Comorbidity studies:



Running Head: SUBTYPES OF BEHAVIOR PROBLEMS PART I

Examining subtypes of behavior problems among 3-year-old children, Part I: Investigating validity of subtypes and biological risk-factors

Elizabeth A. Harvey[1][2], Julie L. Friedman-Weieneth2, Lauren H. Goldstein2, & Alison H. Sherman2

This manuscript is in press at Journal of Abnormal Child Psychology

Online First version was published in December 2006 and is available at

Abstract

This study examined 3-year-old children who were classified as hyperactive (HYP), oppositional-defiant (OD), hyperactive and oppositional defiant (HYP/OD), and non-problem based on mothers’ reports of behavior. Using fathers’, teachers’, and observers’ ratings of children’s behavior, concurrent validity was excellent for the HYP/OD group, moderate for the HYP group, and poor for the OD group. As predicted, both the HYP/OD and HYP groups reported more prenatal/perinatal birth complications and a greater family history of hyperactivity than did non-problem children. Furthermore, the HYP/OD group showed a greater family history of conduct disorder and oppositional defiant disorder (ODD) symptoms than did non-problem children; however, the HYP group also showed a greater family history of ODD than did non-problem children. Results suggest that as early as age 3, these behavior subtypes appear to be linked to biologically-based risk-factors in ways that are consistent with theories of the development of ADHD.

Key words: hyperactivity, oppositional defiance, comorbidity, preschool-aged children

Examining subtypes of behavior problems among 3-year-old children, Part I: Investigating validity of subtypes and biological risk-factors

The past two decades have seen a dramatic increase in the use of stimulant medication among preschoolers (Zito et al., 2000), highlighting the need for more research on young children with behavior problems. Although the preschool years are often characterized by high activity, defiance, and aggression (Campbell, 1995), there is growing evidence that developmentally deviant levels of these behaviors occur among young children (Egger & Angold, 2006; Keenan & Wakschlag, 2004), can cause significant impairment (Keenan & Wakschlag, 2000; Lahey et al., 1998), and are relatively stable over time. Seventy to 80% of older preschool-aged children with attention-deficit/hyperactivity disorder (ADHD) or oppositional defiant disorder (ODD; Lahey, Pelham, & Loney, 2004; Speltz, McClellan, DeKlyen, & Jones, 1999) and approximately half of younger preschool-aged children with behavior problems (Campbell, Ewing, Breaux, & Szumowski, 1986; Lavigne et al., 1998) continue to show clinically significant behavior problems when they reach school-age, although stability rates may be lower when children are identified through community screening (Lochman & The Conduct Problems Prevention Research Group, 1995). ODD and particularly ADHD have in turn been shown to be stable across the school-aged years, although rates vary across studies (e.g., August, Braswell, & Thuras, 1998; Costello, Mustillo, Erkanli, Keeler, & Angold, 2003; Hart et al., 1995), and developmental trajectories vary across children, with some following a chronic pathway and others showing a decrease in symptoms over time (Nagin & Tremblay, 1999).

Thus, although behavior problems are transient for some preschool-aged children, they represent the beginning stages of stable behavior disorders for many others. However, early identification of these disorders is complicated by the challenge of teasing apart early ADHD and ODD symptoms from developmentally normal or transient behavioral difficulties (Campbell, 1995). Moreover, research on older children points to the importance of considering pure ADHD and pure ODD separately from ADHD/ODD (Waschbusch, 2002); however, it is not clear how early these subtypes differentiate or how various subtypes in preschoolers map on to ADHD, ODD, and ADHD/ODD in older children. For example, it is possible that pure ADHD tends to manifest as comorbid hyperactivity/oppositional-defiance during the preschool years or that pure hyperactivity in the preschool years is primarily a transient condition. Only a handful of studies have examined the stability of preschool hyperactivity and oppositional-defiance taking into account their comorbidity (Moffitt, 1990; Speltz, McClellan et al., 1999; Sonuga-Barke, Thomson, Stevenson, & Viney, 1997), but they have only considered comorbidity during preschool and not also at follow-up.

While longitudinal research is needed to address these issues, cross-sectional data may also contribute to our understanding of early subtypes of behavior problems, by examining whether developmentally deviant hyperactivity, oppositional-defiance, and comorbid hyperactivity/oppositional-defiance in preschoolers share etiological correlates with ADHD, ODD, and ADHD/ODD. Moving examination of etiological correlates closer in time to the emergence of these disorders also has the potential to: (a) contribute to the growing body of evidence that developmentally deviant levels of hyperactivity and oppositional-deviance among preschoolers may represent clinically significant syndromes, (b) begin to establish the age at which the effects of different etiological factors become evident, and (c) strengthen support for existing etiological models. Given retrospective reports that ADHD symptoms typically begin to cause impairment at age 3.5 (Applegate et al., 1997), efforts should include young preschool-aged children.

Comorbidity between ADHD and ODD

Approximately 50% of school-aged children with ADHD also meet criteria for ODD or conduct disorder (CD; Barkley, 2006). A growing body of research that has compared children with pure ADHD, pure ODD, and comorbid ADHD/ODD (for a review, see Waschbusch, 2002) is playing a critical role in addressing questions about the etiology of these disorders and their interplay. These studies can tease apart whether etiological factors are linked with ADHD, ODD, or both conditions, and can also begin to provide insight into the nature and etiology of comorbid ADHD/ODD. For example, Waschbusch’s (2002) meta-analysis suggested that ADHD/ODD does not appear to be a separate taxonomic category, but nonetheless differs in important ways from pure ADHD and ODD. Despite the growing recognition of the importance of considering ADHD, ODD, and their comorbidity separately, most studies of preschool behavior problems use broad measures of externalizing problems and do not make these critical distinctions.

Studies that have compared preschool-aged children with ADHD, ODD, and ADHD/ ODD symptoms (Coy, Speltz, DeKlyen, & Jones, 2001; Cunningham & Boyle, 2002; Gadow & Nolan, 2002; Keenan & Wakschlag, 2000; Ross, Blanc, McNeil, Eyberg, & Hembree-Kigin, 1998; Schwebel, Spelz, & Jones, 2002; Sonuga-Barke, Lamparelli, Stevenson, Thompson, & Henry, 1994; Speltz, DeKlyen, Calderon, Greenberg, & Fischer, 1999; Speltz, DeKlyen, & Greenberg, 1999; Speltz, McClellan et al., 1999; Stormont-Spurgin & Zentall, 1995) point to the value of examining subtypes separately. For example, differences across subtypes have been found in family functioning, parenting, maternal depression, and preschool functioning (Cunningham & Boyle, 2002; Stormont-Spurgin & Zentall, 1995); comorbid psychiatric symptoms, developmental deficits, and peer conflict (Gadow & Nolan, 2002); attachment security (Speltz, DeKlyen, & Greenberg, 1999); cognitive functioning (Keenan & Wakschlag, 2000; Sonuga-Barke et al., 1994; Speltz, DeKlyen, Calderon et al., 1999); and stability of behavior problems (Speltz, McClellan et al., 1999). However, these studies need to be replicated, because most correlates have been examined by only one or two studies. They also need to be extended to include other variables that are implicated in etiological models, including family history of behavior problems and prenatal complications. Finally, research is needed to determine whether behavior problems begin to differentiate into subtypes, as evidenced by theoretically consistent patterns of correlates, in children as young as age 3, when evidence suggests that these disorders are likely first emerging. Only one of these studies has focused specifically on children younger than age 4, and very few have even included younger preschool-aged children.

Theory and Research on the Etiology of ADHD, ODD, and ADHD/ODD

Theoretical models of the development of ADHD and ODD point to separate and shared etiological risk factors for these disorders (see Figure 1). The core deficit of ADHD is thought to be behavioral disinhibition (Barkley, 1997), which is largely caused by genetics (Silberg, Rutter, and Meyer, 1996) and biological influences including prenatal and perinatal complications (e.g., Sprich-Buckminster, Biederman, Milberger, Faraone, & Lehman, 1993; Zappitelli, Pinto, & Grizenko, 2001). ODD is thought to develop when largely biologically determined early child characteristics elicit and interact with family stressors and parenting practices, in a dynamic process that continues and progresses over time (e.g., Dodge & Petit, 2003; Lahey & Loeber, 1994; Lahey & Waldman, 2003; Moffitt, 1993; Patterson 1992). The reason for high comorbidity between ADHD and ODD is still not well understood. Both environmental and genetic factors have been implicated (Burt, McGue, Krueger, & Iacono, 2005), but the specific mechanisms are unknown. Barkley (1990) suggested one possible environmental process, outlined in Figure 1. Specifically, children’s ADHD symptoms may exacerbate and interact with certain family stressors, leading to less effective parenting and subsequent ODD.

The Present Study

This study is the first in a three-part series, which examines 3-year-old children who were classified into one of four groups: those who showed elevated levels of hyperactivity (HYP), oppositional-defiance (OD), hyperactivity/oppositional-defiance (HYP/OD), and no problems. Although ADHD is characterized by inattention as well as by hyperactivity, the average age of onset of ADHD predominantly inattentive type is 6 years old (Applegate et al., 1997). Whether this is due to later onset of inattention or to difficulties in assessing developmentally deviant levels of inattention during the preschool years, the present study focused on developmentally deviant levels of hyperactivity rather than inattention.[3] This study begins by examining whether HYP, HYP/OD, and OD subtypes exist at age 3, and if so, whether there is evidence for concurrent validity of these subtypes; these questions have not been previously addressed at age 3. The three studies in this series then seek to further examine the validity of each subtype by testing whether etiological factors identified in theoretical models of ADHD and ODD can be linked with hyperactivity or oppositional defiance among 3-year-old children, taking into account comorbidity. Associations between these factors and developmentally deviant levels of preschool behavior problems do not necessarily establish a link with early ADHD and ODD; many children outgrow early hyperactivity and oppositional-defiance. Nonetheless, linking such factors to developmentally deviant levels of ADHD and ODD symptoms during the preschool years provides an important step in understanding the early development of behaviors that are common precursors to, if not direct manifestations of ADHD and ODD. This first paper examines genetic/biological risk factors, the second paper focuses on family stressors, and the third investigates parenting. Specifically, the present study examines the following questions:

Do fathers’, teachers’, and observers’ ratings of behavior differentiate 3-year-old children with HYP, OD, and HYP/OD? A number of studies have documented that preschool-aged children with ODD/CD and/or ADHD differ from non-problem children on teachers’ reports of behavior (e.g., Speltz, DeKlyen et al., 1999) and observations of behavior (e.g., Kim-Cohen et al., 2005). However, few studies (Cunningham & Boyle, 2002; Gadow & Nolan, 2002; Speltz, DeKlyen, Calderon et al., 1999) have examined whether teachers’ or observers’ ratings discriminate preschool children with ADHD, ODD, and ADHD/ODD from each other. We are not aware of any studies that have done so in children as young as 3 years old, and no studies have examined whether fathers’ reports corroborate mothers’ reports in distinguishing types of behavior problems among preschoolers. Maternal report was chosen as the basis for group formation because clinical decisions for preschool-aged children are typically based on mothers’ reports, and mothers are the most common source of information in research as well. In addition, many 3-year-old children are not yet in preschool and/or do not have a second parent available as an additional informant. It was predicted that fathers’, teachers’, and observers’ ratings of hyperactivity/activity would be elevated among children with HYP and HYP/OD compared to children with OD and non-problem children. Ratings of aggression/defiance and negative affect should be elevated in children with HYP/OD and OD compared to children with HYP and non-problem children. It was expected that observed noncompliance would be elevated for all three behavior groups compared to non-problem children.

Are there differences across subtypes in family history of ADHD and ODD/CD symptoms? Family history of behavior problems has been found to distinguish among different disruptive behavior disorders in older children (Frick, Lahey, Christ, Loeber, & Green, 1991). Given evidence of the stability of behavior problems from preschool to school-age, 3-year-old children with hyperactivity and/or oppositional-defiance should show elevated family histories of behavior problems. However, continuity between specific subtypes of behavior problems among 3-year-old children and later ADHD and ODD has not been well-established, so it is unclear whether family histories of specific types of behavior problems can be linked to specific subtypes of behavior problems as early as age 3. It was predicted that children in the HYP and HYP/OD groups would both show elevated family histories of ADHD, and children in the HYP/OD and OD groups would show elevated family histories of ODD/CD. Establishing these specific links would provide some support for the validity of these subtypes in young preschool-aged children, and ultimately, if supported by longitudinal data, could inform early diagnostic assessments.

Do preschool-aged children with HYP, HYP/OD, and OD have a history of more prenatal and perinatal complications than do non-problem children? A variety of complications have been linked to ADHD and/or ODD, including cigarette and alcohol use during pregnancy, hypoxia, prematurity/low birth weight, maternal emotional and physical health during pregnancy, duration of labor, and use of forceps during delivery (Allen, Lewinsohn, & Seeley, 1998; Hartsough & Lambert, 1985; Sprich-Buckminster et al., 1993; Zappitelli et al., 2001). However, only a handful of studies have linked prenatal complications to behavior problems in preschool-aged children (Orlebeke, Knol, & Verhulst, 1999; Wakschlag & Keenan, 2001). The few studies that have attempted to control for the comorbidity between hyperactivity and conduct problems in older children have not found consistent results (for a review, see Waschbusch, 2002), pointing to the need for further research in this area. If, as predicted, 3-year-old children with HYP, HYP/OD, and OD all show more complications than non-problem children, this would provide further support for each subtype as a clinically significant behavior disturbance.

Method

Participants

Participants were drawn from 258 children and their 258 female primary caregivers (251 biological mothers, 4 adoptive mothers, and 3 grandmothers, to whom we will refer as mothers) and their 178 male caregivers (165 biological fathers, 3 adoptive fathers, 6 stepfathers, 3 grandfathers, and 1 uncle, to whom we will refer as fathers). Children (138 boys and 120 girls) were all 3 years old at the time of initial screening and were 36.43 to 50.30 months (M = 44.15 months, SD = 3.37) at the time of the first home visit (11% had just turned 4 years old). The sample was ethnically diverse; 55% were European American, 18% Latino (predominately Puerto Rican), 12% African American, and 15% were multi-ethnic. The median combined family income was $48,000. Most mothers (87.5%) and fathers (89.4%) had high school diplomas and 33.5% of mothers and 28.7% of fathers had bachelor’s degrees. These are comparable to the median family income and educational attainment in the three counties from which the sample was drawn. (In the year 2000, family income of the three counties ranged from $50,193 to $57,193, rates of high school degrees ranged from 79% to 89%, and rates of bachelor’s degrees ranged from 20% to 38%; U.S. Census Bureau, 2006).

Procedure

All participants were recruited over a 3-year period by distributing screening questionnaire packets through state birth records, pediatrician offices, child care centers, and community centers throughout western Massachusetts. Children with significant externalizing problems (n = 199) and without behavior problems (n = 59[4]) were recruited from 1752[5] 3-year-old children whose parents completed a screening packet containing the Behavior Assessment System for Children – Parent Report Scale[6] (BASC-PRS; Reynolds & Kamphaus, 1992) and a questionnaire assessing for exclusion criteria, parental concern about externalizing symptoms, and demographic information. The present set of studies focused on data collected during the first year (age 3) of a 4-year longitudinal study. Eligible families were scheduled for two 3-hour home visits scheduled approximately 1 week apart, and each parent was paid a total of $200. For the first 2 years of recruitment, all families eligible for the externalizing group were invited to participate. During the last year of recruitment, African American and Latino children who met criteria for the externalizing group were prioritized for invitations to participate, in order to obtain an ethnically diverse sample. European American children who were eligible for the externalizing group were randomly selected to participate whenever there were insufficient numbers of eligible African American or Latino children to meet recruitment goals. Bilingual staff conducted home visits for Spanish-speaking families. All child behavior measures were available in Spanish from the scale developers. The prenatal/perinatal complications questionnaire was translated by professional translators, and bilingual staff on the project compared the translation to the English version to ensure equivalency. Bilingual staff used the Spanish DBRS and DISC to create a Spanish version of the family history interview.

Criteria for all participants included no evidence of mental retardation, deafness, blindness, language delay, cerebral palsy, epilepsy, autism, or psychosis. Criteria for the externalizing group were: (a) parent responded “yes” or “possibly” to the question, “Are you concerned about your child’s activity level, defiance, aggression, or impulse control?” and (b) BASC-PRS hyperactivity and/or aggression subscale T scores fell at or above 65 (approximately 92nd percentile). Of the 1752 children who were screened, 411 met criteria a and b, and were not ruled out for language, mental, or physical problems. Of the 411, 71 were not contacted because European American children were no longer being recruited. For the non-problem comparison children, criteria were: (a) parent responded “no” to the question, “Are you concerned about your child’s activity level, defiance, aggression, or impulse control?” and (b) T scores on the BASC-PRS hyperactivity, aggression, attention problems, anxiety, and depression subscales fell at or below a T score of 60. Of the 1752 children who were screened, 452 met both of these criteria. For matching, 59 children in the externalizing group were identified by selecting every third or fourth child in the externalizing group separately by gender and ethnicity. For each of these children, a child was identified from the pool of non-problem children who was of the same gender and ethnicity[7], and most similar to the target child on parent education and child age. Twenty-three children were matched but could not be scheduled (18 families could not be reached and 5 were no longer interested). When this occurred, a new comparison child was contacted. Thus 59% of behavior problem children and 72% of non-problem children whom we sought to recruit, participated.

Measures

Demographic information. Parents provided information about their income, race/ethnicity, years of education, age, number of children, and marital status.

BASC-PRS and BASC-Teacher Report Scale (TRS). This comprehensive rating scale assesses a broad range of psychopathology in children ages 2-6 and older. T scores (based on general, not gender-specific, norms) for the hyperactivity (16 items for the BASC-PRS, 10 items for the BASC-TRS) and aggression (13 items for BASC-PRS, 12 items for BASC-TRS) subscales were used. These two subscales have demonstrated good reliability for 2- to 3-year-old children (α = .79 and .83 for the PRS, and α = .92 and .88 for the TRS; Reynolds & Kamphaus, 1992). The BASC has distinguished school-aged children with and without ADHD (Ostrander, Wienfurt, Yarnold, & August, 1998). The BASC-PRS was completed at the screening by one parent (247 mothers and 12 fathers) and again at the home visit by both parents. The BASC-TRS was completed by 87 preschool teachers and 68 child care providers.

Disruptive Behavior Rating Scale (DBRS). The ADHD and ODD sections of the DBRS (Barkley & Murphy, 1998) were administered to both parents. This scale presents the 18 DSM-IV ADHD symptoms and 8 DSM-IV ODD symptoms. Parents indicate how often their children displayed these symptoms in the past 6 months using a 4-point Likert scale. The 18 ADHD items have also been published as the ADHD Rating Scale, which has shown good reliability and validity (DuPaul, Power, McGoey, Ikeda, & Anastopoulos, 1998) with older children. Based on data from this study (Friedman-Weieneth, Doctoroff, Harvey, & Goldstein, 2006) the DBRS generally showed good reliability and validity for 3-year-old children, but four of nine items did not load on the Hyperactive/impulsive factor. Based on this factor structure, the following five items were averaged to create the hyperactive/impulsive score for mothers: fidgets, leaves seat, runs about/restless, on the go/driven by a motor, and interrupts/intrudes. The following five items were averaged for fathers: fidgets, leaves seat, runs about/restless, on the go/driven by a motor, and talks excessively. All eight ODD items were averaged. Using these items, internal consistency was good for both the hyperactive/impulsive subscale (mothers’ α = .83, fathers’ α = .80) and the ODD subscale (mothers’ α = .86, fathers’ α = .87).

Diagnostic interview (DISC). The NIMH-Diagnostic Interview Schedule for Children-IV (NIMH-DISC-IV; Shaffer, Fisher, Lucas, Dulcan, & Schwab-Stone, 2000) is a structured diagnostic interview that was designed for children age 6 and up, and has demonstrated adequate test-retest reliability with older children for ADHD (.79) and ODD (.54; Shaffer et al., 2000). The ADHD and ODD sections of the NIMH-DISC-IV were modified to assess behavior in preschool/daycare rather than school, and were administered to mothers in all but three cases in which the father was interviewed because the mother was not able to be at the home visit. A secondary caregiver (94% were the father or mother’s partner) was also present during the interview 57% of the time, but the interviewer used the mother’s responses when there was disagreement. Interviewers all had bachelor’s or master’s degrees in psychology and were trained by the first author who is a licensed clinical psychologist. Training included role playing, listening to experienced interviewers, and receiving feedback on several audiotaped interviews. A factor analysis of the DISC ADHD and ODD symptoms yielded results that were similar to the factor analysis of the DBRS for mothers reported above, suggesting that the same symptoms be used for creating scores. The total symptom counts for the five hyperactive/impulsive symptoms (see DBRS description) and eight ODD symptoms were used. Cronbach’s α was .81 for hyperactivity/impulsivity and .83 for ODD symptoms.

Videotaped assessment of parent and child behavior. Children were videotaped during cognitive testing and during three parent-child interaction tasks: a 5-minute play task, a clean-up task, and a 10-minute forbidden objects task. Research assistants coded the first and seventh 5-minute segments of the cognitive testing (to assess children’s initial responses to the demands of testing and their behavior after the novelty of testing wore off), the parent-child play task, the clean-up task, and the first 5 minutes of the forbidden objects task. Global ratings of child behavior were made after each segment on 5-point scales and the following dimensions were used in this study: noncompliance, defiance/aggression, negative affect, and activity level. High scores represented frequent displays of each of these behaviors. Details of the interaction and coding system can be obtained from the first author. Ratings were averaged across the two cognitive testing segments to obtain a set of testing behavior scores. Ratings were averaged across the three parent-child interaction tasks to obtain a set of parent-child interaction scores. Each videotape was coded by two raters who were unaware of the children’s group status, and each pair of ratings was averaged. Intraclass correlations ranged from .58 (negative affect) to .84 (noncompliance) for cognitive testing behavior (M = .73) and from .78 (activity level) to .90 (noncompliance) for the parent-child interaction (M = .86). Each of the coding variables was skewed so logarithmic transformations were conducted on each variable. Ratings during the parent-child interaction and cognitive testing were significantly correlated for noncompliance (r = .35, p < .001), activity level (r = .19, p < .01), defiance (.32, p < .001), and negative affect (r = .24, p < .001), so they were averaged across the two situations.

Audiotaped assessment of child behavior. Mothers and fathers were each asked to use a micro-cassette player to record 2 hours of interaction with their children, selecting times of day that tended to be challenging. A preliminary review of the tapes suggested that 30 minutes of tape was sufficient for capturing a wide variety of behavior that was representative of the entire 2 hours, and all parents who were willing to take part in this assessment completed at least 30 minutes. The coding system included both event-based and global coding, details of which can be obtained from the first author. Two raters overlapped for 88 participants (24%) and intraclass correlations (ICCs) were calculated. The following child behavior codes were used in this study: noncompliance (ICC = .64), verbal aggression (ICC = .75), and child affect (created by combining an event-based code of child negative affect, ICC = .93, and global child affect ratings, ICC = .61). Verbal aggression, noncompliance, and the event-based negative affect code were skewed so square root transformations were conducted.

Prenatal and perinatal factors. Mothers completed a questionnaire regarding a range of prenatal and perinatal variables (Allen et al., 1998). Based on the existing literature regarding which complications have been linked with ADHD and antisocial behavior, the following six subscales were used: prenatal maternal physical health (vaginal bleeding, premature contractions, swelling of face and hands, high blood pressure, seizures/convulsions, rubella, infectious diseases, diabetes mellitus, anemia, serious injury, and x-rays); prenatal maternal emotional health (depression and anxiety); maternal substance use (cigarettes, alcohol, or marijuana use during pregnancy); difficult delivery (breech birth, use of forceps); prematurity (low birth weight, premature delivery, baby required incubator); acute anoxia/hypoxia (cord around neck, blue baby, slow heart beat, baby did not breathe, baby had convulsions, baby required oxygen).[8] Allen et al. (1998) examined test-retest reliabilities for identifying mothers in the top quartile of the distribution and all kappa’s were .58 or higher for these scales. Each item was scored “1” if it had occurred and “0” if it had not, and then items within each subscale were summed. A total complications score was calculated by summing across the six variables. Subscales were also examined separately in order to evaluate specific types of complications. Prenatal emotional health, prenatal substance use, difficult delivery, prematurity, and anoxia variables consisted of few items each or had small ranges so they were transformed into dichotomous variables; the complication was coded as present if scores were 1 or greater and absent if scores were 0. The total complications and maternal physical health variables were significantly skewed, so square root transformations were conducted.

Family history of behavior problems. Parents were administered a family history interview in which they were asked whether various biological relatives of the child had problems with each ADHD, ODD, and CD symptom before the age of 18. The same caregivers who were present for the DISC interview jointly participated in this interview.[9] The format of the scale is similar to that described by Frick et al. (1991), but was updated for DSM-IV and ODD symptoms were added. The following items were used for CD: fought with other children, vandalized, stole, skipped school, ran away from home, lied, and set fires. The parents reported on both biological parents, the child’s full siblings (18 months and older), and both parents’ full siblings. Symptoms were averaged across siblings and across aunts/uncles. Sibling, aunt/uncle, mother, and father symptoms were then averaged separately for hyperactive, ODD, and CD symptoms. The CD variable was skewed, with a large number of families reporting no CD symptoms. Logarithmic and square root transformations still resulted in a highly skewed variable, so a dichotomous variable was created. Children were considered to have a family history of CD (coded 1) if at least one of the four CD variables (aunt/uncle, mother, father, or sibling CD symptoms) was greater than one.

Analytic Plan

One-way ANOVAs were conducted on continuous variables and chi-square tests were used for dichotomous variables. Significant ANOVAs were followed up with planned contrasts to test predicted differences and with Tukey HSD tests to evaluate differences that were not predicted. To assess effect sizes, partial eta-squares were reported for ANOVAs, and Cramer’s Phi was reported for chi-square tests. Power was somewhat low for some analyses, so using a Bonferroni correction would have resulted in a high rate of Type II error. Using a correction would also make it more difficult to compare the results of the present study with previous studies that have examined subtypes among preschoolers (Cunningham & Boyle, 2002) and school-aged children (Jensen et al., 2001), which did not use Bonferroni corrections. Thus, an alpha of .05 was used in this study. Findings with higher p-values should be interpreted with some caution and will need to be replicated.

Results

Subtype Grouping Process

Traditionally, researchers have used a dimensional approach to identifying preschool-aged children with behavior problems, using cutoffs on behavior rating scales; however, some researchers have used various diagnostic approaches with preschoolers (for a review, see Egger & Angold, 2006). Although there is growing support for the validity of ADHD and ODD diagnoses among preschool-aged children, much of this research is based on children ages 4 and up, and there is not yet consensus regarding diagnosis of ADHD and ODD among 3-year-old children. Therefore, rather than using DSM diagnostic criteria, developmentally deviant levels of hyperactivity/impulsivity and oppositional-defiant symptoms were identified by aggregating across multiple maternal measures and using cutoffs tied to national norms as follows:

First, a hyperactivity index was calculated for each child by standardizing (transforming to z scores) and averaging the following maternal-report measures as described in the Method: screening BASC-PRS hyperactivity subscale, BASC-PRS hyperactivity subscale completed at the home visit, DBRS hyperactivity/impulsivity score, DISC hyperactivity/impulsivity symptom count. An oppositional-defiance index was also calculated for each child by standardizing and averaging the screening BASC-PRS aggression subscale, BASC-PRS aggression subscale completed at the home visit, DBRS ODD score, and DISC ODD symptom count.[10]

Next, children were categorized into one of four groups using these aggregate hyperactivity and oppositional-defiant indexes. Because the BASC has been normed on 3-year-old children, we were able to identify cutoffs that corresponded with developmentally deviant levels of behavior. Specifically, z scores that corresponded to a BASC T score of 65 (1.5 SD above the mean; 92nd percentile based on national norms; Reynolds & Kamphaus, 1992) were selected as the cutoff. In this sample, the mean T scores for the BASC-PRS hyperactivity and aggression subscales completed during the screening were both 65 and the mean hyperactivity and oppositional-defiance indexes were both 0. This means that hyperactivity and oppositional-defiant indexes of 0 corresponded to BASC-PRS screener scores of 65, so z scores of 0 were used as cutoffs. The HYP group consisted of children who fell above a z score of 0 on hyperactivity and below 0 on oppositional-defiance. The OD group fell above a z score of 0 on oppositional-defiance and below 0 on hyperactivity. The HYP/OD group fell above 0 on both hyperactivity and oppositional-defiance. In addition, to increase the distinctiveness of the groups, children in the HYP group were required to score the equivalent of 5 BASC-PRS T score points higher on the hyperactivity index than on the oppositional-defiant index (a z score difference of .31). Similarly, children in the OD group were required to score the equivalent of 5 BASC-PRS T score points higher on the oppositional-defiance index than on the hyperactivity index. Children were classified as non-problem if both behavior problem indexes fell below a z score that corresponded to T scores of 60 on screener BASC-PRS aggression and hyperactivity subscales (z = -.34 for hyperactivity and z = -.28 for oppositional-defiance). Using this method, 220 of the 258 children fell into one of the three behavior problem groups or the non-problem group: 41 in the HYP group, 24 in the OD group, 96 in the HYP/OD group, and 59 in the non-problem group.

Demographic and Symptom Severity Differences Across Groups

Groups were first compared on demographic and other variables that might affect interpretation of results (see Table 1). European American, African American, and multi-ethnic children were fairly evenly distributed across the groups; however, the OD and non-problem groups had few, if any, Latino children, and the HYP/OD group had a disproportionately large number of Latino children. Children in the HYP/OD group also tended to be from less educated, lower-income families than children in the other groups. Children in the HYP/OD group received higher hyperactivity ratings from mothers than did children in the HYP group. In addition, although the HYP group was selected to not be elevated on oppositional-defiance, they did receive higher BASC aggression scores than the non-problem group.

Do Fathers’, Teachers’, and Observers’ Ratings of Behavior Differentiate 3-Year-Old Children with HYP, OD, and HYP/OD?

Table 2 presents comparisons across groups on fathers’ and teachers’ reports of children’s hyperactivity and aggression/oppositional-defiance. Significant correlations were found between fathers’ BASC and DBRS hyperactivity scores (r = .71, p < .001) as well as between fathers' BASC aggression and DBRS ODD scores (r = .69, p < .001). These scales were therefore standardized and averaged to create the variables, fathers’ ratings of hyperactivity and fathers’ ratings of oppositional-defiance. Fathers’ ratings of hyperactivity were significantly higher in the HYP/OD group than in the OD and non-problem groups, and their ratings of oppositional-defiance were significantly higher in the HYP/OD group than in the HYP and non-problem groups. Fathers also rated children in the HYP group as more hyperactive than non-problem children, and they rated children in the OD group as more oppositional-defiant than non-problem children. Teachers’ BASC aggression scores were significantly higher in the HYP/OD group than in the HYP and non-problem groups, and teachers’ BASC hyperactivity scores were significantly higher in the HYP/OD and HYP groups than in the OD group.

Ratings of child behavior during videotaped and maternal audiotaped interactions were modestly but significantly correlated for noncompliance (r = .21, p < .01), defiance/verbal aggression (r = .17, p < .05), and negative affect (r = .14, p < .05), so they were standardized and averaged. [11] As predicted (see Table 2), children in the HYP/OD group received higher noncompliance, negative affect, and defiance ratings than did non-problem children, and were rated as more active than children in the OD group. Children in the HYP group were rated as significantly more active and noncompliant than non-problem children; however, they were also rated as showing more defiance and negative affect. Contrary to hypothesis, ratings of child behavior were not significantly different between the OD and non-problem groups, or between the HYP/OD and HYP groups. Child behavior ratings based on father-child audiotaped interactions showed fewer differences across groups. Although children with HYP/OD were rated as showing more negative affect than non-problem children during father-child interactions, children in the HYP group showed the most negative affect with their fathers.[12]

In sum, there was generally good concurrent validity for the HYP/OD group, moderate support for the HYP group, and little support for the OD group. Because the OD group was small and there was little evidence corroborating mothers’ reports of oppositional-defiance for these children, further analyses did not include this group of children.

Differences in Family History of Behavior Problems and Prenatal/Perintal Complications

Because groups differed on demographic variables that could account for differences in family history of behavior problems and prenatal/perinatal complications, these analyses were conducted controlling for maternal education, ethnicity, and marital status. Maternal education was used as a proxy for SES because there were no missing data on this variable; mothers were categorized as high education (greater than 12 years) or low education (12 years or less). For continuous dependent variables, ANOVA’s were conducted entering maternal education as a blocking variable[13]. For dichotomous variables, multinomial logistic regressions were conducted, entering maternal education as a covariate. These analyses were then repeated twice, first entering ethnicity (1 = non-European American, 0 = European American) and then marital status as blocking variables/covariates. There was not enough power to enter all three demographic controls simultaneously, so we selected SES (maternal education) as the primary control variable because SES likely accounts for much of the confound of marital status and ethnicity. Table 3 presents results of analyses that used maternal education as a control; footnotes in the table indicate any results that changed when ethnicity or marital status were controlled.

Are there differences across subtypes in family history of ADHD and ODD/CD symptoms? Compared to parents of non-problem children, parents of children in the HYP and HYP/OD groups reported greater family histories of hyperactivity, and parents of children in the HYP/OD group reported greater family histories of ODD and CD (see Table 3). Contrary to hypothesis, parents of children in the HYP/OD group did not report greater family histories of ODD (p = .18) or CD (p = .13) than parents of children in the HYP group, although the differences were in the expected direction with small effect sizes (Cohen’s d for the ODD difference = .3; Cramer’s Phi for the CD difference = .18).

Do preschool-aged children with HYP and HYP/OD have a history of more prenatal and perinatal complications than do non-problem children? As predicted, children with HYP and HYP/OD had significantly higher total complication scores than did non-problem children, and this difference appeared to be accounted for by differences in prenatal physical health, prenatal emotional health, and prenatal substance use (see Table 3). Because children in the HYP group were rated by mothers as more oppositional-defiant than non-problem children, it is possible that the difference between the HYP and non-problem groups on pregnancy and birth complications was due to differences in oppositional-defiance rather than to differences in hyperactivity. To address this possibility, forty children (26 children from the non-problem group and 14 children who did not fall in any of the groups) were identified who were not elevated on hyperactivity (z < -.34), but whose oppositional-defiance index fell in the same range as that of the HYP group’s (z between -1 and 0). An ANOVA with maternal education entered as a blocking variable indicated that the HYP group scored higher on the overall prenatal/perinatal complications variable than did these 40 non-hyperactive children who were matched to the HYP group on oppositional-defiance, F (1, 74) = 6.77, p < .05. (ANOVAs with child ethnicity and with marital status entered as blocking variables were also significant, p’s < .05.) Thus, prenatal/perinatal complications were associated with hyperactivity regardless of comorbid oppositional-defiance.

Discussion

The present study examined subtypes of behavior problems among 3-year-old children. The goals were to begin to evaluate the validity of these subtypes by using cross-informants and by testing predictions based on existing theory and research regarding differences in biologically-based risk factors across these groups. Using multiple measures of maternal report, three types of behavior problems were identified: hyperactive only (HYP), oppositional-defiant only (OD), and both hyperactive and oppositional defiant (HYP/OD). Based on fathers’, teachers’, and observers’ ratings of children’s behavior, there was generally good concurrent validity for the HYP/OD group, moderate support for the HYP group, and little support for the OD group. Consistent with existing theory and research on the etiology and age of onset of ADHD, the present study suggested that biological risk factors, including family histories of hyperactivity and prenatal/perinatal complications, are evident among children with developmentally deviant levels of hyperactivity as early as age 3, regardless of comorbid oppositional-defiance. While two of the three sources of cross-informant ratings (fathers and teachers) supported a distinction between the HYP and HYP/OD groups, neither of the two etiological variables (family history of ODD and CD) that were hypothesized to discriminate these groups from each other did so. Parts II and III of this series focus more extensively on additional etiological variables and can more thoroughly address whether there is support for a distinction between HYP and HYP/OD in this sample of young preschool-aged children.

Validity of HYP, HYP/OD, and OD Subtypes

Observational data generally discriminated children in the HYP and HYP/OD groups from children in the non-problem group. However, observations did not discriminate children in the HYP/OD group from children in the HYP group, despite the fact that both teachers’ and fathers’ ratings suggested that real differences between these two groups likely exist. These findings are consistent with prior research on preschool-aged children (Cunningham & Boyle, 2002), but stand in contrast to studies of older children (Abikoff et al., 2002; Barkley, Fischer, Edelbrock, & Smallish, 1991; Johnston, 1996). Our observations may not have been sensitive to the key differences between the HYP/OD and HYP groups, which according to maternal report involved hostility and anger rather than defiance. Future research should focus on specific dimensions of child behavior and develop observational assessment procedures that might better detect differences between preschool-aged children with HYP and HYP/OD.

Teachers’ and observers’ ratings did not discriminate the OD group from the non-problem group. Since fathers corroborated mothers’ ratings of the OD group’s behavior, it is possible that this group had behavior difficulties at home, but not at school, and that our observations were not sensitive to these particular difficulties. It is also possible that the differences between the OD and non-problem groups on parent-reported oppositional-defiance reflect perceived rather than actual differences. The few studies that have used clinic-referred preschool samples to examine children with pure ODD separately from children with ADHD/ODD provide some validity for this diagnosis (Gadow & Nolan, 2002; Speltz, DeKlyen, Calderon et al., 1999). However, it may be that the prevalence of pure oppositional-defiance is too low to identify it in a non-clinic sample as early as age 3.

Differences in Biological Risk Factors

Consistent with previous studies on older children (Frick et al., 1991), children with hyperactivity had elevated family histories of hyperactivity and children with comorbid hyperactivity/oppositional-defiance had elevated family histories of ODD and CD. However, our finding that children with pure hyperactivity had elevated family histories of ODD warrants further investigation. It is possible that the mechanism for the genetic transmission for oppositional-defiance is linked to the transmission of hyperactivity. It also may be that children with pure hyperactivity did not develop ODD symptoms because of low family stressors, whereas some of their hyperactive relatives may have been raised in high stress environments and therefore developed ODD. Studies of family history of behavior problems have focused on Antisocial Personality Disorder and CD rather than on ODD (e.g., Frick et al., 1991). Further research is needed to replicate the findings of the present study in younger and older children.

Our finding of greater prenatal/perinatal complications among children with hyperactivity regardless of comorbid oppositional-defiance is consistent with research and theory on the etiology of ADHD, and with the few studies that have examined complications among preschoolers with behavior problems (Orlebeke et al., 1999; Wakschlag & Keenan, 2001). Examination of specific types of complications implicated physical problems during pregnancy, prenatal emotional health, and prenatal substance use. Differences remained when controlling for maternal education, but other possible confounds cannot be ruled out. For example, in the case of prenatal emotional health and prenatal substance use, links with children’s behavior problems could be accounted for by psychosocial factors such as maternal psychopathology, rather than by biological effects on the child. While breech/forceps delivery, prematurity, and anoxia were not linked with children’s behavior problems, our measures assessed a wide range of complication severity, and effects may be evident only at the severe end of the spectrum.

Limitations and Strengths

There are a number of limitations of the present study that should be considered. First, children in this study showed behavior that places them at risk for later ADHD and ODD (Campbell et al., 1986), but they were not diagnosed with ADHD or ODD, limiting generalizability of findings to these disorders. Second, retrospective parent report of prenatal and perinatal complications and family history of behavior problems were used, and fathers were not always present to report on paternal family histories. Our study represents a step forward by using fathers’ reports of paternal family history for the majority of families (compared to previous studies that have used mothers’ reports), but parents may not accurately recall their siblings’ or their own childhoods. Third, behavior groups were created using maternal reports. This design had the advantage of allowing for an examination of cross-informant validity; however, it is possible that the observed associations between maternal reports of behavior and maternal reports of biological risk factors could have been inflated by method variance. Fourth, although the overall sample was large, sample sizes for the HYP and OD groups for teacher and father measures were small, limiting power for these analyses. Similarly, because group sizes were relatively small, and to remain consistent with previous studies in this area, corrections for Type I error were not made. Therefore, caution should be taken in interpreting results, and findings with weaker significance need replication. Fifth, this was not an epidemiological study; the percentage of parents who returned screening packets was likely not high enough to draw conclusions regarding prevalence rates of these problems, and may limit generalizability. Finally, although ADHD is characterized by both inattention and hyperactivity, this study examined only hyperactivity. To the extent that the present study may provide insight into the early development of symptoms associated with ADHD, it does not address the early development of symptoms associated with ADHD predominantly inattentive type.

Despite these limitations, the present study adds to the small body of research comparing preschool-aged children with hyperactivity, oppositional-defiance, and comorbid behavior problems. It is one of very few studies to compare these subtypes among younger preschool-aged children, and is the first to examine cross-informant validity of maternal reports of these subtypes in 3-year-olds. It is also the first to compare these groups on prenatal and perinatal factors and family history of behavior problems. In fact, even among older children, these variables have rarely been compared across subtypes of behavior problems (Waschbusch, 2002).

Implications and Future Directions

The present study suggests that as early as age 3, developmentally deviant levels of pure hyperactivity and comorbid hyperactivity/oppositional-defiance are associated with biological risk factors that are thought to contribute to ADHD. This provides some support for the validity of both hyperactive subtypes, particularly in conjunction with evidence of generally good cross-informant validity. The degree to which these two subtypes are distinct from each other is unclear, and will be addressed more fully in Parts II and III of this series, though the present study did provide moderate cross-informant validity for the distinction. Note that while these results support the validity of hyperactive and comorbid hyperactive/oppositional-defiant subtypes as clinically significant disturbances among young preschoolers, children with these subtypes may not have, or go on to develop, ADHD. The mean differences found in our study are likely accounted for by the subset of children in the hyperactive groups who are showing early signs of ADHD. Nonetheless, the results suggest that age 3 may not be too early to identify children who are at risk for ADHD, and highlight the need to develop assessment procedures that will discriminate transient from stable early behavior problems, taking into account comorbidity between hyperactivity and oppositional-defiance.

References

Abikoff, H. B., Jensen, P. S., Arnold, L. L. E., Hoza, B., Hechtman, L., Pollack, S., et al. (2002). Observed classroom behavior of children with ADHD: Relationship to gender and comorbidity. Journal of Abnormal Child Psychology, 30, 349-359.

Allen, N. B., Lewinsohn, P. M., & Seeley, J. R. (1998). Prenatal and perinatal influences on risk for psychopathology in childhood and adolescence. Development & Psychopathology, 10, 513-529.

Applegate, B., Lahey, B. B., Hart, E., Biederman, J., Hynd, G. W., Barkley, R. A., et al. (1997). Validity of the age-of-onset criterion for ADHD: a report from the DSM-IV field trials. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 1211-1222.

August G. J., Braswell, L., & Thuras, P. (1998). Diagnostic stability of ADHD in a community sample of school-aged children screened for disruptive behavior (attention deficit hyperactivity disorder). Journal of Abnormal Child Psychology, 26, 345-360.

Barkley, R. A. (1990). Attention-deficit hyperactivity disorder: A handbook for diagnosis and treatment. New York: Guilford Press.

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

Barkley, R. A. (2006). Attention-deficit hyperactivity disorder: A handbook for diagnosis and treatment, third edition. New York: Guilford Press.

Barkley, R. A., Fischer, M., Edelbrock, C., & Smallish, L. (1991). The adolescent outcome of hyperactive children diagnosed by research criteria-III. Mother-child interactions, family conflicts, and maternal psychopathology. Journal of Child Psychology and Psychiatry, 32, 233-255.

Barkley, R. A., & Murphy, K. R. (1998). Attention-deficit hyperactivity disorder: A clinical workbook. New York: Guilford Press.

Burt, S. A., McGue, M., Krueger, R. F., & Iacono, W. G. (2005). Sources of covariation among the child-externalizing disorders: Informant effects and the shared environment. Psychological Medicine, 35, 1133-1144.

Campbell, S. B. (1995). Behavior problems in preschool children: A review of recent research. Journal of Child Psychology and Psychiatry, 36, 113-149.

Campbell, S. B., Ewing, L. J., Breaux, A. M., & Szumowski, E. K. (1986). Parent-referred problem three-year-olds: Follow-up at school entry. Journal of Child Psychology and Psychiatry, 27, 473-488.

Costello, E. J., Mustillo, S., Erkanli, A., Keeler, G., & Angold, A. (2003). Prevalence and development of psychiatric disorders in childhood and adolescence. Arch Gen Psychiatry, 60, 837-844.

Coy, K., Speltz, M. L., DeKlyen, M., & Jones, K. (2001). Social-cognitive processes in preschool boys with and without oppositional defiant disorder. Journal of Abnormal Child Psychology, 29, 107-119.

Cunningham, C. E., & Boyle, M. H. (2002). Preschoolers at risk for attention-deficit hyperactivity disorder and oppositional defiant disorder: Family, parenting, and behavioral correlates.  Journal of Abnormal Child Psychology, 30, 555-569.

Dodge, K. A., & Pettit, G. S. (2003). A biopsychosocial model of the development of chronic conduct problems in adolescence. Developmental Psychology, 39, 349-371.

DuPaul, G. J., Power, T. J., McGoey, K. E., Ikeda, M. J., & Anastopoulos, A. D. (1998). Reliability and validity of parent and teacher ratings of attention-deficit/hyperactivity disorder symptoms. Journal of Psychoeducational Assessment, 16, 55-68.

Egger, H. L., & Angold, A. (2006). Common emotional and behavioral disorders in preschool children: Presentation, nosology, and epidemiology. Journal of Child Psychology and Psychiatry, 47, 313-337.

Frick, P. J., Lahey, B. B., Christ, M. 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 Psychology, 20, 445-451.

Friedman-Weieneth, J. L., Doctoroff, G. L., Harvey, E. A., & Goldstein, L. H. (2006). The Disruptive Behavior Rating Scale-Parent Version (DBRS-PV): Factor analytic structure and validity among young preschool children. Manuscript submitted for publication.

Gadow, K., & Nolan, E. (2002). Differences between preschool children with ODD, ADHD and ODD+ADHD symptoms.  Journal of Child Psychology & Psychiatry, 43, 191-201.

Hart, E. L., Lahey, B. B., Loeber, R., Applegate, B., Green, S. M., & Frick, P. J. (1995). Developmental change in attention-deficit hyperactivity disorder in boys: A four-year longitudinal study. Journal of Abnormal Child Psychology, 23, 729-749.

Hartsough, C. S., & Lambert, N.M. (1985). Medical factors in hyperactive and normal children: Prenatal, development, and health history findings. American Journal of Orthopsychiatry, 55, 190-201.

Jensen, P. S., Hinshaw, S. P., Kraemer, H. C., Lenora, N., Newcorn, J, Abikoff, H., et al. (2001). ADHD comorbidity findings from the MTA study: Comparing comorbid subgroups. Journal of the American Academy of Child and Adolescent Psychiatry, 40, 147-158.

Johnston, C. (1996). Parent characteristics and parent-child interactions in families of non-problem children and ADHD children with higher and lower levels of oppositional-defiant behavior. Journal of Abnormal Child Psychology, 24, 85-104.

Keenan, K., & Wakschlag, L. S. (2000). More than the terrible twos: The nature and severity of behavior problems in clinic-referred preschool children. Journal of Abnormal Child Psychology, 28, 33-46.

Keenan, K., & Wakschlag, L. S. (2004). Are oppositional defiant and conduct disorder symptoms normative behaviors in preschoolers? A comparison of referred and nonreferred children. American Journal of Psychiatry, 161, 356-358.

Keppel, G., & Wickens, T. D. (2004). Design & analysis: A researcher’s handbook. Upper Saddle River, NJ: Prentice Hall

Kim-Cohen, J., Arseneault, L., Caspi, A., Tomás, M., Taylor, A., & Moffitt, T. E. (2005). Validity of DSM-IV conduct disorder in 4 ½-5-year-old children: A longitudinal epidemiological study. American Journal of Psychiatry, 162, 1108-1117.

Lahey, B. B., & Loeber, R. (1994). Framework for a developmental model of oppositional defiant disorder and conduct disorder. In D. K. Routh (Ed.), Disruptive behavior disorders in childhood (pp. 139-180). New York: Plenum Press.

Lahey, B. B., Pelham, W. E., & Loney, J. (2004). Three-year predictive validity of DSM-IV attention deficit hyperactivity disorder in children diagnosed at 4-6 years of age. American Journal of Psychiatry, 161, 2014-2020.

Lahey, B. B., Pelham, W. E., Stein, M. A., Loney, J., Trapani, C., Nugent, K., et al. (1998). Validity of DSM-IV attention-deficit/hyperactivity disorder for younger children. Journal of the American Academy of Child and Adolescent Psychiatry, 37, 695-703.

Lahey, B. B., & Waldman, I. D. (2003). A developmental propensity model of the origins of conduct problems during childhood and adolescence. In Lahey, B.B.. Moffitt, T.E., Caspi, A. Causes of conduct disorder and juvenile delinquency (p 76-117). New York, NY: Guilford Press

Lavigne, J. V., Arend, R., Rosenbaum, D., Binns, H. J., Christoffel, K. K., & Gibbons, R. D. (1998). Psychiatric disorders with onset in the preschool years: I. Stability of diagnoses.  Journal of the American Academy of Child & Adolescent Psychiatry, 37, 1246-1254.

Lochman, J. E., & Conduct Problems Prevention Research Group. (1995). Screening of child behavior problems for prevention programs at school entry. Journal of Consulting and Clinical Psychology, 63, 549-559.

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

Moffitt, T. E. (1993). The neuropsychology of conduct disorder. Development and Psychopathology, 5, 135-151.

Nagin, D., & Tremblay R. E. (1999). Trajectories of boys' physical aggression, opposition, and hyperactivity on the path to physically violent and nonviolent juvenile delinquency. Child Development, 70, 1181-1196.

Orlebeke, J. F., Knol, D. L., & Verhulst, F. C. (1999). Child behavior problems increased by maternal smoking during pregnancy. Archives of Environmental Health, 54, 15-20.

Ostrander, R., Weinfurt, K. P., Yarnold, P. R., & August, G. J. (1998). Diagnosing attention deficit disorders with the Behavioral Assessment System for Children and the Child Behavior Checklist: Test and construct validity analyses using optimal discriminant classification trees, Journal of Consulting and Clinical Psychology, 66, 660-672.

Patterson, G. R. (1992). Developmental changes in antisocial behavior. In Peters, R.D., McMahon, R.J., Quinsey, V.L. Aggression and violence throughout the life span. (p 52-82). Thousand Oaks, CA: Sage Publications

Reynolds, C. R., & Kamphaus, R. W. (1992). Behavior Assessment System for Children. Circle Pines, MN: American Guidance Service

Ross, C. N., Blanc, H. M., McNeil, C. B., Eyberg, S. M., & Hembree-Kigin, T. L. (1998). Parenting stress in mothers of young children with oppositional defiant disorder and other severe behavior problems. Child Study Journal, 28, 93-110.

Schwebel, D. C., Speltz, M. L., Jones, K., & Bardina, P. (2002). Unintentional injury in preschool boys with and without early onset of disruptive behavior. Journal of Pediatric Psychology, 27, 727-737.

Shaffer, D., Fisher, P., Lucas, C. P., Dulcan, M. K., & Schwab-Stone, M. E. (2000). NIMH Diagnostic Interview Schedule for Children Version IV (NIMH DISC-IV): Description, differences from previous versions, and reliability of some common diagnoses. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 28-38.

Silberg, J. L., Rutter, M., & Meyer, J. (1996). Genetic and environmental influences on the covariation between hyperactivity and conduct disturbance in juvenile twins. Journal of Child Psychology and Psychiatry, 37, 803-816.

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

Sonuga-Barke, E. J. S., Thompson, M., Stevenson, J., & Viney, D. (1997). Patterns of behaviour problems among pre-school children. Psychological Medicine, 27, 909-918.

Speltz, M. L., DeKlyen, M., Calderon, R., Greenberg, M. T., & Fisher, P. A. (1999). Neuropsychological characteristics and test behaviors of boys with early onset conduct problems. Journal of Abnormal Psychology, 108, 315-325.

Speltz, M. L., DeKlyen, M., & Greenberg, M. T. (1999). Attachment in boys with early onset conduct problems. Development and Psychopathology, 11, 269-285.

Speltz, M. L., McClellan, J., DeKlyen, M., & Jones, K. (1999). Preschool boys with oppositional defiant disorder: Clinical presentation and diagnostic change. Journal of the American Academy of Child & Adolescent Psychiatry, 38, 838-845.

Sprich-Buckminster, S., Biederman, J., & Milberger, S., Faraone, S. V., & Lehman, B. K. (1993). Are perinatal complications relevant to the manifestation of ADD? Issues of comorbidity and familiarity. Journal of the American Academy of Child & Adolescent Psychiatry, 32, 1032-1037.

Stormont-Spurgin, M., & Zentall, S. S. (1995). Contributing factors in the manifestation of aggression in preschoolers with hyperactivity. Journal of Child Psychology and Psychiatry, 36, 491-509.

U.S. Census Bureau (2006). Fact sheets for Hampshire, Hampden, and Franklin County, Massachusetts.

Wakschlag, L. S., & Keenan, K. (2001). Clinical significance and correlates of disruptive behavior in environmentally at-risk preschoolers. Journal of Clinical Child Psychology, 30, 262-275.

Waschbusch, D. A. (2002). A meta-analytic examination of comorbid hyperactive-impulsive-attention problems and conduct problems. Psychological Bulletin, 1, 118-150.

Zappitelli, M., Pinto, T., & Grizenko, N. (2001). Pre-, peri-, and postnatal trauma in subjects with Attention-Deficit Hyperactivity Disorder. Canadian Journal of Psychiatry, 46, 542-548.

Zito, J. M., Safer, D. J., dosReis, S., Gardner, J. F., Boles, M., & Lynch, F. (2000). Trends in the prescribing of psychotropic medications to preschoolers. JAMA, 283, 1025-1030.

Table 1

Comparison of Subtypes on Demographic Variables and Symptom Severity

|Demographic or symptom |1. HYP | 2.OD |3. HYP/OD |4. Nonprob |F or Χ2 |Tukey HSD |

|severity |#(%) or M(SD) |#(%) or M(SD) |#(%) or M(SD) |#(%) or M(SD) | | |

|Child gender |n = 41 |n = 24 |n = 96 |n = 59 |1.30 | |

| Girls |16 (39%) |11 (46%) |47 (49%) |29 (49%) | | |

| Boys |25 (61%) |13 (54%) |49 (51%) |30 (51%) | | |

|Child ethnicity | | | | |32.14*** | |

| Eur. American |26 (63%) |17 (71%) |33 (34%) |40 (68%) | | |

| Latino |6 (15%) |0 (0%) |33 (34%) |4 (7%) | | |

| African American |5 (12%) |3 (13%) |12 (13%) |7 (12%) | | |

| Multi-ethnic |4 (10%) |4 (17%) |18 (19%) |8 (14%) | | |

|Mother education |14.24 (2.20) |14.79 (3.22) |12.39 (2.70) |14.20 (2.25) |10.74*** |3 < 1***, 2***, 4*** |

|Father education |13.83 (2.71) |14.96 (3.13) |12.59 (2.27) |13.45 (2.33) |5.82** |3 < 2** |

|Income (sqrt. $)a |232.98 (71.75) |254.03 (73.34) |199.42 (61.81) |257.62 (68.16) |11.03*** |3 < 1*, 2**, 4*** |

|Married |29 (71%) |21 (88%) |56 (58%) |45 (76%) |10.71* | |

|BASC HYP screen |71.78 (9.06) |59.67 (8.87) |76.55 (9.63) |48.20 (8.76) |--b |3 > 1*; 2 > 4***b |

|BASC Agg. screen |56.68 (10.67) |73.58 (12.44) |79.91 (14.87) |49.22 (9.87) |--b |1 > 4*b |

Note. HYP = Hyperactive, OD = Oppositional-defiant, HYP/OD = Hyperactive and oppositional-defiant, Nonprob = Non-problem, sqrt = square root, Agg. = Aggression

aTwo family incomes were outliers ( > $300,000) and were changed to the next highest value.

b Comparisons are presented only for groups that would not be different by definition.

*p < .05, **p < .01, ***p < .001

Table 2

Comparison of Groups on Fathers’, Teachers’, and Observers’ Ratings of Children’s Behavior

| |1. HYP |2.OD |3. HYP/OD |4. Nonprob |F |Partial Eta squared|HSD/Planned Comparisons |

|Behavior Variable |M(SD) |M(SD) |M(SD) |M(SD) | | | |

|Father reports a |n = 30 |n = 19 |n = 60 |n = 40 | | | |

| Hyperactivity |0.19 (.88) |-0.28 (.77) |0.60 (.88) |-0.78 (.56) |25.55*** |.35 |1, 3 > 4***; 2 < 1†, 3*** |

| | | | | | | |3 > 1† |

| Oppositional-def. |-0.20 (.81) |0.09 (.86) |0.60 (.87) |-0.74 (.59) |24.03*** |.33 |2, 3 > 4***; 3 > 1***, 2† |

|Teacher BASC |n = 20 |n = 14 |n = 35 |n = 27 | | | |

| Hyperactivity |51.10 (9.15) |45.14 (4.80) |52.94 (10.78) |48.85 (7.18) |2.93* |.09 |2 < 1*, 3**; 3 > 4† |

| Aggression |49.75 (8.17) |49.14 (6.34) |56.26 (12.70) |50.59 (8.91) |2.95* |.09 |3 > 1*, 4* |

|Behavior during testing and |n = 41 | | | | | |

|mother-child interactions | |n = 24 |n = 96 |n = 59 | | |

| Noncompliance |0.07 (.88) |-0.25 (.57) |0.25 (.88) |-0.30 (.69) |6.67*** |.09 |3 > 2*, 4***; 1 > 4* |

| | | | | | | |Table continues |

|Table 2 continued | | | | | | | |

| |1. HYP |2.OD |3. HYP/OD |4. Nonprob |F |Partial Eta squared|HSD/Planned Comparisons |

|Behavior Variable |M(SD) |M(SD) |M(SD) |M(SD) | | | |

| Activity level |0.51 (.31) |0.27 (.18) |0.40 (.28) |0.34 (.20) |5.46** |.07 |1 > 2***, 3†, 4**; 3> 2** |

| Defiance/verbal aggression |0.23 (.89) |-0.00 (.58) |0.15 (.91) |-0.26 (.54) |4.16** |.06 |4 < 1*, 3** |

| Negative affect |0.14 (.84) |-0.09 (.48) |0.15 (.69) |-0.24 (.52) |4.86** |.06 |4 < 1*, 3*** |

|Audiotaped father-child |n = 27 |n = 17 |n = 46 |n = 30 | | | |

|interaction | | | | | | | |

| Noncompliance |1.62 (1.19) |1.07 (.78) |1.75 (1.29) |1.33 (1.16) |1.78 |.04 | |

| Verbal aggression |0.45 (1.10) |-0.32 (.67) |-0.05 (1.01) |-0.11 (.87) |2.81* |.07 |1 > 2† |

| Negative affect |0.40 (.86) |-0.09 (.70) |0.08 (.86) |-0.50 (.73) |6.13** |.14 |4 < 1***, 3*** |

Note. HYP = Hyperactive, OD = Oppositional-defiant, HYP/OD = Hyperactive and oppositional-defiant, Nonprob = Non-problem, BASC = Behavior Assessment System for Children.

aThese scores are based on averages of standardized scores on the BASC and DBRS.

†p < .10, *p < .05, **p < .01, ***p < .001

Table 3

Comparisons of Groups on Family History of Attention and Disruptive Behavior Problems and Prenatal and Perinatal Factors

|Family history or prenatal/perinatal |1. HYP |2. HYP/OD |3. Nonprob |F or Χ2 |Partial Eta2/ Cramer’s |HSD/Planned Comparisons |

|variable |M(SD) or # (%) |M(SD) or # (%) |M(SD) or # (%) | |Phi | |

|Family History |n = 39 |n = 95 |N = 58 | | | |

| Hyperactivity symptoms |0.19 (.74) |0.32 (.84) |-0.48 (.46) |21.73*** |.19 |1, 2 > 3*** |

| ODD symptoms |0.08 (.70) |0.32 (.85) |-0.42 (.53) |16.28*** |.15 |1 > 3**; 2 > 3*** |

| CD symptoms |13 (33%) |50 (53%) |11 (19%) |12.40*** |.30 |2 > 3*** |

|Prenatal/perinatal factors |n = 41 |n = 96 |n = 59 | | | |

| Total score |1.52 (.73) |1.59 (.83) |1.06 (.82) |5.98** |.06 |1 > 3*; 2 > 3*** |

| Prenatal physical health |1.09 (.65) |0.99 (.64) |0.76 (.66) |3.24*a |.03 |1 > 3*; 2 > 3† |

| Prenatal emotional health |9 (22%) |41 (43%) |3 (5%) |26.05*** |.37 |1 > 3*; 2 > 3*** |

| Substance use |8 (20%) |26 (27%) |4 (7%) |6.75* |.22 |1 > 3†, 2 > 3* |

| Difficult delivery |2 (5%) |5 (5%) |5 (8%) |.55 |.07 | |

| Prematurity |7 (17%) |16 (17%) |12 (20%) |.93 |.04 | |

| Anoxia |10 (24%) |27 (28%) |10 (17%) |2.54 |.11 | |

Note. HYP = Hyperactive, HYP/OD = Hyperactive and oppositional-defiant, Nonprob = Non-problem.

aThis effect became a trend when marital status (p = .07) and ethnicity (p = .11) were entered as blocking variables.

†p < .10, *p < .05, **p < .01, ***p < .001

Figure 1. Model of development of ADHD, ODD, & CD

ODD/CD ADHD/ODD/CD ADHD

Note: Although moderator effects are typically modeled using this notation , in this model, they are presented using the following symbol because these variables are conceptualized as interacting with each other (i.e., combining together), rather than one variable moderating the effect of the other (i.e., the variables are conceptualized as moderating each other).

-----------------------

[1] Please address correspondence to Elizabeth Harvey, Dept. of Psychology, Tobin Hall, 135 Hicks Way, University of Massachusetts, Amherst, MA 01003; tel: 413-577-2755, fax: 413-545-0996; email: eharvey@psych.umass.edu

[2] Department of Psychology, University of Massachusetts, Amherst, MA 01003

This study was supported by National Institute of Mental Health Grant R01MH60132.

We are grateful to the families who participated in this study and to staff from physicians’ offices and community centers who assisted in recruiting families. Thanks also to the many graduate and undergraduate research assistants who assisted with data collection, and to Sally Powers and David Arnold who provided suggestions on earlier drafts of this paper.

[3] We did attempt to screen for children with elevated levels of inattention without hyperactivity. However, consistent with findings of Applegate et al. (1997), few children showed elevated attention problems without hyperactivity.

[4] One additional child was recruited into the non-problem comparison group; however, behavioral observations and child care provider ratings indicated clinically significant levels of behavior problems, so he was not included.

[5] Response rates could not be calculated for sites where packets were presented in a display for interested parents to take. At sites where packets were delivered directly to parents of 3-year-olds (e.g., via mail and through some pediatric practices), 20% of packets were completed and returned. However, this likely underestimates the response rate because many parents may have received more than one packet.

[6] Partway through the recruitment phase, we began to use a shortened version of the BASC-PRS at the screening, because staff at recruitment sites reported that the length of the BASC seemed to be a deterrent to many parents with lower education. The shortened BASC contained items from the hyperactivity, aggression, attention problems, anxiety, and depression subscales. The full BASC-PRS was administered to all families at the first home visit.

[7] Because we had difficulty recruiting enough Latino children who were eligible for the comparison group to fully match on ethnicity, European American children were selected to match some of the Latino externalizing children.

[8]Allen et al. (1998) also included caffeine use in the substance use category and local anesthesia in the difficult delivery category. We did not include them because incidence rates were so high for each of these.

[9] Mothers were not able to report on the child’s fathers’ history of behavior problems in 29 of the families. Of the remaining 229 families, fathers were present to report on their symptoms for 147 families. An additional 82 mothers reported that they had sufficient second-hand information about the fathers’ childhood behavior to provide ratings. One hundred ninety-three families reported on the biological fathers’ siblings, and the father was present for 140 of these families. Ratings of paternal and paternal siblings’ hyperactivity, ODD, or CD symptoms did not differ significantly as a function of fathers’ presence during the interview (all p’s > .10).

[10] In the few cases in which only the father completed the DISC or the screening BASC, these measures were not included in the index, so that mothers’ reports could be validated using fathers’ reports.

[11] We considered whether observations in some settings (testing, structured vs. unstructured observation) might be more sensitive to differences across groups, particularly given these modest intercorrelations. Patterns of differences were in fact similar across settings, except that the HYP/OD group was rated as significantly more active than the non-problem group during the videotaped mother-child interaction (p < .05), but not during testing (p = .99).

[12]Because both teachers and fathers identified children in the HYP/OD group as more oppositional-defiant/ aggressive than children in8ÂÕ×óô ! # B f g h u v } ? ‹ £ ° ± the HYP group, but observational data did not, further analyses were conducted to explore this discrepancy. Chi-square tests were conducted to compare children in the HYP and HYP/OD groups on the presence or absence of each of the 8 DISC ODD symptoms. The six symptoms that were significant (all p’s < .001) involved displays of anger and hostile behavior toward others; differences were not significant for argues (p = .11) and defiance (p = .20). This could explain why the HYP/OD and HYP groups did not differ on observed defiance/verbal aggression, but does not account for findings on observed negative affect.

[13]We have chosen to use blocking variables rather than entering education as a covariate because of the potential pitfalls of using ANCOVAs to adjust for pre-existing group differences (Keppel & Wickens, 2004).

-----------------------

Genetics

(family history of ADHD)

Biological influences (prenatal risk factors)

Genetics (family history of ODD/CD)

Family Stress

Time

Antisocial Behavior

Early child characteristics

---------------------------------

ODD symptoms

---------------------------------

Intermediate CD symptoms

---------------------------------

Advanced CD symptoms

Behavioral disinhibition

Parenting

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download