Attention-deficit/hyperactivity disorder (ADHD) is characterized by impairing levels of inattention and/or hyperactivity/impulsivity (American Psychiatric Association 2013) and affects an estimated 7.2 % of children and adolescents in the United States (Thomas et al. 2015). Oppositional defiant disorder (ODD), which is characterized by pervasive anger, irritability, arguing, and defiance, is commonly comorbid with ADHD (e.g., Lavigne et al. 2009). It has been well established that parents of children with ADHD have elevated symptoms themselves (e.g., Starck et al. 2016). However, there are important gaps in our understanding of the relation between parental ADHD and the development of ADHD and ODD symptoms in children. First, much of the research in this area has been conducted with older children, after symptoms are well-established. Examining whether parental ADHD symptoms play a role in the early development of children’s ADHD and ODD symptoms is important. Second, it is unknown if correlates of adult ADHD, including family history of ADHD, comorbid psychopathology, and family adversity, could account for the relation between parental ADHD and child ADHD and ODD. Finally, examining mediators and moderators of the relation between parental ADHD and child symptoms can provide important insights into the processes through which parental ADHD contributes to children’s development. The present study seeks to address these gaps in the literature.

The Relation between Parental ADHD and Child ADHD and ODD Symptoms

The link between parent and child ADHD symptoms is well established. Parents of children with ADHD report higher levels of ADHD symptoms than parents of children without ADHD (e.g., Starck et al. 2016). Conversely, child ADHD is more severe when parents are also affected (e.g., Goos et al. 2007). Studies examining dimensional ratings of ADHD have generally shown moderate-sized correlations between parent and child ADHD symptoms (e.g., van Lieshout et al. 2016), though there is some evidence that the relation may vary as a function of parent gender. Some studies have found parental ADHD symptoms to be more strongly associated with child ADHD symptoms for fathers (Macek et al. 2012; Takeda et al. 2010), whereas others have found stronger relations for mothers (Agha et al. 2013; Goos et al. 2007; Segenreich et al. 2015). Further research is needed to explore possible differences between mothers and fathers.

Though somewhat less well-studied, parental ADHD symptoms have also been associated with children’s ODD. Cross-sectional studies have demonstrated that greater parental ADHD symptomology is related to more ODD symptoms in children (e.g., Ellis and Nigg 2009; Goos et al. 2007; Humphreys et al. 2012; Zisser and Eyberg 2012). When these studies examined both mothers and fathers, maternal ADHD tended to be more strongly associated with ODD than paternal ADHD. However, there is evidence that fathers’ ADHD symptoms are strongly associated with ODD/conduct disorder (CD) when fathers are highly involved in their children’s care (Romirowsky and Chronis-Tuscano 2014). Longitudinal research in this area is sparse, but Brammer et al. (2016) did not find parental ADHD to predict later child ODD symptoms. Research is needed to build on this body of research to longitudinally examine the role of parental ADHD in the development of child ADHD and ODD symptoms.

Factors that May Account for the Relation between Parental ADHD and Child ADHD and ODD Symptoms

Shared Family History

Children with ADHD have been consistently shown to have substantially elevated rates of ADHD in their family histories (e.g., Biederman et al. 2008). This may reflect a combination of genetic, environmental, and behavioral risk factors that are passed down intergenerationally (Doerr and Teng 2012). Because parents and children have overlapping family histories, these shared risk factors likely contribute both to parent and child ADHD symptoms. Thus, the relation between parent and child ADHD symptoms may be accounted for by a shared family history of ADHD; however, to our knowledge, this has not been examined.

Parental Comorbid Psychopathology

Parents of children with ADHD and ODD have higher rates of psychopathology (e.g., Johnston and Mash 2001). In turn, parental psychopathology is thought to increase children’s risk for ADHD and ODD through a combination of genetic liability for psychopathology and environmental risk (Bornovalova et al. 2010). Maternal depression and anxiety have been consistently associated with child ADHD and ODD in cross-sectional (e.g., Segenreich et al. 2015; Tully et al. 2008) and prospective studies (Ashman et al. 2008; O’Connor et al. 2002; Sciberras et al. 2011). Paternal antisocial symptoms are also associated with trajectories of higher levels of hyperactivity-impulsivity and inattention across early childhood (e.g., Galéra et al. 2011) and with ODD in children (e.g., Blazei et al. 2008; Frick et al. 1992). It is possible that in parents with ADHD and comorbid psychopathology, these other disorders account for increased risk of ADHD and ODD in their children. However, it is also possible that the association between parental psychopathology and child ADHD and ODD is due to parental ADHD. Few studies have examined how parental ADHD is related to child ADHD and ODD controlling for other parental psychopathology. Tung et al. (2014) controlled for parent depression and found that the relation between parent and child ADHD in school-age children remained. More research is needed to determine the unique contribution of parental ADHD and comorbid psychopathology to the early development of ADHD and ODD in children.

Family Adversity

There is greater family adversity among children with ADHD and ODD (e.g., Counts et al. 2005; Rydell 2010) and parents of children with ADHD report higher levels of stress (e.g., DuPaul et al. 2001). Although there has been a clear link between child ADHD and family adversity, there is less evidence that parental ADHD is associated with adversity (e.g., Agha et al. 2013). To our knowledge, studies have not evaluated whether the relation between parent and child ADHD can be accounted for by these family factors. However, when controlling for factors like socioeconomic status (SES) and family intactness, familial risk of ADHD in first-degree relatives of children with ADHD remains (Biederman et al. 1992). More research is needed to establish whether family adversity may account for the relation between parental ADHD and child ADHD and ODD.

Moderators of the Relation between Parental ADHD and Child ADHD and ODD

Although family history of ADHD, comorbid parental psychopathology, and family adversity have all been linked with child ADHD and ODD, these relations are generally moderate in size. Thus, children with ADHD and ODD vary in their family functioning, with some exhibiting strong family histories of ADHD, high levels of parental psychopathology, and high family adversity. The effects of parental ADHD on children are likely to vary depending on the larger family context. In particular, parental ADHD may be more likely to lead to the development of child ADHD and ODD when combined with other family risk factors. For example, children with a strong family history of ADHD and with greater family adversity may be more vulnerable to adverse effects of parental inattention and impulsivity. Similarly, just as comorbid child psychopathology confers poorer prognoses for children with ADHD (e.g., Larson et al. 2011), parental ADHD may result in greater risk for children when paired with other forms of parental psychopathology. To our knowledge, no studies have examined moderators of the relation between parental ADHD and child ADHD and ODD. However, research does suggest that more positive family environments buffer the negative effects of stressors more generally (see Grant et al. 2006) and parental psychopathology specifically (e.g., Goodman et al. 2011).

Parenting Practices as a Mediator of the Relation between Parental ADHD and Child ADHD and ODD

It is also critical to understand the pathways through which parental ADHD contributes to children’s development. One potentially important pathway is via disruptions in parenting. Parents who have more ADHD symptoms report more ineffective styles of discipline (Banks et al. 2008), including more overreactivity (Chen and Johnston 2007; Harvey et al. 2003) and lax/inconsistent discipline (Chronis-Tuscano et al. 2008; Harvey et al. 2003). Maternal ADHD symptoms have also been associated with less positive reinforcement (Chen and Johnston 2007), as well as less positive affect and praise and more negative commands and critical statements (Chronis-Tuscano et al. 2008). In turn, negative parenting practices, including low positive regard, intrusiveness, low sensitivity (Keown 2012), and inconsistent parenting (Hawes et al. 2013) have been shown to predict later child ADHD symptoms. However, only a handful of studies have directly tested whether parenting mediates the relation between parental ADHD and child ADHD and ODD symptoms. In a prospective study, Tung et al. (2014) found that corporal punishment mediated the relation between parent and child ADHD. Chronis-Tuscano et al. (2011) found that negative parenting mediated the relation between maternal ADHD symptoms and reductions in child disruptive behavior symptoms following parent training for children with ADHD. Research is needed to build on these studies to examine parenting as a mediator of the relation between parental ADHD and child ADHD and ODD symptoms.

The Present Study

The present study sought to better understand the relation between parental ADHD and child ADHD and ODD symptoms. We examined these relations during the preschool years, when ADHD and ODD symptoms often first emerge (Applegate et al. 1997). In particular, the following questions were addressed:

  1. 1)

    Does parental ADHD symptomatology predict later child ADHD and ODD symptoms, controlling for family history of ADHD symptoms, family adversity, and comorbid parental psychopathology symptoms? We expected that maternal and paternal ADHD symptoms would predict child ADHD and ODD symptoms 3 years later, controlling for early child ADHD symptoms, family history of ADHD symptoms, comorbid parental psychopathology, and family adversity.

  2. 2)

    Do family history of ADHD symptoms, comorbid parental psychopathology, and family adversity moderate the relation between parental ADHD symptoms and child ADHD and ODD symptoms? It was predicted that the relation between parental ADHD and child ADHD and ODD symptoms would be stronger in families with a greater family history of ADHD, more parental psychopathology, and greater family adversity.

  3. 3)

    Do parenting practices mediate the relation between parental ADHD and child ADHD and ODD symptoms? We expected that negative parenting practices (overreactivity, laxness, negative affect, low warmth) would mediate the relation between parental ADHD and child ADHD and ODD symptoms, such that greater parental ADHD would predict more negative parenting practices, which in turn would predict more child ADHD and ODD symptoms.

Method

Participants

Participants included 258 (138 boys; 120 girls) 3-year-old children (M = 44.13 months, SD = 3.39) and their parentsFootnote 1 who took part in a 3-year longitudinal study aimed at understanding the early development of ADHD and ODD among preschoolers. At the time of screening, 199 children had significant externalizing problems (hyperactivity and/or aggression) and 59 children did not have behavior problems. The sample was ethnically diverse, including European American (55.4 %), Latino American (17.8 %; mostly Puerto Rican), African American (12 %), and multiethnic (14.7 %) children. The median family income was $48,000. The majority of mothers (87.5 %) and fathers (89.4 %) had high school diplomas, with a substantial minority of mothers (33.5 %) and fathers (28.7 %) having bachelor’s degrees or higher. Most (86 %) families were two-parent families. Of the 168 behavior problem children who participated in the 3-year follow-up assessments, 36 (20 boys) met criteria for ADHD only, 22 (13 boys) met criteria for ODD only, and 39 (26 boys) met criteria for ADHD and ODD.Footnote 2

Procedure

Participants were recruited over a 3-year period by distributing screening questionnaires through state birth records, pediatrician offices, child care centers, and community centers throughout western Massachusetts. Children with significant externalizing problems were recruited from 1752 3-year-old children whose parents completed a screening packet containing the Behavior Assessment System for Children–Parent Report Scale (BASC-PRS; Reynolds and Kamphaus 1992) and a questionnaire assessing for exclusion criteria, parental concern about externalizing symptoms, and demographics. Inclusion criteria for the behavior problem group included: (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). Inclusion criteria for the control group included: (a) parent responded no to, “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 60. Participants had no evidence of intellectual disability, hearing or visual impairment, language delay, cerebral palsy, epilepsy, autism, or psychosis.

At Time 1 (T1), families were scheduled for two 3-hr home visits scheduled approximately 1 week apart. Each parent was paid for participation in the study. Home visits were then conducted annually, with 243 families participating at Time 2 (T2), 215 families participating at Time 3 (T3), and 223 families participating at Time 4 (T4). The study was conducted in compliance with the University of Massachusetts Amherst’s Institutional Review Board, and written informed consent was obtained from all participating parents.

Measures

Child ADHD and ODD

Parents completed the National Institute of Mental Health Diagnostic Interview Schedule for Children–Fourth Edition (DISC-IV; Shaffer et al. 2000) at T1 and T4. At T1, only the ADHD and ODD sections were administered, with minor modifications made to school-related questions. The full computerized version of the DISC-IV was administered at T4. Interviews were administered jointly to mothers and fathers when both were available.Footnote 3 From these interviews, we created T1 child ADHD and ODD and T4 child ADHD and ODD variables which represent the symptom counts for child ADHD (out of 18 symptoms) and child ODD (out of 8 symptoms). Good reliability was found in this sample for ADHD (α = 0.88) and ODD (α = 0.79) symptoms.

Parental ADHD

Parents completed the Current Symptoms Scale (Barkley and Murphy 1998) at T1 to assess their own ADHD symptoms. The Current Symptoms Scale is an 18-item scale corresponding to DSM-IV ADHD symptoms. Parents rated their symptoms on a 4-point Likert scale ranging from 0 (rarely or never) to 3 (very often), with a 2 or 3 scored as symptom endorsement. The 18 items were used to create parental ADHD symptom counts. Although multi-informant report is ideal, particularly in light of evidence that adults with ADHD may underreport their symptoms (Barkley et al. 2002), self-reports of ADHD symptoms in adults have been shown to correlate highly with informant reports (Barkley et al. 2011). Good reliability was found in this sample for mothers’ (α = 0.89) and fathers’ (α = 0.85) ADHD symptoms. Approximately 12 % of mothers and 6 % of fathers reported at least 5 symptoms of inattention or hyperactivity/impulsivity.

Family History

At T1, the same caregivers who were present for the DISC-IV interview jointly completed a Family History Interview assessing for the presence of symptoms of ADHD, ODD, and CD before the age of 18 in various biological relatives of the child. The format of the scale is similar to that described by Frick et al. (1991), but was updated for the DSM-IV, and ODD symptoms were added. Parents reported the presence or absence of each symptom in the child’s biological parents, full siblings (≥18 months), and biological aunts and uncles. For each symptom, parents were given the option of responding “yes,” “no,” “don’t know,” or “not applicable.” For the present study, ratings of symptom counts in the child’s biological aunts and uncles were used to obtain a measure of family history of ADHD. Given that the number of aunts and uncles in each family varied, an average of the number of ADHD symptoms present within a family was used to represent extended family history.

Self-Report of Parenting

Mothers and fathers completed the Parenting Scale (Arnold et al. 1993) annually. Scores at T2 and T3 were used for the current study, because they temporally preceded T4 symptomatology and followed our assessment of parental ADHD. The Parenting Scale is a 30-item self-report measure, consisting of ratings on a 7-point Likert scale, with anchors that vary across items. This scale yields scores for laxness ranging from 1 (e.g., “When I say my child can’t do something… I stick to what I said “) to 7 (e.g., “When I say my child can’t do something…I let my child do it anyway”) and overreactivity, also ranging from 1 (e.g., “When my child misbehaves… I handle it without getting upset “) to 7 (e.g., “When my child misbehaves…I get so frustrated or angry that my child can see I’m upset”). Scores were calculated by averaging across items that loaded on each factor, with high scores indicating dysfunctional parenting. This scale has demonstrated good reliability and validity with preschoolers (Arnold et al. 1993) and has been validated for use with children with ADHD (Harvey et al. 2001). Parenting practices demonstrated good reliability at each time point: mother’s laxness (α = 0.80) and overreactivity (α = 0.75) at T2; mother’s laxness (α = 0.83) and overreactivity (α = .76) at T3; father’s laxness (α = 0.80) and overreactivity (α = 0.80) at T2; father’s laxness (α = 0.74) and overreactivity (α = 0.83) at T3.

Observational Measures of Parenting

Children were videotaped annually interacting with their mothers during a 5-min play task and a clean-up task. Fathers did not take part in videotaped observations because of time limitations during home visits. Data from T2 and T3 were used in the current study. Each tape was coded by two independent undergraduate research assistants (whose ratings were averaged). Global warmth ratings were used, where warmth represents the extent to which parents positively attended to their children; used praise, encouragement, and terms of endearment; conveyed affection; were supportive and available; were cheerful in mood and tone of voice; and conveyed interest, joy, and enthusiasm in the interactions. Warmth was rated on a 7-point Likert scale with anchors and descriptions at 1 (complete absence of warmth), 4 (moderate warmth), and 7 (high level of warmth). Ratings were made after each task (play and clean-up) and were averaged across the two tasks at each time point. Inter-rater reliability, as measured by intraclass correlations, averaged .69 across the two time points.

Parental Psychopathology

The Millon Clinical Multiaxial Inventory-III (MCMI-III; Millon and Davis 1997), a 175-item questionnaire that assesses DSM-IV symptoms from Axis I and II disorders, was used to measure parental reports of psychopathology symptoms at T1. In a clinical population, internal consistency has been found to range from 0.66 to 0.90 and test-retest reliabilities from 0.84 to 0.96 (Millon and Davis 1997). For the present study, we focused on parental anxiety, depression, and antisocial personality symptoms. To create the measure of anxiety symptoms, the anxiety, somatoform, and post-traumatic stress disorder subscales were each standardized and averaged. To create the measure of depression symptoms, the depressive personality disorder, dysthymia, and major depression subscales were each standardized and averaged. Good reliability was found for the dimensions used: antisocial symptoms (α = 0.80 for mothers, 0.78 for fathers); anxiety symptoms (α = 0.92 for mothers and fathers); and depression symptoms (α = 0.94 for mothers, 0.93 for fathers).

Family Stress

The Life Experiences Survey (LES; Sarason et al. 1978) is a 57-item measure of family stress that was administered at T1. Respondents rated the valence of events that occurred in the past year, with ratings ranging from −3 (extremely negative) to +3 (extremely positive), with 0 indicating that the event had no impact on the respondent. Family stress was calculated by summing across the negative valence ratings separately for mothers’ and fathers’ LES responses. For the families where both parents completed the LES, mother and father LES scores were modestly correlated (r = 0.23, p = 0.002). The LES negative events scale has demonstrated good test-retest reliability (r = 0.72; Sarason et al. 1978). Mother and father LES variables were combined into a single family adversity latent variable (see Results section).

Income

Families reported annual income at T1 was used as a proxy for family SES.

Analytic Plan

Path modeling with Mplus version 7 (Muthén and Muthén 1998-2012) was used for all analyses. The construction of these models is described in the results section. Full information maximum likelihood was used to address missing data. Model fit was evaluated by using χ2/df (< 2 indicates good model fit), Root Mean Square Error of Approximation (RMSEA; < 0.08 represents acceptable model fit and between 0.08 and .1 indicates mediocre model fit), Bentler’s Comparative Fit Index (CFI; > 0.90 indicates acceptable model fit), and Standardized Root Mean Square Residual (SRMR; < 0.08 indicates adequate model fit).

Results

Descriptive Statistics

Table 1 presents means, standard deviations, and intercorrelations for study variables.

Table 1 Descriptive statistics and correlations among family history, parental adhd, child adhd and odd, parenting, and parental psychopathology variables
Full size table

Baseline Model of Parental ADHD as a Predictor of Later Child ADHD and ODD

A base model was constructed in which T4 child ADHD and ODD symptoms were regressed on mothers’ and fathers’ ADHD symptoms and on T1 child ADHD and ODD symptoms (Fig. 1). The four T1 variables were allowed to covary with each other as were the two T4 variables. The base model was just identified, so model fit could not be assessed. Mothers’ ADHD symptoms significantly predicted later child ADHD symptoms, controlling for early child ADHD and ODD symptoms. Fathers’ ADHD symptoms significantly covaried with T1 child ADHD symptoms; however, they did not predict later child ADHD or ODD symptoms, controlling for T1 child ADHD and ODD and maternal ADHD symptoms.Footnote 4

Fig. 1
figure 1

Base model for relations between parental ADHD symptoms and later child ADHD and ODD symptoms, taking into account early child ADHD and ODD symptoms. Standardized coefficients are presented. Only significant paths and covariances are shown in figure. † p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001

Full size image

Relations Controlling for Family History of ADHD, Comorbid Parental Psychopathology, and Family Adversity

First, each control variable was added separately to the baseline model; T4 child ADHD and ODD symptoms were regressed on the family functioning variable. This determined which control variables predicted later child symptoms and therefore should be included in a combined model, because including variables that are not related to the outcome variable can inflate standard errors of other variables in the model (Wooldridge 2012). Latent variables were constructed for comorbid parental psychopathology (anxiety, depression, antisocial as indicators; constructed separately for mothers and fathers) and for family adversity (mothers’ and fathers’ LES and family income as indicators). All indicators significantly loaded on their respective factors; these loadings are not presented in the figure for ease of presentation. Family history of ADHD significantly predicted T4 child ADHD, β = 0.15, SE = 0.06, p = 0.02, and ODD, β = 0.16, SE = 0.07, p = 0.02, symptoms. Fathers’ comorbid psychopathology symptoms significantly predicted greater T4 child ADHD, β = 0.25, SE = 0.09, p = 0.01, and ODD, β = 0.34, SE = 0.09, p < 0.001, symptomsFootnote 5; mothers’ comorbid psychopathology symptoms did not predict T4 child ADHD, β = 0.10, SE = 0.09, p = 0.25, or ODD, β = 0.11, SE = 0.10, p = 0.28, symptoms.Footnote 6 Family adversity did not predict T4 child ADHD, β = −0.07, SE = 0.22, p = 0.75, or ODD, β = −0.12, SE = 0.23, p = 0.60, symptoms. Thus, family history of ADHD and paternal comorbid psychopathology symptoms were retained for the combined model.

In the combined model, T1 and T4 child ADHD and ODD symptoms and mothers’ and fathers’ ADHD symptoms were regressed on family history of ADHD, and T4 child ADHD and ODD symptoms were regressed on T1 child ADHD and ODD symptoms, mothers’ and fathers’ ADHD symptoms, and the latent paternal comorbid psychopathology variable. All T1 variables in the model were allowed to covary. The two T4 variables were also allowed to covary. Covariances that were non-significant were dropped from the model. This model (Fig. 2) demonstrated good fit: χ2/df = 1.30, RMSEA = 0.03, CFI = 0.99, and SRMR = 0.04. In the combined model, mothers’ ADHD symptoms remained a significant predictor of T4 child ADHD symptoms, controlling for T1 child ADHD symptoms, family history of ADHD, and fathers’ comorbid psychopathology symptoms. Family history of ADHD and fathers’ psychopathology symptoms significantly predicted greater T4 child ADHD and ODD symptoms.

Fig. 2
figure 2

Combined path model examining unique relations between parental ADHD and later child ADHD and ODD symptoms, controlling for family history and parental psychopathology. Only significant paths and covariances are shown in figure. ADHD = attention - deficit/hyperactivity disorder; ODD = oppositional defiant disorder † p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001

Full size image

Moderators of the Relation between Parent and Child ADHD Symptoms

To explore whether family history, comorbid parental psychopathology, and family adversity each moderate the relation between parental ADHD symptoms and later child ADHD and ODD symptoms, an interaction variable was created in MPlus between each moderator variable and maternal and paternal ADHD symptoms, with predictor and moderator variables centered. T4 symptoms were then regressed on the moderator variable, maternal or paternal ADHD symptoms, and the interaction term, as well as on T1 child symptoms. Family history of ADHD symptoms and maternal and paternal comorbid psychopathology did not interact with maternal or paternal ADHD symptoms in predicting either child ADHD or ODD symptoms at T4 (all ps > 0.39). Family adversity significantly moderated the relation between maternal ADHD symptoms and T4 child ADHD symptoms, b = 0.03, SE = 0.01, p = 0.01, controlling for child T1 ADHD symptoms. This moderation was further explored to determine which aspect of family adversity was driving the moderation; family income and maternal LES interacted with maternal ADHD symptoms in predicting T4 child ADHD symptoms, β = 0.14, SE = 0.07, p = 0.05 and β = 0.22, SE = 0.11, p = 0.04, respectively. Simple slopes analyses of the relation between maternal ADHD symptoms at low (LES = +1 SD), moderate (LES = the mean), and high (LES = −1 SD) levels of family stress revealed that maternal ADHD symptoms significantly predicted T4 child ADHD symptoms when mothers reported moderate, β = 0.24, SE = 0.08, p = 0.002, and low levels, β = 0.36, SE = 0.11, p = 0.001, of stress, but not significantly when mothers reported high levels of stress, β = 0.13, SE = 0.08, p = 0.10. For family income, maternal ADHD symptoms only predicted later child ADHD symptoms for children from families with high, β = 0.39, SE = 0.14, p = 0.006, and moderate, β = 0.22, SE = 0.07, p = 0.003, family incomes, but not for children from families with low family incomes, β = 0.05, SE = 0.08, p = 0.58. Family adversity did not interact with maternal ADHD symptoms in predicting T4 child ODD symptoms or with paternal ADHD symptoms in predicting T4 child ADHD or ODD symptoms (ps > 0.49).

Parenting as a Mediator of the Relation between Parent and Child ADHD Symptoms

Five parenting latent variables were constructed using T2 and T3 measures: maternal overreactivity, laxness, and warmth; and paternal overreactivity and laxness. T4 child ADHD and ODD symptoms were regressed on these latent variables, as well as on mothers’ and fathers’ ADHD symptoms and T1 child ADHD and ODD symptoms. Parenting variables were regressed on mothers’ and fathers’ ADHD symptoms and T1 child ADHD and ODD symptoms. All T1 variables in the model were allowed to covary. Parenting latent variables were allowed to covary. The two T4 variables were also allowed to covary. Covariances that were non-significant were dropped from the model. The parenting model (Fig. 3) demonstrated good model fit: χ2/df = 1.21, RMSEA = 0.03, CFI = 0.98, and SRMR = 0.05. Each parenting variable loaded significantly on its respective latent variable; these loadings are not presented in the figure for ease of presentation.

Fig. 3
figure 3

Model for relations between parental ADHD symptoms and later child ADHD and ODD symptoms, with maternal and paternal parenting practices as mediators. Only significant paths and covariances are shown in the figure. Standardized coefficients are presented. † p < 0.10, *p < 0.05, **p < 0.01, *** p < 0.001

Full size image

Mothers’ ADHD symptoms significantly predicted greater maternal laxness and overreactivity. Mothers’ overreactivity, but not laxness, in turn predicted greater child T4 child ADHD and ODD symptoms. However, mothers’ ADHD symptoms did not predict warmth, β = −0.01, SE = 0.13, p = 0.92. Maternal warmth was also not significantly predictive of fewer T4 child ADHD or ODD symptoms, β = 0.01, SE = 0.13, p = 0.92, and β = −0.13, SE = 0.15, p = 0.40, respectively.Footnote 7 Mothers’ ADHD symptoms did not significantly predict T4 child ADHD or ODD controlling for parenting practices, β = 0.10, SE = 0.07, p = 0.14, and β = −0.09, SE = 0.08, p = 0.23, respectively. The indirect effect from maternal ADHD symptoms to T4 child ADHD symptoms via maternal overeactivity was significant, β = 0.07, SE = 0.03, p = 0.01, 95 % CI [0.02–0.18]. Similarly, the indirect effect from maternal ADHD symptoms to T4 child ODD symptoms via maternal overreactivity was also significant, β = 0.08, SE = 0.03, p = 0.01, 95 % CI [0.01–0.10]. Paternal ADHD did not significantly predict paternal parenting; however, T1 ODD symptoms significantly predicted greater paternal overreactivity and paternal laxness significantly predicted greater T4 ODD symptoms.

Discussion

This study sought to better understand the role that parental ADHD symptoms play in the early development of children’s ADHD and ODD symptoms and yielded several key findings. First, although both maternal and paternal ADHD symptoms were positively associated with children’s ADHD symptoms at age 3, only maternal ADHD symptoms predicted later ADHD symptoms in children, holding early child symptomatology constant. Second, both family history of ADHD and paternal comorbid psychopathology predicted later child ADHD and ODD symptoms, but they did not account for the association between maternal and child ADHD symptoms. Third, family adversity moderated the relation between maternal ADHD symptoms and later child ADHD symptoms. Finally, maternal overreactivity mediated the relation between maternal ADHD symptoms and later child ADHD and ODD symptoms. Taken together, these findings suggest that maternal ADHD symptoms may contribute to the development of ADHD and ODD symptoms in children through disruption in parenting, and that effects of maternal ADHD may be most pronounced in children with low family adversity. Moreover, fathers’ ADHD symptoms do not appear to directly contribute to the development of ADHD or ODD symptoms over the preschool years. However, other comorbid paternal psychopathology symptoms were found to play a role in the development of ADHD and ODD symptoms.

Relation between Parental ADHD and Child ADHD and ODD

Parental ADHD symptoms were correlated with age 3 ADHD symptoms for both mothers and fathers, though only maternal ADHD symptoms were associated with age 3 ODD symptoms. These findings converge with previous studies linking parental ADHD to child ADHD and ODD (e.g., Goos et al. 2007; LeMoine et al. 2015; Romirowsky and Chronis-Tuscano 2014; Segenreich et al. 2015; Takeda et al. 2010). However, unlike previous studies, our cross-sectional findings did not show substantial differences between mothers and fathers in the relation between parent and child ADHD symptoms, though consistent with some previous research (Goos et al. 2007; Segenreich et al. 2015), relations appeared to be somewhat stronger for mothers than for fathers. To our knowledge, this is the first study to examine whether parental ADHD predicts later child ADHD symptoms controlling for earlier child ADHD symptoms. Our longitudinal findings showed a clearer pattern of parent gender differences; only mothers’ ADHD symptoms predicted children’s later ADHD symptoms controlling for early child ADHD symptoms. This suggests that mothers’ and fathers’ ADHD symptoms may play a role in the early emergence of ADHD symptoms, but only mothers’ symptoms play a role in how these symptoms may shift over time. That is, both mothers’ and fathers’ ADHD symptoms play a role in determining which 3-year-old children have relatively higher levels of ADHD symptoms at age 3; however, within children with the same level of ADHD symptoms at age 3, mothers’ but not fathers’ ADHD symptoms predict which children will have higher versus lower ADHD symptoms at age 6. Additionally, our finding that maternal ADHD symptoms were concurrently related to child ODD symptoms, but did not predict later ODD symptoms is consistent with previous cross-sectional research (e.g., Ellis and Nigg 2009), and converges with the scant longitudinal research in this area (Brammer et al. 2016).

Our results also indicated that although family history of ADHD and paternal comorbid psychopathology predicted later child ADHD, they did not account for the relation between maternal ADHD and later child ADHD. However, the concurrent relation between paternal ADHD and early child ADHD was no longer significant when family history and paternal comorbid psychopathology were controlled. Our findings regarding paternal comorbid psychopathology are consistent with previous research that has found paternal depression (Ramchandani et al. 2005; Ramchandani et al. 2008) and antisocial behavior (Galéra et al. 2011) to be predictive of child ADHD, and with research finding paternal antisocial behavior, but not paternal ADHD to be associated with child conduct problems (LeMoine et al. 2015). Similarly, the observed relations between family history of ADHD and child ADHD are consistent with a number of studies of older children (e.g., Biederman et al. 2008). Our findings extend these bodies of research and suggest that both family history of ADHD and paternal psychopathology are not only associated with child ADHD symptoms, but predict later ADHD controlling for early ADHD. That is, among children with identical ADHD symptoms at age 3, family history and paternal comorbid psychopathology significantly predicted which children would exhibit more symptoms of ADHD at age 6.

In contrast with previous studies linking maternal psychopathology to child ADHD (e.g., Galéra et al. 2011; Segenreich et al. 2015), the present study did not find maternal comorbid psychopathology to be predictive of child ADHD or ODD symptoms at age 6, when taking into account maternal ADHD and early child ADHD and ODD symptoms. However, these previous studies did not take into account comorbid maternal ADHD, suggesting that the link between child ADHD and maternal depression (e.g., Tully et al. 2008) and anxiety (Segenreich et al. 2015) may actually be due to comorbid maternal ADHD symptoms. Indeed, when we estimated a model just with comorbid psychopathology without controlling for maternal ADHD, maternal psychopathology was predictive of later ADHD and ODD.

Moderating Role of Family Adversity in the Relation between Parent and Child ADHD

Results of the present study suggest that the effects of maternal ADHD on children vary depending on the larger family context, such that maternal ADHD is more strongly associated with development of child ADHD in families with less adversity. It may be that among families with high levels of family adversity, other variables play a larger role in the development of child ADHD, thereby reducing the role of maternal ADHD. These findings stand somewhat in contrast to research suggesting that positive family environments can buffer the negative effects of stressors more generally (see Grant et al. 2006), though a number of studies have also failed to find such moderation (Grant et al. 2006). Further research is needed to better understand how parental ADHD interacts with family adversity and other related factors in contributing to children’s development.

Role of Parenting in the Relation between Parental ADHD and Child ADHD and ODD

Our findings converge with studies (Chronis-Tuscano et al. 2011; Tung et al. 2014) suggesting that harsh parenting mediates the relation between parental ADHD and child ADHD and ODD symptoms. Moreover, the present study indicates that these processes emerge in the early development of ADHD. Thus, our results add to the growing body of literature suggesting that maternal ADHD is related to disruptions in some aspects of parenting (Chen and Johnston 2007; Chronis-Tuscano et al. 2008; Harvey et al. 2003), which in turn appear to exacerbate children’s ADHD symptoms. Although ADHD is a largely genetic disorder (Burt 2009; Burt et al. 2012), these findings suggest that at least in young children, symptoms can be influenced by environmental factors, as has been demonstrated in a number of studies finding parent training to decrease ADHD symptoms in young children (e.g., Sonuga-Barke et al. 2001).

Limitations

The results of the present study should be interpreted in the context of several limitations. First, only 6 % of fathers and 12 % of mothers self-reported clinically significant ADHD symptomatology. The low rates of clinically significant parental psychopathology in this sample limit the generalizability of our findings to parents with clinical diagnoses. Second, our measure of parental ADHD was based on self-report which may be limited given evidence that adults often underreport their own symptoms (Barkley et al. 2002). Given the lower levels of ADHD in fathers compared to mothers in this sample, it may be that fathers in this study underreported their own symptoms, which could explain why fathers’ ADHD symptoms failed to predict children’s later symptoms. Future studies should utilize collateral information when assessing parental ADHD. Third, family history of ADHD was measured using parents’ retrospective reports, which may not be fully accurate. Furthermore, family history likely reflects a combination of shared genetic, environmental and behavioral risk factors, and this study was not able to tease apart the effects of these factors. Future research directly assessing whether genetics account for the relation between parent and child ADHD would be important. Fourth, given that a multi-informant approach is the gold-standard for diagnosis of ADHD, there is some concern around the generalizability of the findings of this study given that parent-report of symptoms were solely used to measure both parent and child ADHD. Significant effects found in this study may be inflated by shared method variance or by systematic biases that parents with ADHD may have in rating their children’s behavior. This concern may be somewhat mitigated by the fact that this paper involved predicting later child symptoms controlling for early symptoms, which in effect controls for some of the shared informant variance. Nonetheless, future research is needed to replicate the findings using multi-informant assessment of both parent and child ADHD symptoms. Finally, the effects of paternal ADHD may depend on father involvement and this was unfortunately not addressed in the present study.

Future Directions and Implications

Despite these limitations, the present study has important theoretical and clinical implications. Specifically, targeting comorbid paternal psychopathology and maternal parenting practices has potential for attenuating the effects of parental ADHD on children’s ADHD symptoms. Several studies have begun to explore these possibilities. Chronis-Tuscano et al. (2011) found that the relation between maternal ADHD and improvement in child behavior following parent training was mediated by improvements in negative parenting. Babinski et al. (2014) studied the benefits of behavioral parent training and stimulant medication for parents with ADHD on their parenting and ADHD symptoms, and found parent training (and use of stimulants) to be effective in reducing inconsistent discipline and in improving child behavior. Additionally, several behavioral parent training intervention studies have examined the benefits of targeting parental depressive symptoms (Chronis et al. 2006; Chronis-Tuscano et al. 2013) or a combination of stress, depression, anxiety, and other parenting problems (Sanders et al. 2000) as an enhancement to traditional interventions for families of children with ADHD. However, these interventions have focused more on maternal psychopathology symptoms than paternal. The findings of our study suggest that paternal psychopathology symptoms are linked to both child ADHD and ODD symptoms; thus, more research is needed to examine the benefit of targeting these psychopathology symptoms in fathers with ADHD. In conclusion, the role of maternal ADHD and parenting practices and paternal psychopathology symptoms is important for conceptualizing ADHD development across the preschool years and are potential domains to target in treatment.