduct disorders, were more likely to have fathers with antisocial
personality disorder or alcoholism and mothers who were neurotic than children admitted
for other psychological disorders such as depression. Importantly, when aggressive child
behavior was statistically controlled, the difference in parental diagnoses between the
families of the hyperactive children versus the children with other diagnoses was no
longer significant. The rates of parental disorder in the families of children with
hyperactivity without aggression were the same as in the parents of children with
depression and anxiety. Hyperactivity in parents was not examined. Therefore Stewart
and colleagues concluded that the association of antisocial and neurotic disorders in
parents with hyperactivity in their children was not specific to child hyperactivity.

Instead, increased parental psychiatric symptoms of Antisocial Personality Disorder,

15

alcoholism, and neuroticism were related to the presence of aggression in the children
rather than the presence of hyperactivity per se.

However, Stewart et al.’s (1980) conclusions may not have been entirely
warranted. From studies done in the DSM-III era, it was known that the presence of
hyperactivity was associated with aggression (Hinshaw, 1987). Subsequently it has
remained clear that hyperactivity commonly co-occurs with aggressive behavior
disorders such as CD and ODD (Tannock, 1998). Crucially, Stewart et al.’s definition of
aggression was very broad and included many hyperactive behaviors. Further, their
diagnosis of “hyperactivity” was most similar to a diagnosis made with DSM-III-R:
either hyperactivity or inattentiveness could be sufficient for a diagnosis. As a result,
separating hyperactive boys by the presence or absence of aggression, as Stewart et al.
did, may have simply created a group of hyperactive and a group of inattentive children
by today’s definition. Thus, it is unclear whether parental diagnoses were unrelated to
child hyperactivity or rather unrelated to child inattentiveness. Therefore, it could be that
children with ADHD combined type are more likely to have parents with psychiatric
disorders; alternatively it could be true that parental psychopathology is simply correlated
with the co-occurrence of CD/ODD in ADHD children. Fortunately, other studies have
begun to address this issue more carefully.

Family studies controllingformally for the presence of CD and Aggression.
Implied in the preceding section, understanding the specifically of the association
between parental disorder and hyperactivity in children has been complicated by the high
rates of aggressive and disruptive behavior problems in ADHD children. It therefore

becomes important to consider the relationship between what are today called ODD and

16

CD in children and parental psychiatric disorder and whether these comorbid syndromes
“explain” association of child ADHD with parental disorders other than with ADHD.

As a prelude, it has been shown that child ODD and CD also have characteristic
familial psychiatric correlates. Schacher and Wachsmuth (1990) found that parent
psychopathology was more prevalent in CD and ODD children compared to normal
control families. Frick et al. (1992) found higher rates of paternal and maternal antisocial
personality disorder and substance abuse in families of children with CD and ODD than
in the control families (ADHD was not evaluated). The results point to the possibility that
ODD and CD, ,if not controlled for, could be responsible for observed increased rates of
non-ADHD psychopathology in families of ADHD children in the early studies.

Lahey, Piacentini, McBurnett, Stone, Hartdagen, and Hynd (1988) assessed
psychopathology in parents of children with either ADHD or CD. All children in the
study were outpatients referred to a local Psychological Clinic. Psychiatric diagnoses in
the children were obtained through a semi-structured interview with one custodial
biological parent, usually the mother, according to DSM-III criteria. Information for the
DSM-III diagnoses of both parents was obtained from a structured diagnostic interview
(SADS; Spitzer & Endicott, 197 8) of the mother. Parents of children with CD alone were
more likely than controls to exhibit antisocial personality disorder and substance abuse.
ADHD in children was not associated with any disorder in parents. However, children
with the comorbid diagnosis of ADHD+CD had fathers with elevated rates of aggression,
arrest, and imprisonment compared to the fathers of children with ADI-II) alone or fathers
of children with CD alone. This appeared to suggest that ADI-ID alone was not

associated with an_y disorder in the parents. Rather it was the presence of CD that was

17

correlated with parental disorder and the combination of the CD and ADHD in children
that predict the most associated non-ADHD disorders in fathers.

However that study had two crucial weaknesses. First, ADHD in the parents was
not assessed. Second, child ADHD and CD were associated with the rate of disorder in
the father; yet, diagnoses for the child and both parents were obtained only from the
mother report. Mothers could be biased in their reporting of symptoms in the father or
the mothers of children with conduct problems might be more attuned to the behaviors in
their partners.

Frick and colleagues (1991) also examined how controlling for the presence of
child CD affects the rate of childhood behavior problems in the relatives of children with
ADHD. Children were 96 boys recruited at the Western Psychiatric Institute. Each boy,
his parent, and his teacher wereall interviewed using the Diagnostic Interview Schedule
for Children (DISC; Costello, Edelbrock, Kalas, & Dulcan, 1984) based on DSM-III-R
criteria. Diagnoses for parents and second-degree relatives were made using family
history questionnaires completed by one of the biological parents, most often the mother
of the boy. Boys with ADHD were significantly more likely to have a mother, father,
one relative, or two relatives who exhibited ADHD symptoms during childhood. Fathers
of boys with CD were more likely to have used substances (tobacco, alcohol, or drugs)
before the age of 15 or to have had CD themselves. Boys with comorbid ADHD+CD had
relatives with elevated levels of hyperactivity similar to the ADHD alone group.
Relatives of boys with ADHD plus CD also had high rates of CD and substance use
similar to the parents of boys with CD alone. Frick et al. did not find that the children

ADHD plus CD had higher ratings of hyperactivity in the children than ADHD alone,

18

suggesting that the ADHD plus CD group was not simply a more severe ADHD group.
Therefore, ADHD in parents was associated with ADHD in children and parental
substance abuse was associated with CD in children. The higher number of disorders in
parents of children with ADHD plus CD was due to the co-occurrence of the two
disorders in the child; CD and ADHD in children each predicted separate psychiatric
disorders in parents.

These studies taken together suggest two different possibilities forassociation
between ADHD comorbidity and parental disorder. One possibility as presented by
Lahey et al. (1988) is that child ADHD alone is not associated with non-ADHD disorders
in the parents of these children, but when ADHD and CD co-occur, the severity of non-
ADHD parent disorders is qualitatively worse than would be suggested by either the
presence of ADHD or CD alone. Because Lahey et al. did not assess for ADHD in
parents of the children, it remains unclear whether the combination of ADHD + CD in
children would also predict a higher rate of m in their parents. This differs from the
results presented by Frick et al. (1991), which suggest that it is a diagnosis of ADHD and
CD in children that predict distinct disorders in the parents. The associated disorders in
parents of children with both ADHD and CD were merely the additive afi’ect of what
would be predicted by each of the child disorders alone. The relationship between parent
psychopathology and childhood ADHD and childhood CD is not clear; the presence of
ADHD and CD could represent the simple co—occurrence of 2 disorders, or ADHD plus
CD could be a distinct syndrome with unique correlates. As a result of this ambiguity, it
is essential to control for comorbidity with CD when interpreting the relationship

between child ADHD and parent disorder.

19

Kuhne, Schachar, and Tannock (1997) examined the effects of comorbid child
CD and ODD and severity of child ADHD symptoms and aggression, in relation to
maternal reports of their own experience with psychiatric symptoms, as assessed using
the Symptom Checklist. Child diagnoses were made using questionnaires completed by
parents and teachers based on DSM-IH-R criteria. ADHD symptoms were more severe
in both ADHD+ODD and ADHD+CD children compared to ADHD only children.
Mothers rated their children with ADHD+CD as more aggressive than the mothers of
either children with ADHD or children with ADHD+ODD had rated them. Mothers of
the ADHD+CD children had elevated levels of self-report psychopathology in areas of
interpersonal sensitivity, anxiety, and hostility compared to mothers in the ADHD-only
group. These findings are generally consistent with earlier findings by Lahey et al.
(1988) and Frick et al. (1991). _

F araone et al. (1997) examined rates of ADHD, ODD, CD, and Antisocial
Personality disorder in the parents, siblings, and other relatives of controls and of
children with either ADI-ID alone or ADHD and comorbid externalizing disorder. They
found that relatives of all three child groups (ADHD+CD, ADHD+ODD, and ADHD
alone) were all at greater risk for both ADHD and ODD compared to controls. However
only the relatives of the ADHD+CD group had an increased risk of CD and antisocial
personality disorder. The relatives of the ADHD+ODD and ADHD only did not difi‘er in
rates of ADHD, ODD, CD, or antisocial personality disorder.

Nigg and Hinshaw (1998) looked more specifically at the presence of ADHD
along with other psychiatric diagnoses and their associated features in parents of children

with ADHD. The participants in the experimental groups were the parents of boys

20

enrolled in a summer research program; boys were diagnosed with ADHD (defined by
DSM-III-R). The control group was parents of boys without ADHD. Both groups were
recruited from the community. All children were also assessed for comorbid CD and
ODD using parent rating scales data. Parent ADHD diagnosis was made using direct the
Wender Utah questionnaire (W ender, 1985), other non-ADHD diagnoses were
determined using the Quick Diagnostic Interview Schedule (QDIS). The children were
then initially placed in one of three groups: comparison, ADHD, and ADHD+ODD/CD.
The mothers in both ADI-II) groups had higher rates of recent maternal depression: 25%
of the mothers of the ADHD group and 39% of mothers of the ADHD+ODD/CD group
compared to 0% of mothers of the comparison group. When the parents of the boys with
ADHD+CD were examined apart from ADHD+ODD, the mothers of the boys with
ADHD+CD had higher rates of childhood ADHD than all other groups, including the
mothers of ADHD+ODD children. Fathers of the ADHD group (41%) were more likely
than control fathers (0%) to have a childhood history of ADHD, based on the Wender
rating scales. Furthermore, the fathers of the boys with ADHD+CD were more likely to
have antisocial personality disorder as compared to all other fathers.

DSM-IV familial subtypes? In a study to determine whether or not the subtypes of
ADHD in families were the same as those found in children, Faraone, Biederman, and
Friedman (2000a) examined ADHD subtypes in relatives of boys with ADHD. Children
for the study were recruited from referrals to a psychiatric clinic. ADHD diagnoses were
initially made using DSM-III-R criteria. The investigators estimated DSM-IV subtypes
retrospectively using a proxy procedure to estimates DSM-IV subtype from DSM-III-R

data. The relatives of boys with ADHD were more likely to have had ADHD themselves,

'21

replicating previous studies. However, relatives were not more likely to have the same
subtype as their child. In other words, relatives of boys with combined-type ADI-1]) were
not more likely to have combined-type themselves. The authors concluded that the child
subtypes of ADHD are not related to familial or genetic differences, but rather are
different manifestations of one disorder that may vary according to environmental
influences. The latter reflected potentially important evidence in opposition to the
conclusions of Milich et al. (2001), who argued that the inattentive type is a distinct
disorder.

However the study by Faraone et al. (20003) had key limitations. Most important,
the study was not done with actual prospectively assessed DSM-IV criteria, but rather by
approximating DSM-IV subtypes from the DSM-III-R criterion set. Secondly, the
children were fiom a clinic-recruited sample, which may make results vulnerable to
referral bias (Goodman et al., 1997). Third, parents were not independently examined;
rather all the relatives were combined in the analyses. Finally, all diagnoses in the
relatives were determined from a family history questionnaire filled out by the mother of
the participants, rather than by direct structured interview of each parents.

A similar study was also completed with the families of girls. Faraone and
colleagues (2000b) looked at the rates of ADI-ID subtypes in relatives of girls diagnosed
with ADHD. The girls for this study were recruited from the same psychiatric clinic as
reported in Faraone et al. (2000a). As was the case in the previous study, higher rates of
ADI-II) were found in relatives of girls with ADHD; but the relatives of girls with ADHD
were _no_t more likely to be diagnosed with the same subtype as diagnosed in the child.

One advantage of this study is that diagnoses in the relatives of the girls were assessed

22

using semi-structured clinical interviews of each relative. However, this study also
suffered from the same key limitations as Faraone et al. (2000a). Specifically, (1) girls
with ADHD were diagnosed using DSM-III-R, and DSM-IV subtypes were
approximated, (2) the sample was clinic-referred, and (3) results of analyses were
reported for all relatives, but not for mothers and fathers separately.

In addition, the authors concluded that their findings demonstrate that familial
transmission of ADHD in girls is similar to the families of boys, and therefore, previous
studies using male samples generalizes to girls with ADHD. This broad conclusion was
probably premature for several reasons, including the methodological limitations already
noted. Even by the authors’ own report, this is the only known study of familial
transmission of ADHD in girls with ADHD (F araone et al., 2000b); the extent of
generalizability between results-of studies using male samples to female populations
warrants replication and extension.

Cgmment on Past Studies

These studies taken together suggest an association between child ADHD and
parental psychopathology. Specifically the parents of children with ADI-ID were more
likely to exhibit ADHD themselves. Along with having ADI-ID, mothers of children with
ADHD were also more likely to have been diagnosed with depression. Fathers of ADHD
children are more likely to have antisocial personality disorder. However, the
relationship between child ADHD and parental non-ADHD psychopathology was not
straightforward; parental disorder was also associated with other disruptive behavior
disorders. Parents of children with ADHD plus ODD or CD were more likely to exhibit

more antisocial and hyperactive symptoms than the parents of children with ADHD

23

alone. Mothers of children with ADHD were more likely to be depressed than the
mothers of control children. However mothers of children with ADHD plus CD were
even more likely to be depressed than mothers of children with ADI-1D alone.

Taking these studies together, it appears that the co-occurrence of ADHD plus CD
in children could suggest a unique, severe combination with different parental correlates
greater than simply the sum of these found in either CD/ODD alone or ADHD alone.
Stewart et al.’s (1980) experimental group (labeled as hyperactive and aggressive) could
have represented this severe sub-group.

All of the previous studies were done using previous editions of the DSM, prior to
the DSM-IV; many were done with the DSM-HI-R Only two studies even considered
the subtypes found in the DSM-IV (none even examined DSM-HI subtypes), yet the
DSM-IV subtypes differentially co-vary with comorbid disorders. The relationship
between parental psychopathology and DSM-IV ADHD subtypes therefore becomes an
important question, for which insufficient data exist.

What we know about family context and validity of ADHD subtypes is
incomplete. Family psychiatric studies provide one type of data that can be informative.
Past studies found that children with ADHD were more likely to have parents with
ADHD. In general, although raw proportions were not reported in all studies, rates of
ADHD in parents were 3 to 8 times higher for ADHD than control children. Table 2
summarizes these data. It indicates that there was considerable variability. However
these rates suggested guidelines for what to anticipate in the current DSM-IV study.

A crucial point is that no prior study examined subtypes of ADHD prospectively

assessed using the DSM-IV. This gap in the literature leaves questions of both family

24

environmental effects and current subtype validity unanswered. In addition, several other
limitations exist within the relatively small family psychiatric literature. In addition to
outdated diagnostic criteria, many of the studies used samples that were 100% male. In
addition to contributing to the under-representation of girls in ADHD studies, clinic-
recruited samples have other limitations. Many times, clinic-recruited samples are not
representative of youth with mental disorders. Clinic-recruited samples are more
impaired, less competent, more likely to have comorbid disorders, and more likely to be
non-Hispanic white (Goodman et al., 1997). Therefore, previous research needs to be
balanced using community recruited samples of children.

. Summag

The ADHD subtypes apparently differ in social adjustment and associated
behavioral problems. It is less clear, however, whether these differences indicate
‘ different underlying etiological factors and possible different disorders (Milich et al.,
2001) or rather different manifestations of the same underlying disorder of attention
(F araone et al., 2000). Finding patterns of ADI-ID subtype transmission in families would
lend support to continued theoretical and clinical differentiation between subtypes and
possible different etiologies, whereas replication of no difference (as in Faraone et al.)
could counter that argument.

Children diagnosed with different subtypes of ADHD have differing rates of
associated disorders, with hyperactive children being more likely than inattentive
children to receive an additional externalizing disorder diagnosis, such as CD or ODD.
ADHD children with comOrbid CD were more likely to have parents with both ADHD

and CD themselves. Paralleling that profile, exposure to other types of parental

25

psychopathology was also more common in ADHD families compared to control
families, in several studies. Table 3 summarizes these findings across all studies.
Determining whether parents of ADHD children defined by DSM-IV subtypes have
distinct patterns of psychiatric disorder, other than ADHD, would clarify further whether
the differentiation of the subtypes is a reflection of etiology differences. If the subtypes
of ADI-II) are different disorders, it would be expected that children with different ADHD
subtypes would have parents with different patterns of psychopathology. If the latter case
is true, and in fact the two subtypes do not differ, parents of all ADHD children,
regardless of subtype, would show similar patterns of psychopathology.

It is important to understand the relationship between ADHD subtypes in children
and (a) parental disorder along with (b) comorbid disorders in the children. It has already
been shown that child CD and ODD have their own associated parent psychopathologies.
Thus it is crucial to distinguish parental disorders correlated with child ADHD from
parental disorders correlated with other comorbid child disorders. There has been a
sparse history of research in the area of transmission of disorder in the families of
children with ADHD. There have been no studies that examine the diagnosis of
prospectively defined DSM-IV ADHD subtypes in children and the correlated disorders

in parents.

26

Current Study

Although past studies have shown a link between parent ADHD and child ADHD,
no studies have been done using prospectively defined DSM-IV child diagnosis. In
addition, many of the studies failed to include girls in the study and nearly all used a
clinic-recruited sample. Further, many relied solely on maternal self-report of psychiatric
disorder in both parents, and few used structured interviews. The present study examined
the relationship between DSM-IV ADHD subtypes in children and parental
psychopathology, while correcting limitations in the literature. Specifically it looked at
the association between child DSM-IV ADHD subtypes and the rates of DSM-IV ADHD
subtypes and other diagnoses in parents. Due to the relatively low prevalence of the
hyperactive subtype, only two child DSM-IV ADI-II) subtypes were considered, ADHD-
combined type and ADHD-inattentive type.

Thus this was the first family psychiatric study of ADHD using children
prospectively diagnosed by DSM-IV criteria. The sample included both boys and girls
recruited from the community. Both biological parents were asked to report their own
psychological history in separate structured clinical interviews. Child symptoms were
rated by parent and teacher report.

Primagy Hypotheses.

Following the logic just summarized on the previous page, the following

hypotheses were tested.
Hypothesis 1: Comparing Parents of Children with ADHD-any type.
(I_-A)_P_arent ADHD (any type). In order to replicate past studies (see Table 2) it

was predicted that the parents of children with ADHD (any type) would have a higher

27

prevalence of ADHD (any type) themselves as compared to the parents of the control
children.

(,1-_B)£_arent other (non-ADHD) disorders. It was also predicted that the rates of
psychiatric diagnoses would be higher in the parents of children with ADHD (any type)
as compared with the parents of control children.

Hypothesis H: Comparing Parents of Children with and without comorbid
ODD/CD

In order to control for the effects of child comorbidity, rates of ADHD and other
non-ADHD diagnoses in parents of children with ADHD plus ODD/CD were compared
with diagnoses in parents of children with ADHD alone and parents of control children.

(H-A) Parent ADHD (any fine). It was hypothesized that the parents of children
with ADHD alone and parents of ADHD + ODD/CD would both have a higher
prevalence of ADHD than parents of control children.

(II-B) Parent ADHD subtypes. Although it was predicted that the both parents of
the child ADHD groups (ADHD alone and ADHD+ODD/CD) would have higher rates of
ADHD (any type), it was predicted that there would be no differences between the
parents of the child ADHD groups on rates of specific ADHD subtypes.

(II-C) fluent other (non-ADHD; disorders. It was also predicted that the parents
of children with ADHD alone and parents of ADHD+ODD/CD would have higher rates
of both subtypes of ADHD compared with controls. However there was no predicted
difference between the parents of the two child ADHD groups (ADHD alone and

ADHD+ODD/CD) for prevalence of subtype.

28

Hypothesis HI: Comparing Parents of Children with different DSM-IV ADHD
subtypes

Finally and centrally, ADHD subtypes were also examined. The hypothesis that
ADHD-combined subtype and ADHD-inattentive subtype are distinct disorders (as
presented in Milich et al., 2001) was tested by three predictions.

(III-ALParent ADHD (any type). As a preliminary test, it was predicted that the
parents of children with ADHD-Combined type and of children with ADHD-Inattentive
type would have a higher prevalence of ADHD (any type) themselves when compared
with the parents of control children.

Q-B) Parent ADHD subtypes. It was also predicted that the diagnosis of ADHD
subtype in the children would be the same as the ADHD subtype in the parents.
Specifically [1] parents of children with ADHD inattentive type will be more likely to
have ADHD inattentive type compared with parents of control children and parents of
children with ADHD-combined type. Also, [2] parents of children with ADHD-
combined type will be more likely to have ADHD-combined type, compared with parents
of children with ADHD-inattentive type or parents of controls.

(III-Q) Parent other (non-ADHD) disorders. Child ADHD subtypes will differ in

rates of parent other (non-ADHD) disorders.
Secondagy Anglyses

I. Since there are relatively few studies of childhood ADHD and parental
psychopathology, and no studies of DSM-IV ADHD subtypes and other
psychopathology, additional analyses explored the relationship between child ADHD and

key individual psychiatric diagnoses in parents, including Major Depression, Generalized

29

Anxiety Disorder, Antisocial personality disorder, and Alcohol and Drug dependence. In
addition, the relationship between specific diagnoses in parents and ADHD subtypes in
children was also examined.

11. Almost every previous study failed to include girls (see Table 1). The one
study that specifically used an all female sample did not use prospectively diagnosed
DSM-IV ADHD, nor did they look at mothers or fathers separately to explore unique
gender interactions. Therefore, additional analyses separating girls and boys were
conducted to explore the differences between genders. These were folded in as I

secondary analyses within each of the hypotheses listed earlier.

30

Methods

Overview

Data were gathered fiom children and their biological parents who participated in
the Michigan State University Child Attention Study directed by Dr. Nigg. The families
were community-recruited fi'om the Lansing area. They completed a multi-stage
screening procedure (detailed below) before they were invited to participate in a six-hour
battery broken into two visits of three hours. The present report used data collected
through 2001 with a total N of 144 families participating. The author has been closely
involved with the project for over two years, assisting with data collection.
Participants

Two groups of children and their biological parents were included in the study.
Adoptive and stepparents were excluded from the study to hold consistent the possibility
of genetic transmission. The children were initially included in one of two groups: [1]
children diagnosed with ADHD, any type (n = 95) and [2] control children (n =49). A
total N = 245 parents were available for the study. They included It =159 parents (n= 90
mothers and n = 69 fathers) of children with ADHD and n = 86 parents (n = 49 mothers
and n = 37 fathers) of control children. For additional analyses, children with ADHD
were divided into several groups depending on hypothesis. For hypothesis H, ADHD
children were divided into groups according to comorbidity: [1] children with ADHD
only (n = 48) and [2] children with ADHD plus ODD/CD (n = 46). For hypothesis HI, the
original group of children with ADHD (n = 95) was then restratified according to DSM-
IV ADHD subtypes: [1] children with ADHD Inattentive subtype (n = 28) and [2]

children with ADHD Combined type (n = 66). Additional demographic information for

31

children and their parents (i.e. age, ethnicity) is included in Table 4 and firrther discussed
in the Results section later. All participants were recruited into the study and diagnosed
using the screening process described in the following section.

Procedure

Initial Recruitment. All parents of third, fourth, and fifth graders in the Lansing

 

school district were contacted through mass mailings with information about the study.

In addition, advertisements were placed in local newspapers. Families were also
recruited through local support groups and advertisements in local pediatric clinics.
Although the majority of the sample was community-recruited, a minority of ADHD
children (n=12) was recruited from a local pediatric clinic specializing in ADHD.

Control subjects for those children were recruited from a general pediatric clinic in the
same facility. Those parents who responded to the mailing or newspaper advertisement
completed prescreening to determine the eligibility of their child for the study. A total of
N = 836 families either called or sent in postcards expressing interest in participating in
the study.

Multi-stagp recruitment process: Screen. Those (N = 836) were then screened in
several stages. First the family was contacted by phone and had to meet certain
qualifications. These included a willingness to keep the child off psychostimulant
medication for 24 hours prior to the visit for short-acting medication and 48 hours prior
to the visit for slow-acting medication. English was required to be the first language of
both parents and the child. The initial phone screen also precluded any children with
neurological impairments, seizure history, head injury, other major medical conditions, or

those taking slow-acting psychotropic medication (e.g., antidepressant medication)

32

according to parent report. Children were also excluded if they had a prior diagnosis of
Mental retardation or autism. If the families were not screened out by phone, both the
child’s parents and schoolteachers were invited to complete normative behavior rating
scales (N = 455). Those children who exceeded empirically supported cutoffs on at least
one normed parent rating scale and one normed teacher rating scale (described later) were
eligible as possible ADHD. Children whose scores fell below empirically supported
cutoffs on all normed rating scales were eligible as possible controls.

Eligible children and their mother or primary caregiver were then invited to attend
an on—campus diagnostic screening visit, during which the child completed the Wechsler
Intelligence Scale for Children-third. edition (WISC-III, Wechsler, 1991) and the
Wechsler Individual Achievement Test-Reading (WIAT, Wechsler, 1992) and the mother
completed the Diagnostic Interview Schedule for Children (DISC-IV; Shaffer, Fisher, &
Lucas, 1997). Children who had an IQ above 75 on the WISC-III, and either (a) met
criteria for a diagnosis of ADHD (as defined below) or (b) did not meet criteria for
ADHD, ODD or CD were invited, along with their parents, to participate in the stud (N =
155)

Once afamily was admitted into the study, the biological parents, the child, and a
sibling participated in two additional visits; each visit was three hours long. The children
completed a neuropsychological battery (reported elsewhere), in addition to filling out
questionnaires. The parents also completed a neuropsychological battery (reported
elsewhere), questionnaires (described below) and a structured diagnostic interview about
their own psychiatric history. The interview, the Diagnostic Interview Schedule. is also

described in detail later.

33

Measures
Child Diagnpstic Assignment Measures.

Pre-Screen. The initial prescreen measures alluded to earlier were a DSM-IV
symptom rating scale, a broadband behavior rating scale, and a disruptive behavior rating

scale. For data collected from 1997-1999 these were the Parent and Teacher DSM-IV

 

Checklist (SNAP-IV Swanson et al., 1998), the Child Behavior Checklist (CBCL;

 

 

Achenbach, 1991a), the Teacher Report Form (TRF; Achenbach, 1991b) and the Parent

and Teacher Conners’ Rating Scales-Revised—Short Forms (Conners; Conners, 1997).

 

Data collected from 1999-2001 used the School and Home versions of the ADHD Rating
Scales (DuPaul, Power, Anastopoulos & Reid, 1998), the Parent and Teacher Behavior

Assessment System for Children (BASC; Reynolds & Kamphaus, 1992), and the Parent

 

and Teacher Conners’ Rating Scales-Revised-Short Forms (Conners; Conners, 1997).

 

The ADHD rating scales and the BASC replaced the SNAP-IV and the CBCL,
respectively, because they had superior, empirically validated cut-off scores while
providing comparable symptom coverage.

Psychometric characteristics of prescreen measures. The test-retest reliability for
the School and Home versions of the ADHD Rating Scales were .90 and .85 respectively
(DuPaul et al., 1998). The validity coefficient between the Hyperactivity-Impulsivity
scale of the School version of the ADHD Rating Scale and the Hyperactivity scale of the
Conners Teacher Rating Scale-48 was .79. For the same questionnaires respectively, the

validity coefficient between the Inattention scale and the Daydream-Attention Problems

34

scale was .85. The validity coefficient between the Hyperactivity-Impulsivity scale of the
Home version of the ADHD Rating Scale and the Impulsivity-Hyperactivity scale of the
Conners Parent Rating Scale-Revised was .78 (DuPaul et al., 1998). Internal reliability
(alpha) can be reported from our own data. Internal reliability for the parent version of
the ADHD rating scales was .98. The school version’s reliability was .97. Agreement
between the Home and School versions was .98.

The 15-day test-retest reliability of the CBCL was .90 for the attention problems 1
scale. With respect to validity, the correlation was .86 between the (_JILCL aggressive
behavior and the Conners' conduct problems scales (Achenbach, 1991a). Internal
reliability for our own data was acceptable with alpha equal to .87.

The 15-day test-retest reliability of the m was .96 for the attention problem

scales. In terms of validity of the TRF, the correlations were .80 between the TRF

 

attention and the Conners' inattention-passivity scales, .67 between the m aggressive
behavior and the Conners' conduct problem scales, and .83 between the two measures
total problems scales (Achenbach, 1991b).

With respect to the reliability of the Parent and Teacher ByA_S_§_, the one-month
test-retest reliabilities for the Teacher M attention problems, hyperactivity, and
aggression scales were .92, .92, and .91 respectively (Reynolds & Kamphaus, 1992). The
one month test-retest reliabilities for the Parent m attention problems, hyperactivity,-
and aggression scales were .92, .84, and .69 respectively (Reynolds & Kamphaus, 1992).
In terms of the validity of the B_A_S__C, the correlations were .78 between the Parent BASC

attention problems and the CBCL attention problems scales, .82 between the Parent

35

BASC aggression and the CBCL aggressive behavior scales, and .52 between the Parent

BASC anxiety and the CBCL anxious/depressed scales (Reynolds & Kamphaus, 1992).

 

For the Parent CONNERS the test-retest reliability coefficients were .62, .85, and
.72 for the oppositional, hyperactive, and ADHD scales (Conners, 1997). Internal
reliability from our study for the parent CONNERS had alpha equal to .98 for the ADHD
scale. For the Teacher CONNERS, the test-retest reliability coefficients were .84, .72,
and .80 for these scales respectively (Conners, 1997). Our study’s internal reliability for
the Teacher CONNERS ADI-ID scale was alpha = .97. With respect to the validity of the
Parent CONNERS the correlations between the short and long factor-derived scales were
.98 and .97 for males and .97 and .97 for females for the oppositional and hyperactive
problems scales respectively (Conners, 1997). These correlations for the Teacher
CONNERS were .99 and 1.00 for both males and females (Conners, 1997). In summary,
the screen scales are widely used scales with more than adequate reliability and validity.

Screening Visit. After screen-in cut-offs as possible ADHD were exceeded,
diagnoses of ADHD were confirmed with the help of the National Institutes of Health
Diagnostic Interview Schedule for Children (DISC-IV: Shaffer, Fisher ’& Lucas, 1997)
completed with the mother or primary caregiver. The DISC-IV is a computer-guided,
structured interview that assesses onset, duration, and impairment criterion from the
DSM-IV as well as assessing individual symptoms of disorder. A graduate student
interviewer who first completed 10 hours of training administered the DISC-IV. The
quality of the interviews was checked by having the interview video recorded and 10 %

reviewed by a certified trainer (either Dr. Nigg or Dr. Fisher at Colombia University).

36

DISC-IV. With regard to the reliability of the DISC, test-retest reliability
coefficient for ADI-II) diagnosis was .80. The reliability coefficients for ODD was .73;
.59 for CD. The reliability for Depressive Disorders and Anxiety Disorders were .56 and
.64 respectively (Schwab-Stone et al., 1996). In terms of validity, the correlations with
clinician symptom ratings were .82 for ADI-ID, .73 for ODD, .49 for CD, .67 for
depressive disorders, and .53 for anxiety disorders (Schwab-Stone et al.). In summary,
the DISC-IV isa widely accepted and a cited research diagnostic interview with
acceptable reliability and validity for ADHD.

Final Diagnosis. Children who scored above the 80th percentile on at least one
parent ADI-ID rating scales and above the 90th percentile on at least one teacher ADHD
rating scales were considered as possible ADHD. The diagnosis was confirmed using the
DISC-IV supplemented with an.“or” algorithm. If children met age of onset, duration,
impairment, and cross-situational criteria, then the diagnostic assignment was determined
by adding the endorsed symptoms of the DISC-IV with the teacher reported symptoms to
establish the final ADI-ID subtype. Thus if either teacher or parent reported the symptom
as present, it was counted as present. This method was chosen to approximate the DSM-
IV field trial data (Lahey et al., 1994). Which indicated maximal validity for the DSM-
IV symptom count cut—offs using an “or” algorithm. -

Control children failed to meet cut-offs for all parent and teacher ADHD rating .
scales at the 80th percentile. In addition, they exhibited 4 or fewer symptoms of ADHD
by the “or” algorithm. Those children exhibiting 5 symptoms of overactivity or
inattention by the “or” algorithm were excluded from all groups based on field trial data

indicating that borderline cases might have ADHD (Lahey et al., 1994). Controls also

37

had to have never been diagnosed with ADHD in the past and could not have a sibling
diagnosed with ADHD. Prior parent diagnoses were not considered in the screen
process.

Comorbid Child Diagnpses. The DISC-IV interview was used for establishing the
presence of child Oppositional Defiant Disorder and Conduct Disorder by DSM-IV
criteria.
Parental Diagnostic Assignment

ADHD in parents. Because DSM-IV does not provide adult-specific diagnostic
criteria for ADHD, parents completed a Structured diagnostic interview of their childhood
behavior as well as current behavior. As a supplement, parents also completed a series of
normative self-report questionnaires concerning adult behaviors related to problems with
attention, activity, and impulsivity. The questionnaires and the interview ask participants
to retroactively assess ADHD symptoms in their childhood, as well as current symptoms.

Participants completed the Self-SNAP a revised version of the Swanson, Nolan, and

 

Pelham DSM ADHD rating scale (SNAP-IV) (1998). This version was revised using
the same language as the original SNAP-IV in order for the parents to retrospectively rate
their own behavior. All participants were also given the ADHD module of the Diagnostic
Interview Schedule-IV (DIS-IV), based on DSM-IV (1994) ADHD criteria. The DIS-IV
asks parents to report on their own childhood symptoms of ADHD as well as any
symptoms of the disorder during adulthood. It also assesses impairment and age of onset.
A total ofN = 160 parents completed both the DIS-IV and the Self-SNAP.
Agreement between the instruments for diagnosing the presence of ADHD (any type)

was 90.0% (Kappa = .58, p< .001). For the diagnosis of the DSM-IV ADI-ID subtypes,

38

the instruments had 94.0% agreement on the diagnosis of ADHD-inattentive type (Kappa
= .642, p = .000) and 96.0% agreement on the diagnosis of ADHD-combined type
(Kappa = .480, p = .000). A final diagnosis of ADI-ID and DSM—IV subtype in parents
was defined when criteria for that subtype in the parent’s own childhood was met on the
DIS-IV. Diagnoses were made for the parents were based on parents’ retrospective
report of their own symptoms in childhood. Individuals must report at least 5 or more
symptoms of inattention or hyperactivity’. Impairment on the DIS-IV must be rated
moderate or severe, have a reported age of onset before 7 years. Parents who were just
shy of meeting diagnostic criteria (4 symptoms of either inattention or hyperactivity)
were excluded from parent ADHD analyses (n = 2). DSM-IV subtypes were determined
if criteria was meet for one area, but met 3 or fewer symptoms in the other area. Parents
who met criteria for one symptom area, but endorsed 4 symptoms in the other were
counted as positive for ADHD (any type) but were excluded from DSM-IV subtype
analysis (n = 2). Due to constraints on the number of testers and total length of the
testing battery, n = 28 parents did not complete the DIS-IV. For those parents, diagnoses
were based on endorsed symptoms on the Self-Snap, justified by the acceptable
agreement between the instruments.

Parental Diagnostic Assignment-other disorders. Other psychiatric disorders in
assessed parents were: Generalized Anxiety Disorder (GAD), Major Depression Disorder
(MDD), Dysthymia, Bipolar Disorder, Antisocial Personality Disorder (ASPD), Alcohol
Abuse, and Drug Abuse. All were assigned on the basis of the DIS-IV interview.

However some parents used a DSM-III-R version for the diagnosis, as explained next.

39

DI; and ODIS. In addition to the DIS-IV ADHD module, each parent also
completed either the rest of Diagnostic Interview Schedule-IV (DIS-IV) or the Quick
Diagnostic Interview Schedule (QDIS) for the DSM-III-R in order to assess current and
lifetime occurrence of non-ADHD psychiatric disorders. The two are versions that are
computer assisted, structured interviews administered by trained interviewers. The
interview is divided into modules based on diagnostic categories. Each module follows a
Probe Flow Chart in which each question is followed up by a series of probe questions
exploring severity and alternative explanations (e.g. medical causes) of any symptom.
Each question appears on the screen to be read by the interviewer exactly as written. The
interviewee responds to most questions with a simple yes or no answer recorded by the
interviewer, and the next question to be asked appears on the screen. The Q-DIS-III-R
assessed DSM-III-R diagnoses; once a disorder was either established or ruled out, the
interview moved to questions about the next disorder rather than assessing all the
symptoms of a disorder. N = 39 parents completed the Q-DIS-III-R, which was utilized
early in the project due to the lack of availability of the DIS-IV program for DSM-IV.
The DIS-IV became available in 1998 and was then used.

Although for most disorders, DSM-III-R and DSM-IV diagnostic criteria are
similar, analyses were conducted to evaluate whether the rates of diagnoses were
significantly different based on type of interview used. Complete results of those
comparisons are provided in Appendix B. In brief, 2x2 chi-square comparisons were
done comparing the fiequency of each disorder assessed in the DIS-IV and QDIS-III-R
(GAD, Dysthymia, drug dependence, ASPD, Alcohol dependence, depression). Overall,

no differences in prevalence were found between versions at less than p=. 31. An

40

additional 2x2 Chi-square was done to compare the presence or absence of any disorder
across the two versions. Again, no difference was found (x2(1)=. 803, p = .370). Finally,
using a one-way ANOVA, the QDIS-III-R and DIS-IV were compared on the total
number of disorders diagnosed for each person. The two interviews did not differ in
average number of disorders diagnosed (F (1)=1.26, p = .26). Therefore it was deemed
appropriate to pool across both instruments in testing the hypotheses.

DIS Reliability. In initial field studies done with the DIS-IV, inter-rater reliability
(Kappa) was .80 for Alcohol Dependence, .77 for Drug dependence, .63 for Mania, .63
for Antisocial personality disorder, and .63 for depression (Robins et al., 1981) The QDIS
has been shown retain equally reliability as the early paper versions of the firll DIS as
well as having acceptable reliability to the full DIS itself (.69) (Bucholz et. al., 1996).
The newer computer version of the DIS also shows acceptable reliability with previous
paper versions of the DIS (.64) (Erdman et al., 1992). Overall, the DIS shows high
reliability across administrations. It is widely used and accepted as a valid research and
diagnostic tool designed for community samples.

Data Analysis

In order to assess the differences between the groups of children (those with
ADHD (any type) and control children), a series of 2x2 chi-squares were performed. First
child groups were compared in terms of presence of parent ADHD (any type). For power
analysis purposes, population affect size estimates were made using rates of ADI-ID
found in Table 2. The difference in rates of ADHD (any type) between parents of
children with ADHD (any type) and parents of control children was estimated to be 14%

(average in Table 2). For a control sample size of n = 86 parents (mothers and fathers

41

combined) and a ADHD (both subtypes) sample size of 159 parents, power was equal to
.96 to detect a 14% difference between groups using a chi-square with 2-tailed alpha test
(.99 for a l-tailed alpha). Mothers and fathers of the two child groups were then analyzed
separately. Using the same method for estimating population prevalence, a 23
percentage-points difference was assumed between fathers of children with ADHD and
fathers of control children in the rate of ADHD (any type). For the available fathers,
power was equal to .74 to detect this effect for 2-tailed test of significance (.90 for 1-
tailed alpha). Average difference between the mothers of the two child groups was
estimated to be 16 points, which yielded a power equal to .80 (2-tailed) or .88 (l-tailed
alpha) for the available n of mothers for the two groups. . '

Additional 2x3 chi-squares were used to assess the differences between the
parents of the child DSM-IV ADI-ID subtype groups. Because this was the first time
ADHD subtypes as defined by the DSM-IV were examined in parents, there were few
prior data on which to base an estimate of effect size in the population. However if the
subtypes are etiologically distinct disorders, as proposed by Milich, Balentine, and
Lynam (2001), differences comparable to those between ADI-1]) (any type) and control
groups might be expected and population estimates from Table 2 were again used. As
before the differences between all parents (mothers and fathers) were examined. Sample
sizes for parents of children with ADHD-combined was n = 106 and for parents of
children with ADHD-inattentive is n = 51 (mothers and fathers combined). An estimate
of a 14-point difference in prevalence between parents of children with ADI-ID-
Inattentive type and parents of children with ADHD-Combined type would yield power

equal to .83 for a 2-tailed alpha test (.95 for l-tailed alpha). The power to detect a 14-

42

point difference between parents of control children and parents of children with ADHD-
Combined type equals .93 for 2-tailed alpha (.99 for l-tailed alpha). To detect the same
difference between parents of control children and parents of children with ADHD-
Inattentive type power for a 2-tai1ed test of significance equals .80 (.94 for I-tailed
alpha).

Mothers and fathers of the three child groups were again analyzed separately
using the population estimates as described above. Power to detect a 16—point difference
between mothers of children with ADHD-Combined type (II = 61) and mothers of
children with ADHD-Inattentive type (11 = 28) was .58 for 2-tailed alpha (.70 l-tailed
alpha). A 2-tailed test would have a power of .55 (.64 for 1-tailed) to detect a 16-point
difference between mothers of control children and mothers of children with ADHD-
Inattentive type; power equals .73 (.83 for l-tailed alpha) to detect the same difference
between mothers of controls and mothers of children with ADHD-Combined type.

With an average estimated difference of 23 percentage points between fathers of
children with ADHD-Inattentive type and fathers of children with ADHD-Combined
type, power = .62 (2-tailed alpha) to detected a 23-point difference (.74 for l-tailed ’
alpha). Power = .75 (.84 l-tailed alpha) to detect the same difference between fathers of
controls and fathers of children with ADHD-Inattentive type. Power = .75 (.84 l-tailed
alpha) to detect a 23-point difference between fathers of controls and fathers of children
with ADHD-Combined type. However it is recognized that the power was more limited
to detect potentially important smaller differences between the subtypes; this is

considered later in data interpretation.

43

Results
Participants

Demographic information for children and parents are included in Table 4. A
total of 144 children were included in the study. The two groups of children, control
children (n = 49) and children with ADHD (n = 95), did not significantly differ in
ethnicity or in IQ. Despite efforts to include an equal amount of girls with ADHD, more
girls were included in the control group (40.8%) than in the ADHD group (26.3%). Also
control children were slightly older (mean age = 9.78 yrs.) than children with ADHD
(mean age = 9.42 yrs); however, this difference was shy of significance. Differences in
behavior ratings were consistent with diagnostic groupings; children with ADHD were
rated as having a greater amount of inattentive and hyperactive symptoms compared with
control children.

Mothers of the control children (n = 49) and mothers of children with ADHD (any
type) (n = 90) were similar in age, ethnicity, IQ, educational level, and full time
employment status (measured by number of months employed firll time in the last 12
months). The fathers of control children (n = 37) and fathers of children with ADHD (n
= 69) also did not differ in respect to the same variables. However, mothers and fathers
of control children were more likely to be married than the mothers and fathers of
children with ADHD.

Hypothesis 1: Comparing ADHD (any type) with controls.
To address the first hypothesis, children were placed into two groups: [a] children

with ADHD (any subtype) and [b] control children.

44

(I-A) Parent ADI-ID (any type). The two groups were first compared on the
fi'equency of parent ADHD diagnosis (any subtype). Table 5 lists all results. Significance
was determined if groups differed at p<. 05 two-tailed, however because of directional
hypotheses, l-tailed significance is also reported. Parents of children with ADHD were
more likely to have had ADHD (21.4%) compared with the parents of control children
(4.7%). When examined separately, both mothers (20.0%) and fathers (23.2%) of
children with ADHD were more likely to have had ADHD themselves compared with the
mothers (4.1%) or fathers (5.4%) of controls. Thus, Hypothesis I-A was supported.

In order to look at the effects of gender in separate analyses, the parents of girls
and parents of boys were examined independently (Table 6 list the results separately for
girls and boys). The parents of girls with ADHD had a higher rate of ADHD themselves
(22.5%) when compared with parents of control girls (6.1%). The same was true for
parents of boys (21.0% vs. 3.8%). Mothers of girls with ADI-II) were also more likely to
have had ADHD themselves (20.8%) when compared with the mothers of control girls
(0%). However fathers of girls with ADHD (25.0%) were not more likely to have
ADHD themselves compared with the fathers of control girls (15.4%). The apparently
high prevalence may have been an artifact of the low It (total affected fathers, n = 2). The
opposite was true of mothers and fathers of boys. Fathers of boys with ADHD (22.6%)
were more likely to have ADHD themselves compared with the fathers of control boys
(0%). There is not a significant difference between the prevalence of ADHD in mothers
of boys with ADHD (19.7%) and mothers of control boys (6.9%).

(l-B) Parent other (non-ADHD) disorders. The two child groups were also

compared on the prevalence of parental non-ADHD psychiatric diagnoses (Generalized

45

Anxiety Disorder, Major Depression Disorder, Dysthymia, Bipolar Disorder, Antisocial
Personality Disorder, Alcohol Abuse, and Drug Abuse) using a 2x3 chi—square as follows.
Parents were separated into three groups, a) no diagnosis, b) only one other non-ADHD
diagnosis, or c) two or more other non-ADHD diagnoses. Parent ADI-1D was excluded
fiom analysis. Parents of children with ADHD did not differ in rates of 0, 1, or 2 or more
diagnoses when compared with the parents of Control children. Differences in the total
number of disorders were assessed using a one-way ANOVA. Parents of children with
ADI-1D had a higher number of disorders compared with parents of control children,
however results were short of significance (F (1) = 3.15, two tailed p = .077).

When mothers and fathers were examined separately, the number of mothers of
controls and mothers of ADHD children with only one disorder did not differ. However
more mothers of children with ADHD had 2 or more diagnoses than mothers of control
children. Mothers of children with ADHD also had a greater number of disorders than
mothers of control children (F (1) = 4.99, p = .027). Fathers of controls and fathers of
ADI-ID children did not differ in the chance of having 1 or 2 or more other disorders.
Fathers of the two child groups also did not differ in total number of disorders (F (1) =
.18, p = .672). Thus overall, hypothesis I-B was only partially supported, the only
differences found were in rates of maternal non-ADHD disorder.

No significant differences were found between the parents of girls or boys with
ADHD compared with the parents of control girls or boys for rates of non-ADHD
disorders. (Results found in Table 6) When mothers and fathers were examined
separately, mothers of girls with ADHD were not more likely than control mothers to

have 1 disorder. However, mothers of ADHD girls were more likely to have 2 or more

46

disorders compared with the mothers of control girls. Mothers of control girls were more
likely have 1 disorder compared to more than 2 disorders, whereas mothers of girls with
ADHD were more likely to have 2 or more disorders compared to only one disorder
(x2(l)= 4.219, p=. 040). Fathers of girls with ADHD were more likely to have one
disorder compared to fathers of control girls. They were not more likely to have two or
more disorders compared with fathers of control girls. There were no differences in rates
of non-ADHD disorder between the mothers and fathers of boys with ADHD compared
with the mothers and fathers of control boys.

For additional analyses, the two groups of children were also compared on rates
of individual parental disorders. When examining all parents, parents of controls and
parents of children with ADI-ID did not differ in rates of any disorder (GAD, MDD, BPD,
Dysthymia, ODD/CD, ASPD, Drug dependence, and Alcohol dependence). Results are
found in Table 7. Fathers of the two groups also did not differ in prevalence of any
disorder. Mothers of children with ADHD were more likely to have GAD (19.0%)
compared with mothers of controls (6.7%), however the difference did not reach
significance at two-tailed significance.

Hypothesis 2: Comparing child comorbid groups.

To address the second hypothesis, children were placed into three groups: (a)
children with ADHD (any subtype) alone (b) children with ADHD plus ODD or CD and
(c) control children. The three child groups were first compared on the fi'equency of
parent ADHD (any type) diagnoses. Secondly the child groups were compared on the

frequency of parent ADHD subtypes. All comparisons were made using chi-squares;

47

Table 8 lists results of all 3-group chi-squares. Ifthe omnibus chi-square was significant,
follow up pair-wise chi-squares were performed (represented in Table 8 by superscripts).

(A)_P_arent ADHD (any typ_e). Parents of both the ADHD - alone group (12(1)=
8.339, p= .004) and the ADHD + ODD/CD group (x2(1)= 11.681, p= .001) were mOre
likely to have had ADHD compared with parents of the control children. The two ADHD
groups did not significantly differ fi'om each other (x2(1)= .449, p= .503). This pattern
remained significant when mothers and fathers were analyzed separately. Mothers of
children with ADHD alone (x20): 4.029, p= .045) and mothers of children with ADHD
+ ODD/CD (x2(1)= 7.270, p= .007) were more likely to have had ADHD compared with
the mothers of control children. However the mothers of the ADHD-alone and ADHD
+ODD/CD groups did not significantly differ from each other (x20): .633, p= .426).
Fathers of children with ADHD‘alone(x2(1)= 4.202, p= .040) and fathers of ADHD +
ODD/CD (x2(l)= 4.454, p= .035) were also more likely to have had ADI-ID. Again,
there were no differences between the fathers of the ADHD groups (x2(1)= .025, p=
.874). Thus hypothesis H-A was supported.

fluent ADHD subtypes. When rates of parental ADHD-subtypes were
compared across child comorbid groups, parents of children with ADHD-alone (12.6%)
were more likely than parents of controls (3.7%) to have ADHD-Inattentive type (78(1):
4.899, p= .027). Parents of children with ADI-ID + ODD/CD were not more likely to
have ADHD-Inattentive type compared with the parents of controls (x2(1)= 1.46, p= .23).
The parents of the two child ADHD groups (ADHD alone and ADHD+ODD/CD) did not
differ in rates of ADHD-Inattentive type (x2(l)= .816, p= .366). When parents were

examined separately, the three groups of mothers and the three groups of fathers did not

48

significantly differ in rates of ADHD-Inattentive type, perhaps due to a reduction in
power.

When the child comorbid groups were compared on rates of parental ADHD-
Combined subtype, both parents of children with ADHD —alone (x2(1)= .275, p= .039)
and parents of children with ADHD + ODD/CD (x2(1)= 11.951, [F .001) were more
likely to have had ADHD-Combined type compared to parents of controls. Compared
with parents of children with ADHD —alone, the parents of ADHD + ODD/CD children
had higher rates of ADHD —combined type although the difference was significant by 1-
tailed test (x2(1)= 3.239, p= .036). The same analyses were repeated for mothers and
fathers separately.

Mothers of children with ADHD + ODD/CD had a higher rate of ADI-1D-
Combined type than the mothers of control children (x2(1)= 5.297, p= .021). This
difference was also significant for the same two groups of fathers (x2(1)= 6.667, p= .010).
The mothers (x20): 2.957, p= .086) of the ADHD- alone group did not differ
significantly from either the mothers of controls or from the mothers (x20): .134, p=
.714) of the ADHD + ODD/CD group in rates of ADHD combined type. Rates of
ADHD-combined type also did not differ between the fathers of children ADHD-alone
and fathers of controls (x20): 2.008, p: .156) or between the fathers of children with
ADHD alone and fathers of children with ADHD+ODD/CD (x2(1)= .794, p= .373).

Because the pattern for parent ADHD-subtypes appeared to differ between the
child ADHD-alone and child ADHD+ODD/CD groups, those were compared directly.
Using 2-group chi-squares, the groups were tested for an interaction between parent

ADHD subtype by child comorbid subtype. Parents of children with ADHD alone were

49

more likely to have ADHD-Inattentive type than ADHD-Combined type whereas parents
of children with comorbid ODD/CD are more likely to have ADHD-Combined type than
ADHD-Inattentive type (1,20): 4.14, p= .042). This interaction was not significant when
comparing the mothers (x2(1)= 1.167, p= .280) or the fathers (78(1)= 3.233, p= .072) of
the ADHD groups by a 2-tailed test, perhaps due to loss of power. Thus hypothesis II-B
was not supported, and in fact there was a significant difference in rates of parent DSM-
IV subtype between the child ADHD-alone and the child ADHD+ODD/CD groups.

(CLPtarent other (non-ADHD) disorders. When comparing the rates of other non-
ADHD parental (all parents, and mothers and fathers separately) diagnoses (0 disorders
versus 1 disorder versus 2 or more disorders), no differences were found between the
three groups of children. When comparing the total number of parental non-ADHD
disorders, there were again no differences between child groups (F (2) = 1.39, p= .250).
There were also no differences between the 3 groups of paternal non-ADHD disorders (F
(2) = .57, p = .566). There was, however, a difference between child groups in rates of
maternal disorder (F (2) = 3.07, p = .050). Mothers of children with ADHD+ODD/CD
were more likely to have a greater number of diagnoses compared with mothers of
controls (p = .03 6). There were no differences between mothers of children with ADHD-
alone and mothers of controls (p= .537) or between mothers of children with ADHD-
alone and mothers of children with ADHD+ODD/CD (p = .340). Thus, hypothesis II-C
was partially supported.
Hypothesis 3: Comparing ADHD Subtypes.

To address the third hypothesis, all children were restratified into one of three

groups: (a) ADHD combined subtype, (b) ADHD inattentive subtype, and (c) control

children. An omnibus chi-square compared the three child groups for fiequency of each
parent ADHD subtype.

(Aliment ADHD (any type). First, rates of ADHD (any type) in the three groups
of parents were compared. Parents of children with ADHD-Combined type were more
likely to have ADHD themselves than parents of the controls (38(1)= 14.23, p= .000).
Parents of children with ADHD-Inattentive type had higher rates of ADHD compared
with parents of controls, however the difference was only significant by one-tailed test
(x2(1)= 3.57, p= .03). Parents of the two child ADHD groups did not significantly differ in
rates of ADHD (38(1): 2.42, p= .120). Both mothers (x20): 7.784, p= .005) and fathers
(x2(1)= 6.483, p= .011) of children with ADHD-Combined type were more likely to have
ADHD themselves when compared to controls. Mothers of children with ADI-1D-
Inattentive type did not significantly differ in rates of ADHD from mothers (38(1)= 2.58,
p= .108) of controls or mothers (78(1): .893, p= .345) of children with ADHD-Combined
type. The rates of ADHD did not differ between fathers of children with ADHD-
Inattentive type and fathers of controls (38(1): 1.08, p= .298) or between the fathers of the
two groups of children with ADHD (inattentive and combined) (x2(1)= 1.64, p= .200).

The parents of boys and girls were then examined independently. Parents of girls
with ADHD combined type were more likely to have ADI-1]) themselves compared with
parents of control girls (x2(1)= 3.90, p= .048). There were no other differences between
parents of the other groups of girls. Parents of boys with ADHD-Combined type were
more likely to have ADI-ID compared with parents of control boys (12(1): 10.12, p=
.001). Parents of boys with ADHD-Combined type were also more likely to have ADHD

compared with parents of boys with ADHD-Inattentive type (x2(1)= 3 .04, p= .04)

51

 

 

 

althOugh the difference was only significant by a one-tailed test of significance. There
were no difference between the parents of boys with ADHD-Inattentive type and parents
of control boys (78(1): 1.22, p= .270).

Mothers of girls with ADHD-Combined type were also more likely to have
ADHD (any type) compared with mothers of control girls (x20): 5.52, p= .019).
Mothers of girls with ADHD-Inattentive type (16.7%) were more likely to have ADHD
compared with the mothers of control girls (0%) however it fell short of significance at a
Zetailed alpha level (1,20): 3.56, p= .059) perhaps due to lack of power. There were no
differences between the fathers of the three groups of girls.

This was not the case with the groups of boys. Fathers of boys with ADHD-
Combined type were more likely to have ADHD themselves compared with the fathers of
control boys (78(1)= 7.76, p= .005). Fathers of boys with ADHD-Combined type also
had a somewhat higher prevalence of ADHD compared with the fathers of boys with
ADHD-Inattentive type (38(1): 2.65, p= .103) however the difference fell short of 2- A
tailed significance. There were no differences between rates of ADHD in fathers of boys
with ADHD-Inattentive type and fathers of control boys (x2(1)= 1.64, p= .200). Also,
unlike the mothers of girls, there were no differences in the rate of ADHD in the mothers
of the three groups of boys. Thus, support for hypothesis IH-A was mixed.

(B) Parent ADHD subtypes. The 3 groups of children were also compared for
rates of parental ADHD subtypes. Compared with parents of control children, parents of
children with ADHD-Inattentive type (78(1) = 4.40, p= .036) and parents of children with
ADHD-Combined type (36(1)= 9.67, p= .002) were more likely to have ADHD-

Inattentive type themselves. However parents of the two groups of children with ADHD

52

did not differ fiom each other in rates of parent ADHD-Inattentive type. Mothers of
children with ADHD-Combined type had higher rates of ADHD-Inattentive type
compared with parents of control children (78(1) = 4.155, p= .042). Rates of ADHD-
Inattentive type in mothers of children with ADHD-Inattentive type did not differ from
mothers of controls (78(1) = .24, p= .623) or from mothers of children with ADHD-
Combined type (78(1) = 1.45, p= .793). Rates of ADHD-Inattentive type in fathers of the
three groups of children did not differ.

Rates of ADHD-Combined type in parents of children with ADHD-Inattentive
type (78(1) = 5.41, p= .020) and parents of children with ADHD-Combined type (78(1) =
9.67, p= .002) were higher than in the parents of control children. The rates of ADHD-
Combined type did not differ in the parents of the two different ADHD child groups.
Mothers of children with ADHD-Inattentive type were more likely to have ADHD-
Combined type than mothers of controls (78(1) = 5.54, p= .019). There was no difference
in the rate of maternal ADHD-Combined type between mothers of controls and mothers
with ADHD-Combined type (78(1) = 2.91, p= .088) or between the mothers of the two
child ADHD groups (78(1) = .677, p= .411). However rates of ADHD-Combined type in
fathers of children with ADHD-Combined type was higher than in fathers of controls
(78(1) = 6.76, p= .009) and higher than in fathers of children with ADHD-Inattentive type
(78(1) = 3.96, p= .047). Furthermore, fathers of children with ADHD-Inattentive type
were more likely to have ADHD-Inattentive rather than ADHD-Combined type
themselves, while fathers of children with ADHD-Combined type were more likely to
have ADHD-Combined type rather than ADHD-Inattentive type (78(1) = 3.82, p= .051).

This interaction was not significant when comparing all parents of the child ADI-fl)

53

groups (78(1) = .048, p= .827) or the mothers of the child ADHD groups alone (78(1) =
2.36, p= .124).

When boys and girls were examined separately, an interesting pattern emerged.
There was no significant difference between the 3 groups of girls on parental rates of
ADHD-Inattentive type. However, parents of girls with ADHD-Inattentive type (78(1) =
3.81,p= .051) and parents ofgirls with ADHD-Combined type (78(1) = 3.81,p= .051)
were more likely to have ADHD-Combined type than parents of control girls. There was
no significant difference between parents lof girls with ADHD-Inattentive type and
parents of girls with ADHD-Combined type (78(1) = 0, p=1) in rates of ADHD-Combined
type. Boys with ADHD-Combined type were more likely to have a parent with ADI-ID-
Inattentive type (78(1) = 4.35, p= .037) and more likely to have a parent with ADHD-
Combined type (78(1) = 5.95, p=. .015) than control boys. Parents of boys with ADHD-
Inattentive type did not differ fiom parents of control boys in rates of ADHD-Inattentive
type (78(1) = 1.17, p= .279) or in rates of ADHD-Combined type. Also, the parents of
boys with ADHD-Inattentive type and parents of boys with ADHD-Combined type did
not differ in rates of ADHD-Inattentive type (78(1) = .715, p= .398) or in rates of ADHD-
Combined type (78(1) = 1.50, p= .221).

Further analysis separating the mothers fiom fathers of boys and girls, uncovers
even more remarkable results. The 3 groups of girls did not differ in rates of mother or
fathers with ADHD-Inattentive type. Rates of parental ADHD-Combined type were
higher in mothers of girls with ADHD-Inattentive type compared with mothers of control
girls (78(1) = 3.89, p? .049). Rates of parental ADHD-Inattentive type did not differ

between the mothers of girls with ADHD-Combined type and mothers of control girls

54

(78(1) = 2.07, p= .150) or between the mothers of girls with ADHD-Combined type and
mothers of girls with ADHD-Inattentive type (x20) = .95, p= .330). There were no
significant differences between the rates of ADHD-Combined type in the fathers of the
three groups of girls (ADHD-Combined, ADHD-Inattentive, and control).

For boys, rates of maternal ADHD-Inattentive and ADHD-Combined type did not
differ between the three groups. Also rates of paternal ADHD-Inattentive type did not
differ between the 3 groups of boys. However, boys with ADHD-Combined type were
more likely to have a father with combined type compared with control boys (78(1) =
4.82, p= .027). They were also more likely to have a father with ADHD-Combined type
when compared with boys with ADHD-Inattentive type (78(1) = 2.92, p= .044), when
using a one-tailed test of significance. There was no difference between boys with
ADHD-Inattentive type (0%) and Control boys (0%) for rate of paternal ADHD-
Combined type. Overall, hypothesis III-B was not supported.

Merit other non-ADHD disorders. The 3 groups of children were compared
on the presence of parent non-ADHD Disorder. The 3 child groups did not differ in total
number of non-ADHD disorders in parents (F (2) = 2.05; p = .131) or in fathers (F (2) =
.17; p = .841). However, mothers of children with ADHD-Combined type were more
likely to have 2 or more disorders when compared with parents of control children (78(1)
= 5.48, p= .019). Mothers of children with ADHD-Combined type were also more likely
to have a greater total of disorders compared with mothers of controls (p = .018). Rates of
maternal non-ADHD disorders did not differ between mothers of children with ADHD-

Inattentive type and mothers of controls (78(1) = .74, p= .390) or between the mothers of

55

the two child ADHD groups (78(1) = 1.31, p== .253). Thus Hypothesis III-C was supported
only when mothers were examined.

When looking at individual diagnoses in parents, the parents of three child groups
only differed in rates of drug dependence. Parents of children with ADHD-Combined
type had higher rates of drug dependence (12.8%) compared with parents of controls
(78(1): 4.09, p= .043). Rates of drug dependence did not differ in parents of controls and
parents of children with ADHD-Inattentive type (78(1)= .466, p= .495) or between
parents of children with ADHD-Inattentive type and parents of children with ADHD-

Combined type (78(1)=1.16, p= .281).

56

Discussion

F amily prevalence studies provide a key step in considering possible
developmental context, etiological pathways, genetic effects, and validity of nosological
distinction of psychopathological syndromes. In the case of ADHD, a promising early
literature has lacked completion in terms of DSM-IV subtypes. Existing family studies
have for the most part overlooked the subtypes, failed to control for comorbidity, and all
but ignored possible differences in gender effects. These and other problems have lefi
the ADHD family literature with an incomplete understanding of parent psychopathology
in relation to child ADHD as now defined. The present study was the first to examine the
prevalence of parent psychopathology (both DSM-IV ADHD subtypes and other'non-
ADHD disorders) in the families of children with prospectively defined DSM-IV ADI-ID
subtypes.

The clearest and most easily understood result of the present study was the
replication of past findings that children with ADHD were more likely than control
children to have parents who also have ADHD. Confirmation of this basic fact using
DSM-IV isreassuring, but not surprising. Nearly all, prior family studies have shown an
elevated rate of the then-current form of ADHD in relatives of children with ADHD.
However, more interesting in the present study was the new data obtained concerning
comorbidity, subtypes, and gender effects. Each of these findings is discussed in detail;
contribution to understanding possible etiology and the value of current diagnostic
classification of ADHD is also considered.

Child Comorbidity and Parent ADHD. When the presence of comorbid child

disorders was controlled, the main finding remained significant: parents of children with

57

ADHD, regardless of comorbid behavior disorders in the children, were more likely to
have ADHD themselves. Parents of children with ADHD + ODD/CD had similar rates
of ADHD as the parents of children with ADHD alone, filrther supporting the specific
validity of ADHD. For example past studies have asserted that it is comorbid disorders
that “explained” the association between parental psychopathology and child ADHD

- (Lahey et al. 1988), however the results of the current study found that the presence of
comorbid disorders did not explain the relationship between parent and child ADHD. To
this extent, the results were consistent with the findings of Frick et al. (1991). Like their
data, the coooccurrence of ADI-1D plus ODD/CD in children in our sample did not predict
higher rates of ADHD in parents than did child ADHD alone.

Results also demonstrated a significant relationship between DSM-IV ADHD
subtypes in parents and child comorbidity, which was not consistent with the findings of
Frick et a]. (1991). Parents of children with ADHD alone were more likely to have the
Inattentive subtype, whereas the parents of children with ADHD+ODD/CD were more
likely to have ADHD-Combined subtype. Although there is controversy about whether
the DSM-IV subtypes are best viewed as degrees of severity of ADHD, if one was to
think of ADHD-Combined type as a more severe manifestation of ADHD (individuals
have both inattentive and hyperactive/impulsive symptoms) compared with ADHD-
Inattentive type, then perhaps our data are consistent with prior studies. Although the
presence of child ODD/CD alone did not account for the presence of parental ADHD, it
suggests that the child ADHD+ODD/CD group could demonstrate a more severe disorder

than ADHD alone. The former is more likely to be correlated with more severe parental

58

psychopathology. This study also looked at non-ADHD disorders in parents in order to
explore this fiirther.

Non-ADHD Disorders in Parents. Past research was not replicated when
examining rates of non-ADHD disorders in all parents of children with ADHD. When all
parents of children with ADHD were examined, they did not have higher rates of non-
ADHD disorders when compared with control children, nor were any specific non-
ADHD disorders in parents related to ADHD in children. Also striking was that the
number of non-ADHD disorders in parents was unrelated to child comorbidity status.
This result is contrary to findings of either Lahey et al. (1988) or Frick et al. (1991) using
earlier definitions of ADHD in samples consisting only of boys. Both studies found that
a diagnosis of child ADHD plus CD predicted higher rates of non-ADHD disorders in
parents when compared with parents of children with ADHD alone. Several explanations
for the contrasting results are possible.

First of all, unlike prior studies, here mothers and fathers were examined
separately. There were no significant differences in fathers’ rates of non-ADHD disorder
between fathers of children with ADHD and fathers of controls. However, mothers of
children with ADHD, regardless of comorbidity, were more likely to have non-ADHD
disorders than mothers of control children, replicating past studies. Unique to our study,
non-ADHD parental diagnoses were examined in relation to DSM-IV subtypes. Mothers
of children with ADHD-Combined type had a greater number of disorders than mothers
of children with ADHD-Inattentive type and mothers of control children. Specifically
mothers of children with ADHD-Combined type were more likely to have a diagnosis of

Major depression and drug dependence.

59

The current finding that only mother’s disorders were related to child ADHD
could be merely a result of the different diagnostic procedures. In both Lahey et al.
(1988) and Frick et al. (1991), parental disorders were assessed by asking the mother of
the child to rate themselves and the child's father, whereas the present study directly
interviewed both parents. Perhaps mothers had over-reported psychiatric symptoms in
the child’s father in previous studies, leading to over-estimating the relationship between
paternal pathology and child ADHD. On the other hand we may have underestimated.
Because a significant number of fathers did not participate, it could be that the fathers
who were unavailable for the study had more disorders than the fathers that were still in
the family. Thus the current study could be under-representing the number of non-
ADHD disorders in the fathers of ADHD children. In addition, both of the
aforementioned studies used clinic recruited samples in which child disorder and familial
correlates could be more severe than would be found in the largely community recruited
samples used here, although many of the community recruited children in our study had
previously obtained clinical services.

However, non methodological interpretations warrant consideration. The higher
rates of non-ADHD disorders in mothers of children with ADHD-Combined subtype
could be interpreted in two contrasting ways. It could mean [1] findings could lend
support the hypothesis that ADHD combined type is merely a more severe manifestation
of ADI-ID or [2] results could also be interpreted as ADHD-Combined type having
distinct parental correlates compared with ADHD-Inattentive type, thereby could validate

the distinction between the subtypes. Either hypothesis, however, does suggest a possible

60

etiology, in that maternal non-ADHD disorder contributes to either the severity or the
manifestation of ADHD.

Maternal psychopathology has been correlated with a number of psychological
and behavioral outcomes in children (F rick, 1994). For example it has been suggested
that maternal psychopathology may degrade parenting ability, disrupt the marital
relationships, and even effect work performance and result in lower socioeconomic
status. Therefore, it is not entirely surprising that maternal pathology might also
contribute to the severity of behavior problems in children. Ifthis were the case, it would
be expected that more severe maternal psychopathology would result in more severe and
a greater number of ADHD symptoms. Conversely, if maternal disorder is correlated
with the manifestation of ADHD beyond the severity and/or number of symptoms of the
subtypes, it could suggest that maternal psychopathology contributes unique risk factors
for the development of ADHD-combined type independent of the risks for the Inattentive
subtype and lend support to the distinction between the subtypes. However, in order to
further determine the nature of familial transmission of ADHD, it is important to clarify
the relationship between parental DSM-IV ADHD subtypes and DSM-IV subtypes in
children.

Parent and Child DSM-IV ADHI) Subtypes. When all parents were examined
together, DSM-IV subtypes in parents did not directly correspond to the subtype
diagnosed in their child. As was found in previous studies of DSM-IV ADHD, subtypes
were not found to “breed true” per se. For instance, Smalley et al. (2000) concluded that
inattention and hyperactivity seem to share the same familial underpinning, and do not

have unique effects. Alternatively, it has been suggested that subtypes based on

61

symptom expression Mbeen shown to cluster in families (Todd, et al., 2001). This
suggests that differential expressions of ADHD could be transmitted in families and
exploring possible mechanisms for symptom transmission becomes essential. One area
that has been neglected in past research has been gender.

In prior studies mothers and fathers were not directly interviewed, instead all
relatives were assessed using a questionnaire completed by the mother of the child.
Further, neither of the previous studies reported results for mothers and fathers separately
or for child boys and girls separately. Contrary to past studies, the present study assessed
parental ADHD using a structured clinical interview with each parent and directly
assessed gender effects for family transmission. When considering mothers and fathers,
and boys and girls separately, gender differently predicted parental rates of ADHD (any
type) in boys and girls with ADI-1D. Specifically girls with ADHD had mothers with
higher rates of ADHD, whereas boys with ADHD have fathers with higher rates of
ADHD themselves.

Gender and DSM-IV ADI-ID Subtypes. When boys and girls or mothers and
fathers were combined, the subtypes did not appear to be distinct disorders. However, as
was true with ADHD (any type), subtype transmission appeared to be gender specific. In
general, there was a relationship between child subtype and ADHD status in the parents.
Girls with ADHD combined type compared to control girls were more likely to have
mothers with ADHD (any type), there was no significant relationship between girls
ADHD subtypes and father ADHD status. The opposite was true for boys. Boys with
ADHD-Inattentive type and boys with ADHD-Combined type were both more likely to

have a father with ADHD compared with fathers of control boys. Also, boys with

62

combined type were more likely to have a father with ADHD, when compared with boys
with ADHD-Inattentive type. There was no significant relationship between a diagnosis
of ADHD subtype in boys and mother ADHD status.

Further, fathers of boys with ADHD-Combined type were more likely to have
. ADHD-Combined type versus both the fathers of controls and fathers of boys with
ADHD-Inattentive type. Fathers of boys with ADHD-Inattentive type were more likely
to have Inattentive type themselves versus fathers of the other two child groups. This
was not true for mothers of boys. Thus, subtypes did appear to be validated with boys
and fathers, but not girls and mothers.

Results suggested that the ADHD subtypes directly transfer only within boys.
This was clear when looking at boys and fathers; results for girls and their mothers were
less clear. Perhaps one reason for this was the low number of girls with ADHD-
Combined type (n = 12). However, beyond the sample’s limitations, the results could also
hint at important gender differences in the manifestation of ADI-ID. Previous studies
have found that girls with ADHI) have a greater chance of having a parent with a
diagnosis of ADHD (Smalley, et al., 2000) and other studies have suggested that girls
with ADHD require a greater loading of familial influences to develop ADI-11). Thus it
may be that symptoms of ADHD in boys are present with genetic loading of ADHD,
whereas for girls, genetic interaction with environmental factors is more crucial for
symptom expression. Therefore to examine familial transmission in families of girls with
ADI-II) would require including measures of environmental risk that may interact with

symptom expression.

63

 

Additionally, differential familial transmission for boys vs. girls also may be a
result of the lack of research with girls with ADHD. DSM-IV field trials were done with
primarily male samples. It could be that the subtype differentiation could be more valid
with boys than with girls. This could in part explain why gender-specific transmission
was found with mothers and girls for ADHD-any type, but not for ADHD subtypes. As
found in Todd et al., it may be beneficial to organize symptoms in alternative ways,
especially when examining familial transmission in girls.

Overall, this gender-specific transmission pattern requires replication, but is
provocative. Unlike past studies that have merely found correlations between parent and
child ADI-ID, gender-specific transmission hints at possible etiological pathways. For
example it could be consistent with some emerging results suggesting that molecular
genetic correlates of ADI-1D arebetter understood when gender-specific transmission is
considered (Fisher et al., 2002). Also, assuming non-sex linked genetic markers,
differential familial transmission could be a result of social modeling; perhaps the girls
are more likely to model the behavior after their mothers whereas the boys are more
likely to model their behaviors after their fathers. In addition to suggesting etiological
mechanisms, gender-specific familial transmission could help understand past
inconsistencies in the literature. However, it is important to determine potential
confounding factors in the pattern of familial transmission.

Limitations. Although one of the strengths of the present study was the use of a
structured interview with each of the parents, interviews were designed to be
administered using lay interviewers. Although this interview is appropriate and

acceptable for the community-recruited sample we studied, a clinician-guided interview,

64

such as the SCID-IV-TR (First, Spitzer, Gibbon, & Williams, 2001) might provide more
sensitive and valid diagnostic assignments and could alter some findings if ADHD was
over or under diagnosed in the present study. Parent ADHD also relied entirely on self
report; it would be ideal to have observer and childhood observer ratings of parents own
behaviors to assess their ADHD status.

Another possible limitation to results was the use of the QDIS-III-R for a subset
of parents. The QDIS-III-R, based on DSM-III-R criteria, was used to diagnose non-
ADHD disorders in some parents who did not complete the DIS-IV. Although agreement
between the two versions demonstrated reasonable agreement to warrant pooling the data
for this study, the inconsistency between diagnostic interviews could have influenced
interpretation of associations between child subtypes and parental non-ADHD disorders
in ways that are indeterminate. .

Finally, although a strength of the present study was that both boys and girls were
included, when groups were stratified by both parent and child sex and ADHD subtype,
the number of girls in some cells was very small, especially in the ADHD combined type
group. As previously mentioned, unclear gender subtype validity with the parents of girls
may have been a result of this unequal distribution between subtypes.

Conclusions. Overall, this study made several significant contributions to the
ADHD family studies literature. Overall, this was the first study to do a familial
transmission study of prospectively defrned DSM-IV ADHD in a community sample of
children. This study replicated and extended previous studies, in that this study examined
mothers and fathers separately and found gender specific ADHD transmission in families.

In addition, this study also corrected for short-comings of previous family studies of

65

ADI-II). The current study used boys and girls diagnosed with ADHD subtypes while
controlling for comorbid behavior disorders in children. Results showed that parental
ADHD is related to ADI-ID in children, regardless of child comorbidity. It also revealed
intriguing hints of gender specific transmission of ADHD in families. Therefore this
study underscores the importance of future research that explores in more depth gender
based theories of family effects on ADI-ID symptoms in children. Overall, specific
patterns of ADHI) subtype transmission within families raises further implications for
both genetic and psychosocial transmission processes related to gender and ADI-II)

subtypes.

 

66

 

‘ Although there are 3 ADHD subtypes as defined in the DSM-IV, recent literature has hypothesized that

the hyperactive subtype is a precursor to the combined subtype (Hard, et al., 1994). By the time children

reach school age, there is a relatively low prevalence of the hyperactive type. For these reasons, only two
DSM-IV subtypes (Inattentive and Combined) will be considered here and in the final analyses.

2 The use of 5 symptoms, rather than 6, has been shown to be acceptable to use with adult populations
(Barkley, 1998).

67

Appendix A

Differentmodels and criteria have been proposed for establishing the validity of a
medical and psychiatric illness (e. g. Cronbach & Meehl, 1955; F eighner et al., 1972;
Cantwell & Baker, 1988). One approach to the validation of psychiatric disorders arises
from the medical model tradition in classification (F eighner et al., 1972). This approach
is often traced to Emil Kraeplin’s classic observations of schizophrenia. Early in the 20th
century Kraeplin advocated that mental disorders be identified through the grouping of
people with similar observable behaviors and symptoms. Once a clinical description of l
the disorders was established, he believed that the cause of the disorder could be gleaned
from the study of this group of people with the particular syndrome. Kraeplin, following
the medical tradition, believed the cause might be traced to a single gene or pathogen.
Although that approach was out. of favor for much of the next half century, it was
renewed with the advent of the contemporary DSM approach. Its renewal was directed in
part when Feighner et al. (1972) published an article that updated the medical model
approach to include five phases of validating diagnostic categories: clinical description
(observable behaviors, symptoms and other associated factors), laboratory studies, the
addition of criteria excluding overlap of other syndromes, follow-up studies of syndrome
course, and family prevalence studies. That article set the stage for the DSM-III, in
which criteria for diagnoses were revised fiom the purely theory-based descriptions to the
directly observable symptom descriptions that are found in the DSM today. The medical
model assumes that once a homogenous group of people with the same syndrome is
isolated, the cause will be more easily located; this cause was believed by some to be

biogenetic and found in one gene or virus/bacteria, such as true with most physical

68

disorders. However most contemporary scientists have discarded this metabolic view of
psychopathology. It is believed that there are several different contributing etiological
factors (Depue & Monroe, 1991). Nevertheless family prevalence studies may provide
clues as to whether these etiological agents are present in systematic ways within

families.

69

 

>325... w

Goa—Elan :8 065 8 =5 U—m-~<
m8. 653288 .5 92.2.2. 5.9.22.8

 

 

 

 

 

 

 

 

 

 

$2.2: 982.2. e\e $32.8 SE. e\e $32.8 5.5. 03-9.5...» $
Emoaon 3.. 055 $822. 3.. U—m-_<
02.32.85. >853. 59x. :.Ae\e xofiv ”.05 .wem
982.2 8%.
32335 0.x. 2.x. me u H. see
cam $82.88 m. E. mesa x“... n .8 i.
>27mo2m. $28.55.. SEX. 3.1x. XNA: n .3 .w 3
982.2 39.9
cognac: Nu._e\e 36.x. xofiv H mm .m3
285 98338 3%.. . 2.x. xsE n .3 .3.
>3 areas 8.x. 3.3.. xx: u .8 .30
H08. 5:352 om ><n n .ow AmUubov ><o n .mw C .9: m M So .Nom

98308...

 

 

 

 

 

 

70

REFERENCES

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

 

Achenbach, T.M. (1991b). Marml for the Teacher’s Report Form Jand 1991
Profile. Burlington, VT: University of Vermont Department of Psychiatry.

American Psychiatric Association. (1952). Diagnostic and stagisticalfimanual of
mental disorders (1" ed.) Washington, DC: Author.

American Psychiatric Association. (1967). Diagnostic and statistical manual of
mental disorders (2n ed.). Washington, DC: Author.

American Psychiatric Association. (1980). Diagnostic and statistical manual of
mental disorders (3r ed.). Washington, DC: Author.

American Ps chiatric Association. (1987). Diagnostic and statistical manual of
mental disorders (3r ed.- revised). Washington, DC: Author.

American Psychiatric Association. (1994). Diagnostic and statistitl manual of
mental disorders (4th ed.) Washington, DC: Author.

August, G. J., & Stewart, M. A. (1983). Familial subtypes of childhood
hyperactivity. Journal of Nervous and Mental Disease, 171, 362-368.

Barkley, R. A. (1998). Comorbid disorders, social relations, and subtyping. In
R. A. Barkley (Ed). Attention-Deficit Hyperactivity Disorder (2"d ed.). New York: The
Guilford Press.

Barkley, R.A., DuPaul, G.J., & McMurray, MB. (1990). A comprehensive
evaluation of attention deficit disorder with and without hyperactivity. Journal of
Consulting and Clinical Psychology, 58, 775-789.

Barkley, R. A., Guevremont, D. C ., Anastopoulos, A. D., DuPaul, G. J ., &
Shelton, T. L. (1993). Driving-Related risks and outcomes of Attention Deficit
Hyperactivity Disorder in Adolescents and young adults: A 3- to 5-year follow-up
survey. Pediatrics. 92, 212-218.

Biederman, J., Milberger, S., Faraone, S. V., Kiely, K., Guite, 1., Mick, E.,
Ablon, J. S., Warburotn, R., Reed, E., & Davis, S. G. (1995). Impact of adversity on
fimctioning and comorbidity in children with attention-deficit hyperactivity disorder.
Jouml of the American Academy of Child and Adolescent Psychiatry, 34, 1495-1503.

71

Biederman, J., Newcorn, J ., & Sprich, S. (1991). Comorbidity of attention
deficit hyperactivity disorder with conduct, depressive, anxiety, and other disorders.
- American Journal of Psychiatry, 148. 564-577.

Breen, M. (1989). Cognitive and behavioral differences in ADHD boys and
girls. Journal of Child Psychology Psychiatryyand Allied Disciplines, 30, 711-716.

Bucholz, K. K., Marion, S. L., Shayka, J. J., Marcus, S. C., & Robins, L. N.
(1996). A short computer interview for obtaining psychiatric diagnoses. Psychiatric
Services, 47, 293-297.

Cantwell, DP. (1972). Psychiatric illness in the families of hyperactive children.
Archive of General Psychiatry, 27. 414-417.

Cantwell, DP. (1975). Genetics of Hyperactivity. Joumjal of Child Psychology
and PsychiatryJS, 261-264.

 

Cantwell, D. P. (1997). The scientific study of child and adolsecent
psychopathology: The attention deficit disorder syndrome. Journal of the American
Academy of Child and Adolescent Psychiziry, 36, 1033-1035.

Cantwell, D.P. & Baker, L. (1988). Issues in the classification of child and
adolescent psychopathology. Journal of the American Academy of Child and
Adolescent Psychiatry, 27, 521-533.

Conner, C. K. (1997). Conners Rating Scales-Revised. Toronto: Multi-Health
Systems, Inc.

Cronbach, L. J. & Meehl, P. E. (1955). Construct validity in psychological tests.
Psychological Bulletin, 52, 281-302.

DuPaul, G. J ., Power, T. J ., Anastopolous, A. D., & Reid, R. (1998). ADHD
Rating Scale IV: Checklists, Norms, & Clinical Interpretation.

 

Depue, R. A. & Monroe, S. M. (1991). Psychopathology research. In M. Hersen,
A. E. Kazdin, & A. S. Bellack (Eds) The Clinical Psychology Handbook (2"d Ed;.
New York: Pergamon Press.

Erdman, H.P., Klein, M.H., Greist, J .H., Skare, S. S., Husted, J.J., Robins, L.N.,
Helzer, J. E., Goldring, E., Hamburger, M., & Miller, JP. (1992). A comparison of two

computer-administered versions of the NIMH diagnostic interview schedule. Journal of
Psychiatric Research, 26, 85-95.

Faraone, S. V., Biederman, J ., Jetton, J. G., & Tsuang, M. T. (1997). Attention

deficit disorder and conduct disorder: longitudinal evidence for a familial subtype.
Psychological Medicine, 27, 291-300.

72

Faraone, S.V., Biederman, J ., & Friedman, D. (2000a). Validity of DSM-IV
subtypes of attention-deficit/hyperactivity disorder; A family study perspective. Journal
of the American Academy of Child raid Adolescent Psychiatg, 39, 300-309.

Faraone, S. V., Biederman, J ., Mick, E., Williamson, S., Wilens, T., Spencer, T.,
Weber, W., Jetton, J ., Kraus, I., Pert, J ., & Zallen, B. (2000b). Family study of girls
with attention deficit hyperactivity disorder. American Jouml of Psychiatry, 157,
1077-1083.

Faraone, S.V., Biederman, J ., Weber W., & Russell, R. L., (1998). Psychiatric,
neuropsychological and psychosocial features of DSM-IV subtypes of attention-

deficit/hyperactivity disorder: Results from a clinically referred sample. Journal of the
American Academy of Child and Adolescent Psychiatry, 37, 185-193.

 

Feighner, J.P., Robins, E., Guze, S.B., Woodruff, R.A., Winokur, G., & Munoz,
R. (1972). Diagnostic criteria for use in psychiatric research. Archives of General
Psychiatry, 16, 57-63.

Fisher, S.E., Francks, C., McCracken, J .T., McGough, J .J ., Marlow, A.J.,
MacPhie I.L., Newbury, D.F., Crawford, L.R., Palmer, C.G., Woodward, J .A.,
DelHomme, M., Cantwell, D.P., Nelson, S.F., Monaco, A.P., & Smalley, S.L. (2002). A
genome-wide scan for loci involved in attention-deficit/hyperactivity disorder.
American Journal of Human Genetics, 70, in press.

Frick, P. J. (1994) Family dysfunction and the disruptive behavior disorders: A
review of recent empirical findings. In TH. Ollendick & R. J. Prinz (Eds) Advances in
Clinical Child Psychology, 16. New York: Plenum Press.

 

 

Frick, P. J., Lahey, B. B., Christ, M. A. G., Loeber, R., & Green, S. (1991).
History of childhood behavior problems in biological relatives of boys with attention-

deficit hyperactivity disorder and conduct disorder. Journal of Clinical Child
Paychology, 20, 445-451.

Frick, P. J ., Lahey, B. B., Loeber, R., Stouthamer-Loeber, M., Christ, M. A. G.,
& Hanson, K. (1992). Familial risk factors to oppositional defiant disorder and conduct
disorder: Parental psychopathology and maternal parenting. Journal of Consultingm
Clinical Psychology, 60, 49-55.

Gomez, R. & Sanson, A. V. (1993). Mother-child interactions and
noncompliance in hyperactive boys with and without conduct problems. Journal of
Child Psychologyand Psychiatry, 35, 477-490.

Goodman, S.H., Lahey, B.B., Fielding, 3, Dulcan, M., Narrow, W., & Regier,

D. (1997). Representativeness of clinical samples of youths with mental disorders: A
preliminary population-based study. Journal of Abnormal Psychology, 106, 3-14.

73

Hinshaw, SP. (1992). Extemalizing behavior problems and academic
underachievement in childhood and adolescence: Causal relationships and underlying
mechanisms. Psychology Bulletin, 111, 127-155.

Hinshaw, S. P. (1987). On distinction between attentional deficits/hyperactivity
and conduct problems/aggression in child psychopathology. Psychology Bulletin, 101,
443-463.

Hinshaw, SP. (2001). Is the inattentive type of ADHD a separate disorder: A
comment on Milich, Balentine, & Lynam. Clinical Psychology: Science and Practice,
8, 498-501.

Houghton, 8., Douglas, G., West, J ., Whiting, K., Wall, M., Langsford, S.,
Powelll, L., & Carroll, A. (1999). Differential patterns of executive function in children
with attention-deficit hyperactivity disorder according to gender and subtype. Journal
of Child Neurology, 14, 801-805.

 

Kuhne, M., Schachar, R., & Tannock, R. (1997). Impact of comorbid
oppositional or conduct problems on attention-deficit hyperactivity disorder. Journal of
the American Academy of Child and Adolescent Psychiatry, 36, 1715-1725.

Lahey, B. B., Applegate, B., McBurnett, K., Biederman, J ., Greenhill, L., Hynd,
G. W., Barkley, R. A., Newcorn, J ., Jensen, P., Richters, J ., Garfinkel, B., Kerdyk, L.,
Frick, P. J ., Ollendick, T., Perez, D., Hart, E. L., Waldman, I., & Shaffer, D. (1994).
DSM-IV field trials for Attention Deficit Hyperactivity Disorder in children and
adolescents. American Journal of Psychiatry, 151, 1673-1685.

Lahey, B. B., Piacentini, J. C ., McBurnett, K., Stone, P., Hartdagen, S., & Hynd,
G. (1988). Psychopathology in the Parents of Children with Conduct Disorder and
Hyperactivity. American Academy of Child and Adolescent Psychiatry,27. 163-170.

Lalonde, J., Turgay, A., & Hudson, J. I. (1998). Attention-deficity hyperactivity
disorder subtypes and comorbid disruptive behaviour disorders in a child and adolescent
mental health clinic. Canadian Journal of Psychiatry, 43, 623-628.

Lahey, B. B. & Loeber, R. (1996). Attention-deficit/hyperactivity disorder,
oppositional defiant disorder, conduct disorder, and adult antisocial behavior: A lifespan
perspective. In D. M. Stoff & J. Breiling (Eds). Handbook of Antisociflehavior. 51-
59.

Milich, R., Balentine, A.C., & Lynam, DR. (2001). ADHI) combined type and
ADHD predominately inattentive type are distinct and unrelated disorders. Clinical

Psychoggy: Science and Practice,8, 463-488.

Morrison, J. R. & Stewart, M. A. (1971). A family study of the hyperactive
child syndrome. Biological Psychology, 3, 189-195.

74

Nigg, J .T., Blaskey, L.G., Huang-Pollock, C.L., & Rappley, MD. (in press).
Neuropsychological executive functions and ADHD DSM-IV subtypes. Journal of the
American Academy of Child and Adolescent Psychiatga

Nigg, J .T. & Hinshaw, SP. (1998). Parent personality traits and
psychopathology associated with antisocial behaviors in childhood attention-deficit
hyperactivity disorder. Journal of Child Psychology_and Psychiatry, 31, 145-159.

Reynolds, C.R., & Kamphaus, R.W. (1992). Behavioral assesment system for
children manual. Circle Pines, MN: American Guidance Service.

Robins, L. N., Cottler, L. B., Bucholz, K. K., Comptom, W. M., North, C. S. &
Rourke, K. M. (1995). The Diagnostic Interview for DSM-IV (DIS-IV). St. Louis:
Washington University.

Robins, L. N., Helzer, J .E., Croughan, J ., & Ratcliff, KS. (1981). National
institute of mental health diagnostic interview schedule: Its history, charactoristics, and
validity. Archives of General Psychiatry, 38, 381-390.

Schachar, R. J. & Wachsmuth, R. (1990). Oppositional disorder in children: a
validation study comparing conduct disorder, oppositional disorder and normal control
children. Journal ofChild Psychology and Psychiatry, 31, 1089-1102.

Schwab-Stone, M.E., Shaffer, D., Dulcan, M.K., Jensen, P.S., Fisher, P., Bird,
H.R., Goodman, S.H., Lahey, B.B., Lichtman, J H Canino, G., Rubio-Stipec, M., &
Rae, DS (1996). Criterion Validity of the NIMH Diagnostic Interview Schedule for
Children Version 2.3 (DISC-2.3). Journal of the American Academy of Child and
Adolescent Psychiatry, 35, 878-888.

Seidman, L.J., Biederman, J, Faraone, S.V., Milberger, 8., Norman, D., Seiverd,
K., Benedict, K., Guite, J ., Mick, 13., & Kiely, K. (1995). Effects of family history and
comorbidity on the neuorpsychological performance of children with ADHI):
Preliminary findings. Journal of the American Academy of Child and Adolescent
Psychiatry, 34, 1015-1024.

 

Shaffer, D., Fisher, P., & Lucas, C. (1997). NIMH Diagnostic Interview
Schedule for Children-IV. New York: Ruane Center for Early Diagnosis, Division of
Child Psychiatry, Columbia University.

 

Sherman, D.K., Iacono, W.G., & McGue, MK. (1997). Attention-deficit
hyperactivity disorder: A twin study of inattention and impulsivity-hyperactivity.
Journal of the American Academy of Child flid Adolescent Psychiatry, 36, 745-753.

Stewart, M.A., deBlois, S., & Cummings, C. (1980). Psychiatric disorder in the
parents of hyperactive boys and those with conduct disorder. Journal of Child

Psychology and Psychiatfl, 21, 263-291.

75

Szatmari, P. (1992). The epidemiology of attention-deficit hyperactivity
disorders. In G. Weiss (Ed), Child and adolescent psychiatric clinics of North
America: Attention deficit hyperactivity disorder (pp.361-3 72). Philadelphia: Saunders

Tannock, R. ( 1998). Attention deficit hyperactivity disorder: Advances in
cognitive, neurobiological, and genetic research. Journal of Child Psychology and
Psychiatry, 39. 65-99.

Taylor, E., Sandberg, S., Thorley, G., & Giles, S. (1991). Hyperkinetic disorder:
Prevalence, definition, and associations. In G. Russell, B. H. Anderton, P. Williams, &
S. E. Hague (Eds) Maudsley Monographs, 33: The Epidemiology of Childhood
Hyperactivity Oxford: Oxford University Press.

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

76

H.020 .

$03.... $ 0.10.0.0 9.5.00 0.. >UHU 9:00 .3:

 

 

 

 

 

 

 

 

 

 

 

max... GEE 20800.... 2 e\e 95.0.0 $2.0... G052... 8.560 $100.20 $25.9
$82.2. H00. 2.0.0 85.00 3.24.0: E .832.
a E832...
$0.03..
.2038: .360? Ue.-$0..0... .8 00.x. . €808.00 20 20 3.60.083... 50.. 00 8.0.0..
an 9020.. 00.7.0 E88... .0 $2.030. 0.00.5.2...
30.... GEE , 3.0.05.0. 0.... 80.2.0.3.
359.030
N. G820: 300.. $0.0... .8 .ooe\e . C8083. 20 20 $0.00.... 0.. 0E.....0.. 2....
A .38 00.70 3.8... 30080.33. 20.0 30.0 ES...
GEE .0 .30 360.0030 053.030
$2.830
u. mg 0. .300..- .0608... .m0 .oo..\e . 9.0.0.0.. 20 20 GEE £000.00. .5. >U...U.
0.. Como. 00.7.3. E0800... . 00.0.an.0.. £00 0.80.0.2. as...
8.0.5.0800. E8302 .0
00.0....
0. E8. 0. >U...U m. Umg mm fies. . m>Um <00 20 GEES: as... >U...U $.00 GU
0.. :83 GU U...” 20.0 0.0.0 ES... .0 .30 0..
0380.0. 00.0... ...0.. 0E....0..
2.... >U..5 0.. GU 0.000.
m. $10.. 0. >U...U an Umz .3 BOX. . 000300038 <00 20 . GEEB: SE. >Um..U. GU. o.
0.. Com... GU U.-.” >U...U {GU 20.0 32.0 ES...
.0 ..0<0 00.0.... 8.... ...0 8.30
E882. 0m E08002. E ...0
0EE..0...
0. >U...U UmZ ~00 .ooe\e . m>Um 20 20 .....0 0.5.00. 0. 30.030.
90003.0: 5-x E08008 200 88.0. .0
0. 0.. 38m. 0EE.0.. 2.... >Um.U.

 

 

 

 

 

 

 

 

77

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.. .9000. >U...U. Um... 0. .00.. mGr..w..0o <00 20 3090.0 0.. 0EE..00, 8....

8.50.5... .0. OUU. 0.0 E.” >U...U 0.00 GU 500.80

000000.. GU ..0<.0m 0505000.... 30.0

:00... 00..0..0.0m.00. 0.0.0.030 ...00
>9... 0.000.

0. 00.0000 >U...U. Um: Mme .oo.\o mG.U <00 20 0000.50 0.. 0.0.9.00 3....

0. 0.. 3003 UUU. 0.. E-” >U...U 0.00 GU .5<0

GU 0505000.... Em..0.. 00.00 0..

.00... 0.8500 0.00 >U...U 0.
GU 0.000.

0. 2.mm R >U...U. Um: IN .8... «<96 m. <00 20 7.0.00.0 0.. >U...U 0EE..00

2.00.53 OUU. 0.0 E-” a U5...” .50 Em..0. 00.00 0..

. .008 GU Um??? 000500.000. 0090.0 .50 30.0
>U..HU. 7.0.00.0 0..
>U...U+GU 000.0 30.0 ES...
.0 .5<0 >U...U. 00.0000.
00.00.... 0000.00 0000000000.

, 00.. >m0U.

.0. 00.0000. >U...U Umg-.< ~00 30.x. . mG.U....-.~ 20 <00 ”0.0.500 0.. >U...U 0.0.0.00

9000.300. >003..- 000.0 30.0 5.0... .0 ..0<0

@ 00.00050 .0550 >U...U. .00. .. $.00 00.

©0000. 8.00.0.2. 2.... 0030 00.....00
00 0.0m00000 .0 0EE.00.

... 00.0000 >U...U Um??? NS 0.... mG.U....-.~ 20 <00 ”0.0.500 0.. m...0 8.... >U...U

0. 0.. >008..- 00.0 Em..0. 00.00 0.. >U...U

©0000. 0.5.0.. 0.0300500. .0... .. .0 00. 5.0.0..
.0 00.....00 00 0.080000 .0
30.0.

 

 

_. .u G..0.0 303.80
Nu G03300.... 0003.50

78

55.0 N
wags—ES 3. tummy m: $5.88 2:— worzmfia a. 02:3.“ 8512.2— 8

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5:25 3. GEE—.2. i=— >UEv. is. ©5906.
mafia 055 2955233: mmmimnwi
02:3. . >UMU .. Bulb 2% CU a. ”in“ on gmflv :— w A .3
52.2.8 3.
>Umndnno3no—
nun—.03, 39:52. mun—51m 30:52.. 5.283 Earn; fins—2.». 392.03 3.253 23:83
9526: :38 N..\.. 0.x. 3.x 5.x. - - w; 90 UVm 2m
maow on a. 3005 5.x. _ 1x. 3.x. Ame; Amfo 3X. :3 “I: Fov Pa Fov Pa
2mmm mm 3398: oak fix. 3.x. ix. 5.x. 5.x. 5; Hum; ch. o Zm
233
3:53 2 3. Good >= 5.3:: >= 3.3:: >= 5.8:: >= 5.3:; >= 5.3:;
_.. man—37$ ”£332 Han—ago“ man—anram ”£338
mo\o Ex. 3ng 9x. “9.x. 8.x. w L A. m; OVVVm on
38030 9 a. 388 >= awn—“:76“ 0+ EH, 1
.I.
fix. max. 5.x. 3.x. - - PB: Ov $3.1 v m

 

 

 

 

 

 

 

 

79

 

 

.. 3558 Q a. Cwod a5 :3 98850 30:52 85 $an Enougqonzw. .93 28 50:53 m: 8596», 3.25 Econ—am 5 So

mum—«mam.
imamosa 3 a 688 3958 w: 3533 m: m aim—o mamEam.
++ 0 n Oochsoa 9:569 m; u $558035 365.33 98369 H n Hanna—:20 mccceo.

N38 3. Huang??? 9.82.03 A: 5.358 on noise—u 853—6.. 8
Hun—6:3 3. GEE—.2. £5. Bug 2:— 5.3.5 o». GEE—.2. i=— >UNU+OU

Haw—o u

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5:63 Ema—don wag. 02:3. >UIU >UMU+OU
man—.2. 33:2. $.32. 33.5. 3:52. 2332.
68583: A. 39 39* A: Sb... wo.Aw*
o. 0 Am... G 8* me...
00:95 Unmoaoa w. A* A* m... w* 3*
>389”: $30333 _.+ o 0 mm o -
98:5 N.+ m o W., o -
A. Kg 0 Mo 0 mN.A* mu?
m. on, Q.“ 0* 0... N1.
0. _A m A o A
28:2 >95» A. NAA o wub mu -
N. u o 3 A... - -
A. _o 0 m6 m6 gas, AA
0. Nu no 3 AA A A
Dam >_u:mo A. 9Q o o 90 Nw.m* AA
0. 3* m o... A 0
H08. 2:362 0». A. Afim wad SQ AWN -
9830; N. S o 3* :w... -

 

 

 

 

 

 

 

 

 

.._ M mmmammowa Emmaoaoo €A.o$ c328: mac?
+ n 3 RE? A. ea N. 90 89823 28 5o Emmsoam 88.

80

How—o A

 

 

 

 

 

 

 

 

 

 

 

 

 

 

UoBomSuEOm
033 U. 85 S u 3.3 5.03%
083: Bum—035$” Bum—0.398 >UIU+OUUBU >35; >35- n 0.8%
9:53: .

2 A0 0m Am Aa 3 am -
>3. wmo 5 33m 05» 0.A~ 0:3 0.3 0.w 0.5a bum
$603 0.5 35 .2; .2: w: no.0 .30...
$838 and mmh . «9A w0.q mob 0_.q ._00

>3. 5 8.50 898 SubA 895.». A300 _oAuN .umn

>3. A o». #5333 mx. .3 q. 5 90A .5; 0.8 «pm boo.
>3. A om=§o§3 mx. .3 ubA 93 95 ~50 93 boo...
2093..

2 A0 00 Au AN 5w 2 -
>3. own um.~A 3.3 wwbm 3.5 .00.“: 090A .30
$238 3m m0.m 3.0 0PM 0~.w «a; .Amm

>3. 5 80.00 Scum 20.8 30.3 :93 80.3 .30
.x. .338 00.0 .2 .a .2 .0 3.4 3.0 3.0 .80...
>3. A E Oman—=83: 5.3 5.8 55¢ 5A4 5.3 5.8 .muo
>5. A :5. 96.88. 93 93 Md 9%. A. : 98 .30

3an

2 3 .3 3. N0 8 a -
>3. wmo Ao.AA Ac. 5 A98 250 3.x 0.0.3. .mAN
$238 2.0 m0.m mum 0mg 00.0 0P0 .500

>3. 5 SW05 25.0w _5.u0 :on :o.A0 :o.A0 .58
o\.. 32.18 Bob 3.0 «mm «So 3;». mum .80...
>3. A «8 3.39083: 5.3 800 0.0o 5.5m 85a 35a .3;
>5 A 30. 96.30; N 0.8 #3 So :5 s8 s8 .8”

 

 

_ a3 «38 v.83 8538 08.3: as .633 Ex.
N 9:3: 39 23.5 963 Eamon 38 23 95.33 88 a». «3:3 man Enos». 85.3% 93$ 0: racopramu A: Eamon Oran—.8 £9 Bum—u ammon— on
85035 3388 E .358 958: 23. >35 38a on Um??? 9.368. m8 8% m3 935 0538:?
u >3. A Bo. mat—38$.“ Banana 2588 o» 3935 9»: So 0033 3% 3.28.8 m2 9: 83 A38. 8 9. «.85 9m: 3 :95 02. £85 5 can <85

81

H.923 m

MESA—.8». A” Goa—515E >55 A95 .5515. 02:37

 

 

 

 

 

 

 

583 Egg; 02:3— 9:53: DEER: 259.5 9.3. 330 5 Ana—3 macaw 5v
3 u A8 A: u 30 03-358 _ NEE 73:8
>UEU>5 Q03
.25 08.339... AZ n mov AZ u 30V
3.x. Bax. xx: u :8 .02... .88...
53583.. 95 n A00 95 u 08
3.x. 590$ x50 u 98 .S 1. 8%..
59533... 95 n 5.3 A2 u 00v
MAX. 555$ 0% CV." 5.3 .059. .30...
223965 382.33
A: 08.85.. S. n m5 A2 n HAS
0 E8803 Sufi. A0808
382.2 3.5.x. 8.5.x. . x” as .... Na .8 .E
5 amen—o; 5.5.x. 5A. E.
52583.. 3. "A3 A2 umAv
0&8..qu aA.A.x. SAX.
33082. 590$ 55.x.x. X» 50 H Aha .: _ .08
5 E8803 $0.8 5w.m..\o
5.18:3“... A2 u 53 95 u md
o 38:33 3.8.x. 3.8.x.
3302—32.. 85:. 505:. an A50 u 5.3 .8 .5
5 E8808 505$ 5A.o..\o

 

 

 

 

_ 5x5 names—Rm m3 20?ng 388on 8333 o $8303. A $882. ca 5 on :53 E8832. 5 38:8 3. 035 B96.

82

Haw—o a
Macon—.8!“ a“ Goa—.915 tum—v A35. any in. 6256.»

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A91... 2:— wcwdv
533 9833 98.96 35$ wowm p-396 93$
A 03.358 OE-ma:m8
0033— >UmU XNCV n ML 0253— >UEU xunC u Mn.
>35 33. 3.3
. ANN wags? Q< n uwv AZ unov A2 "mu v AZ NH 5
_ 91x. Sufi. 9mm .Om— w.m..\o 8.9x. mbw boa...
32:3,... A2 u n8 AZ u wk: A2 .1. n3 3 n o3
0.x. uoax. Ago .80... 99x. Six. N3 .Zu.
fiaSma... A2 n 5v 3. n 3v A2 n N3 A2 u mwv
3.5.x. 8.0.x. .3 .mNm 9x. 56.x. 93 .S 1.
$8 u N am XNE u u at
ZSEPUEV 383a:
ANN $335... AZ “3 v AZ "mg A2 "#3 AZ "53
o “:8an flux. 3.9x. 3.4:. mumfi.
Emmoaoq flux. 36.x. Bum .ma 3.9x. N553 WK .3
N .3833 39.x. wodfo 8.5.x. 5.0.x.
303mg. AZ u _8 A2 n NNV A2 unov A2 ”as
c 98303 oubx. 8.9x. 3x3: mu.~..\°
Emmoaq 36.x. She; Pd .8 ~91; aux. CG .3
N amenaoa who; u _ ax. Safe B .o.\..
NuaSmJ... A2 .1. 73 A2 u EV AZ n MB 3. n 3V
0 ammo—dog wagon. 3L9} awhfi. 3.0.x.
ammo—as. Nix. mo.oo\o mum .oa... Six. 39.x. NS .3
N E833 3.1x. 5.9x. 0.1x. Nuuf.
{8.5.8. .

g uxm 05-353... men 2972va magma; 8338 o Egan? _ 98am: 9. N 2. 803 E8303 5 338.0. 3. 035 macaw.

83

fine—o q

5.96.2.3 3. $2.03 23.56% 2332. m: GEE—.2. 1.5. >6: 8:52.2— 15. 02:3:

 

 

 

 

 

38:» awn—Macaw 0033. GEES: GEES: :1ng $3 $.ro 2.» CV u accoAa—oa h

EN $335. I AZ Had Q4 "$3

996 mafia _w.m°\o ~63 .85
ECU BAX. 53> .qom .52
Eco—8 Swab 85$ 98. .mmo
63353» 9x. max. Hum .Nnm
OUQOU 50.x. 36.x. m8 .Aww
3500mm. vb 98x. flux. :0 .qu
9cm 33338 Fox. 8.4.x. New .8.“
28:2 33388 80.x. 59.x. .Ba .93
30:83. I Az u 3v AZ n wav

Q>U ago; 89$ 93o .Omo
KOO 39.x. VCR. n. SN .73
Eve—E 91x. 56.x. 5qu .wa
Sagan o..\.. u .m..\.. _ .8 . Bu
OUUBU N~.m..\o woéx. b3 .34
Eamoam. Eu ~.No\o Suck. fifim hue
Dam agagoo ~.~..\o 8.0:. N84 .Hoa
Boog— qouoaagoo mix. g _ .oo\.. o: .53
wag; . AZ n wuv AZ u m3

0% mafia Max. .Sm .moa
ECU 3.0.x. SEX. :3» an:
Eco—ma H H. 1x. Muck .3 .Auq
Suzi—am 9x. 9x. - -
OUQOU 39x. 3.9x. .ooN boa
$5383». Eu 03.x. Sufi. .oao .moq
Cam 33888 92x. Safe .Aoq .53
2852 332.38 3.1x. . 35.x. box 9%

 

 

 

 

 

 

84

 

 

 

 

Haw—n a
flags—.8”. u” non—calumgmu 13. >53 + 63:05
383 Ema—don 0033— 0358: 0358: 232.6 GEES: w arch—-mnocw E
9:88 2\>Uma+OUU\OU Giana—m8 $358 78:8

>29
éwfiaaau A2 A5 Sum: E 3:
>25 :5 £3 93... sax... 8.3.... x” E n fig .8... .8».
>U§é§§<a wax... 5.9x... 93.. x” E u i: .23 .03...
égéaasa o3. 3.x... 5.4.x..." X» E u 5.3. .08... .2: ..
3°33- 3 £3 E A: 2 ME
>26 93 2.3 9%.. 3.9x... “3.x... x» E u do .8” .80
gauémnoaza N. 1x. 5%.. q. _x x“ E u N8 BM .5
Egéasaa o3. 8.x. 3.x... x” E n mg .03 8%
WE‘RE. - A2 u u& Q< u #3 A2 u n8 N
>26 93 2.3 max... Box... 8.3... x E u in .:o So
55:33:? 3.x. 3.0.x. 99.x. x” E n 8” .E .2:
35-8358 9x... 93.. 3?... x” E n E 8.: .2»...
22.4535 E8353
\5 $8.85.. A2 «um v A2 .I. gov A2 "on V

o cacao; 9 .mfi. 3.9x. 3.9x.

_esaa 8%. 8.3... Ni: x» E u 98 S E

N asaam 3.9x. 8.3.. B .39
33:3,. Eur: 3 £5 Si:

c 9832” 8.0.x. fl. xx. 3%.. N

E832 Bax. NIX. 8?. x E u Moo .3 .Na

N cacao; 0.1%.... B .A..\.. unbfi...
haSma... AZ u wé AZ u wé A2 u N:

o ammo—.38 mm.m..\o 3.~o\o 3.8.x. w I

awoaa Fax. Nix. 85.x. x E I :8 .8 .3...

N Amman—a3 mobfi. Ncbfo 35.x.

 

 

 

 

 

 

 

 

#8538 38 Egan 96333. :5. Emai— 2 vvbu.

_ uxw mafia—83m m2 Zo=.>UmU emanaoa 8833 o amen—n? _ .3832. 2 u 2. :53 E839“ 5 333m 3 0E5 «88.

85

as...» w
avian—.88 u” Goa—bung Um??? >55 9.35.8

 

 

 

 

H.938 $8308 0033_ 0353: 259.5 03585 «a Bug u 93:: macaw
$33.. .. 333.20 a...» .. 932...... E... 9-8.8.5.. 8.8: Ia.

.65
835...»... A2 "83 Sum: Eu 53

>26 93. 3a 3.x... 3.8.x. 8.8.x... x. E u 3.8 .8... .88...
886-3833 8.8.x... 8.8.x. 8.8.... x. E u .8 ._ 8 .o8
8.6.8398 9.x... 99.x... 5.x... x. E u 85 88 .8...
3283. I A? u aov A2 u N3 AZ u as 8
>96 $3. 3.3 P 1.x... 3.8.x. 8.9.x... x. E u 88 .88... .88..
88.3335 8.0.x... 8...: :88. x E u +8 8.. 88...
5.96.8888 0.x... 8.x. +3.... x. E u .8 .28 .88...
33.3,... - A2 u ad 9.. u va A2 u u& 8

>26 $3. 8.3 is... 8.0.x. 8.0.x... x .E u :8 88 8%.
.65-...»2385 Max. 38.x. 8.9.x. w E u _. a .8 88
3630603352. o..\.... o..\.... _m.a..\.... x AB u 8.3 boo... bow...
22?.»65 fine—do:
883...... A2" .5 Eu E Eu 8v

0 982—08 mo.m..\.. 39.8 388 N

3.82.... 8.8.x. 8.9.x. 8.8.x. x E u 8.8 .3 .88
N “:8..qu Sufi. 38.x. 35.x.
3882.... A2 u 3v Q< n va A2 u m8
o Ewan—an 38.x. 8.0.x. $.88 8

3.832 8.8.x. 8.0.x. 88.x. x E u is E .8
N cacao; who; 3.0.x. N918.
$283.. A2 N NE A2 u Nov A2 u E
o “50an 36.x. 3.9.8. .388. 8

5.832 3.8.x. 8.9x. 8.8.x. x E u 88 .3 8
N E8238 Nq.m..\.. No.0..\.. Nq.m..\o

 

 

 

 

 

 

 

 

m. 8558 :36 among. 83813. 93. 95.28 a vvbu.

. ”:0 OE-ma§ m9. 23.36 8833 8833 o 9833... _ E839. o... N 2 :53 98033.. E 38:5 3. 9:5 «8%.

86

as...» 3
Evian—.8? w" 02:31:” >53 9.32.8

 

 

 

 

 

 

 

 

 

 

 

 

 

A91» 2:. 9.59
333 H3832. Q...» 9.8.96 9.3 $6.96
08:2 >26-» 66h a. E u n. 932 >66-» 66h x. E u a.
>36
ENwfinaa I Eu »3 Eu »3 Eu »8 Eu E A2" »: 9?. as
>66 93. Q3 91.x... 8.9.x. 8.3.... 5.3 .8... ».».\.. 3.x. 3...: :3 .8»...
66-33895 a. 1x. 8.0.x. 8.9.x. .3 3a 3.x. 3.x. 8.»: .3... .8»
66-83258 3.. 8.0.x. 8.3.. 9% .3» 9.x. 3.x. 3.x. 98 .8»...
3%.»; I 9." »8 9?. E 3.... E Eu E 9." E A2" E
>66 3..» 3.3 9.x... 8.»: 8.9.x... N8 .8» 93.. 50.x. »».§ 3» .»8
>66émagmé ox 8.x. 3.x. No» .»»a o..\.. 92.x. 5.»... ».»o .8»
6.6.8.352. 9x. Sax. m.»..\.. .23 .5. 9.x. 3.x. A. E. 3» .3a
may»; - 9? E 9? 3 Eu E Eu ».c Eu E Eu 3V
>66 93. 3.3 Sex. 8.3. »m.o..\.. . .8 .2» 9.x... 3.x... 3.9x... 2.3 .8»...
66-53835 Sex. 59.x. 8.9.x. .8 .m»_ 0.x. 3x. ». E. ».3 .»2
66.8353 o..\.. 50.x. 3.. 3» .»»» 9x... 9.x. 3.»..\.... 3m .8»...

 

 

#8668 3:6 @599: 669813». :5. @528 a uvbm.

_ 23.23.

87

5.36.2.8 3. 5:6... Ezifivme $32.9. 5. GE..— macaw—K

Hue—a _—

 

 

 

 

 

0932 0353a 0358: 259.5-“ DEER: i hum—v-0 mnnv n w
$963.” I AZ u ad AZ n £5 AZ u 03
Q>U max. 0.1x. 3.9%.. #8 LB
ZUU 85$ Six. 30.x. Ema .wS
93:: Sax. «6.x. 26.x. .mwm ham
Dawning? 0.x. 9X. 05.x. uhow .NS
OUUBU 30.x. 8.1x. u g .o..\.. Numm bum
32.83% Ev m.~..\o Aux. 0.0.x. Sum .moq
cam agags “.3... 9%; 5.me +35 Sm
Booze. 33388 Sufi. 3.9x. _ _.N..\.. .43 .mmo
32:8.“ I AZ u #3 A2 u umv AZ n m8
Q>U Qqfi. 5.0.x. 89$ .93 .73
E5 Bax... 3.9.x... 3.3.x... :6 .89.
Eve—ma a. 1x. 3.2x. 53.x. _bo .mNA
Dwarfs? 0.x. 0.x. 39.x. _dN ham
OUUBU 3.0.x. B .m..\.. 89.x. N. 5 who
23,89»— Eu ~.~o\o 5.9x. wax. who .3»
Cam nova—588 ~.No\o. 0.x... 73.x... «bu Sm...
28:0. momma—88 Qix. mbfi. 59x. _b boo
WE‘RE - AZ u uuv 2 u H8 m2 u u&
9.6 .39.. 0.x. «6.x. :3 .33
KUU 3.0.x. 3.9x. 5.1x. w. _o .BN
Eva—E. 3.1x. oo\o who; Hum .30
U5??? 9x. 0.x. oo\o - ..
OUUBU 8.2x. 36.x. 33x. .mo .mno
>55an Eu wax. Max“ fax. 83 .mmm
Uam 33338 aux. 30$, pix. row .mmn
Encro— aobosaaaoo 3.1x. 3.0.x. 0.9x. Pg :3. G

 

 

 

 

 

 

 

88