The Association between ADHD Symptoms and Antisocial Behavior: Differential Effects of Maternal and Paternal Parenting Behaviors

Abstract

The present study examined the moderating role of maternal and paternal behaviors on the association between ADHD symptoms and antisocial behavior. The data are from a Swedish community-based study with 2886 adolescent participants, aged 13–15. We analyzed the adolescents’ self-reports of negative and positive parental behavior, ADHD symptoms, and antisocial behavior. Results show that ADHD symptoms significantly predicted levels of antisocial behavior in both maternal- and paternal-moderated regression models. Also, several significant interaction effects emerged. Fathers’ attempted understanding and mothers’ warmth significantly buffered the association with antisocial behavior. In turn, mothers’ negative behaviors—both coldness and rejection and angry outbursts—significantly enhanced the association with antisocial behavior, whereas only fathers’ angry outbursts emerged as a significant moderator. Thus, there is support for differential effects of maternal and paternal behavior on the association between ADHD symptoms and adolescent antisocial behavior.

1. Introduction

Attention-deficit/hyperactivity disorder (ADHD) is a syndrome of inattention, hyperactivity, and impulsivity that affects both cognitive and behavioral functioning in academic, social, and family contexts [1]. It is one of the most prevalent psychiatric disorders among children and adolescents, affecting between 5% and 10% of youths in general populations worldwide [2,3]). The central deficit of the disorder is low behavioral inhibition, which is thought to interfere with the child’s ability to regulate and organize their behavior in different contexts, most prominently in school and at home. Low behavioral inhibition is also linked to antisocial behaviors in adolescence, but the mechanisms explaining the link between ADHD and antisocial behavior is unclear. One potential explanation might lie within the family context, as it constitutes an important context for the child’s development of behavioral inhibition. Specifically, how parents interact with their adolescents might be of importance to explain why ADHD might be linked to antisocial behaviors for some adolescents and not for others. In this study, we disentangle the effects of mothers’ and fathers’ positive and negative behaviors and examine them as potential moderators of the association between ADHD symptoms and antisocial behaviors.

It is well established that ADHD symptoms are related to a range of negative behaviors. These include externalizing problems, anxiety and mood disorders, poor academic performance, as well as antisocial behavior and delinquency in adolescence [4]. Research shows that adolescents with symptoms of inattentiveness and restlessness are likely to be arrested, convicted, and reoffend at a higher rate than youths without these symptoms [5]. Prevalence rates of offending for youths with ADHD vary between 30% and 50% in adolescence and 25% and 46% in adulthood [6]. In addition, children and youths with ADHD are also at elevated risk for criminal behaviors in adulthood [4,7,8], as well as other disruptive disorders, such as conduct disorder and oppositional defiant disorder, in adolescence [8]. Furthermore, results from a longitudinal study showed that children who displayed symptoms of ADHD were more likely to be involved in and convicted of crime as adults [9]. Thus, symptoms such as inattentiveness, hyperactivity, and impulsivity (i.e., ADHD) contribute to the emergence and continuation of problem behaviors throughout early development and into adulthood.

Theoretically, the presence of ADHD symptoms is thought to influence the development of antisocial behavior in two ways. Given that the central deficit of the disorder is low behavioral inhibition, one way that symptoms of ADHD may increase the risk of antisocial behaviors is through impulsive and irresponsible behavior. These behaviors are thought to interfere with youths’ ability to regulate and structure their behavior in line with the demands of the social context. Indeed, low self-control is one of the most studied predictors of crime and delinquency across the lifespan, and there is ample evidence of associations with learning difficulties in school, impaired peer relationships, and various conflicts and disturbances within the family among adolescents with and without ADHD [10]. Another way that symptoms associated with ADHD may increase the risk of antisocial behavior is through genetic vulnerability. For instance, studies show that the heritability of ADHD is strong and accounts for 71–73% of the variation in inattention and hyperactivity symptoms [11]. Given this large genetic overlap between parents and children in hyperactivity, impulsive behavior, and inattention, it is reasonable to assume that parents of youths with ADHD symptoms may be more likely to use ineffective parenting practices that may further increase the risk for poor adjustment and externalizing problems in adolescents. Results from twin studies offer some support for this assumption, showing that a highly heritable general liability to externalizing problems accounted for the parent–child resemblance and was manifested in behavior problems such as oppositional defiant disorder and conduct problems [12]. In summary, characteristics that designate ADHD are highly heritable and place children and youths at an increased risk for the development of antisocial behavior later in life.

Although ADHD has been tied to problematic behaviors and delinquency, less is known about the associated family factors. Clearly, many adolescents with ADHD manage the difficulties associated with the disorder and function well in school and at home, which emphasizes the importance of understanding the factors that attenuate and exacerbate the link to negative outcomes. From a developmental perspective, parents’ behaviors can and do influence youth outcomes. The relation between ineffective and neglectful parenting and youths’ externalizing problems is well documented and is typically viewed as evidence that poor parenting is a salient predictor of the development of antisocial behavior throughout adolescence [13,14,15]. In contrast, positive parenting behaviors, such as parental warmth, is typically associated with decreasing problem behaviors across adolescence [14,16]. Research has also recognized that difficult and hard-to-manage children are particularly at risk of eliciting poor parenting behavior and that these re-occurring and cyclical exchanges increase the risk for the development of antisocial behavior. In this study, we examine different parenting behaviors as moderators to explore what aspects of the parent–child relationship might be of particular importance to explain the link between ADHD and antisocial behaviors.

Parenting behavior seems to play a role in the development of antisocial behavior in adolescents with ADHD [17]. In particular, negative parenting behavior seems to account for the link between ADHD and delinquency behavior [18]. Furthermore, Walther et al. [19] examined the relations between parenting behaviors and adolescent-reported substance use and delinquency among adolescents with and without a diagnosis of ADHD. Adolescents’ report of parental knowledge was a significant correlate of delinquency and substance use above and beyond other parenting variables. More knowledge was associated with less delinquency and substance use for all participants, but parental knowledge was more strongly associated with alcohol use for adolescents with ADHD.

Research on the effects of maternal and paternal parenting on adolescent problem behavior is scarce. Typically, mothers’ and fathers’ behaviors are averaged into a composite parenting score not considering their unique contributions. The handful of studies that have parsed the effects of maternal and paternal behavior on adolescent outcomes show that there are unique effects depending on the parent’s gender [15] and that the effects vary in relative importance in predicting adolescent problem behavior [20]. For instance, the longitudinal study by Hoeve at al. [15] show that fathers’ neglectful parenting style, but not mothers’, influenced children’s delinquency across five years. In contrast, other results suggest that negative parenting, measured as psychological control, positively predicts externalizing behavior problems (i.e., aggression and role-breaking) but only when exerted by mothers [21]. In the same study, no significant effects of paternal negative behavior on adolescent externalizing behavior emerged. The research by Yang et al. [20] shows that maternal hostility is relatively more important than paternal hostility in predicting subsequent aggression over three years of data. Interestingly, paternal warmth was more important than maternal warmth, and there seem to be different effects for boys and girls in the study. The authors suggest that maternal hostility may be in conflict with expected maternal care-taking behaviors and therefore may be the relatively stronger influence on adolescent aggression behaviors. Thus, there is evidence to suggest that mothers’ and fathers’ specific parenting behaviors may have different links with adolescent problem behaviors.

Given that ADHD symptoms are a risk factor for antisocial and criminal behavior, it is important to further explore the potential role of parenting in this development. In this study, we test the link between ADHD symptoms and antisocial behavior among mid-adolescents. We move beyond the study of overall parenting behavior, comprising mothers’ and fathers’ responses averaged into a composite, and examine how maternal and paternal parenting behaviors impact the link between ADHD symptoms and antisocial behavior. Specifically, we compare mothers’ and fathers’ positive (i.e., warmth, attempted understanding) and negative (coldness and rejection, angry outbursts) behaviors and how they predict antisocial behavior for adolescents with ADHD symptoms.

2. Materials and Methods

2.1. Participants and Procedure

Participants were 13- to 16-year-old youths and their parents in a mid-sized city in Sweden. Students in all 7th through 9th grade classes in the city (N = 2886) composed the target sample for the study. They took part in the study unless their parents returned a postage-paid postcard stating that they did not want their child to participate (less than 1% did so). In total, 1007 boys (51.9%) and 934 girls participated; M_age was 13.79, SD = 2.60.

Adolescents filled out the questionnaires during regular school hours in sessions administered by trained research assistants. They were given information about the study and were informed that their participation was voluntary. Neither teachers nor school administrators were in the room during the time the adolescents competed the questionnaires. Parents were sent questionnaires to their home addresses with a postage-paid return envelope. Neither adolescents nor parents were paid for their participation. The study, including all measures and procedures, was approved by the university’s Ethics Review Board (CF-529-2003).

2.2. Measures

2.2.1. ADHD Symptoms

To measure ADHD symptoms, we used the Swanson, Nolan, and Pelham rating scale (SNAP-IV) [22]. SNAP-IV is a parent-reported scale and was developed to assess ADHD symptoms (i.e., hyperactivity/impulsivity and attention problems) according to the criteria in the DSM-IV and can be used in community-based populations [23]. Parents reported how well 19 statements applied to their adolescent’s behavior. Response options ranged from does not apply at all (1) to applies exactly (4). Hyperactivity was measured with five statements, and examples of these are, “Often leaves his/her place and has difficulties sitting still in situations that demand sitting still” and “Often has difficulties with keeping his/her hands or feet still or can’t sit still”. Impulsivity was measured with three statements, for example, “Often has difficulties waiting for his/her turn” and “Often answers before the question has been asked”. Attention problems was measured with nine statements, and examples of these are, “Is often distracted by things that happen around him/her”, “Often has difficulties with maintaining attention during a longer period”, and “Often has difficulties to follow instructions and therefore often fails to complete tasks”. The total score of all 19 items (Cronbach’s α = 0.92) was used to represent the construct.

2.2.2. Antisocial Behavior

Adolescent-reported antisocial behaviors were measured with 19 items similar to other commonly used, validated self-report measures [24] that have been validated in a Swedish sample in a comparison with official records [25]. Items referred to delinquent behaviors during the past year, ranging from acts of vandalism to shoplifting or breaking into a building. Items were rated on a 5-point scale ranging from no, it has never happened (0) to more than 10 times (4), and a sum score based on all items was used. The Cronbach’s alpha was 0.92.

2.2.3. Parents’ Positive Behaviors

Attempted Understanding. Parents’ attempted understanding was measured with five items [26], and adolescents reported on mothers and fathers separately. The stem question was, “What happens if you have done something your parent really dislikes?” Adolescents rated five statements about their parents’ typical behaviors and reported on mothers’ and fathers’ behaviors separately. Examples of the five items are, “Talks to you at once”, “Are clear about what they think, but are open for discussions”, “Honestly wants to understand why you did what you did”. The response options were “never”, “sometimes”, and “most often”. The alpha reliability for this scale was 0.73 for mothers and 0.78 for fathers.

Warmth. Parental warmth was measured with six statements, and adolescents reported on mothers’ and fathers’ behaviors separately [26]. There were three response options, ranging from “never” to “most often”. Some examples of parental warmth statements are, “Praises you for no special reason”, “Shows they care for you with words and gestures”, “Does small things that make you feel special (e.g., wink, smile)”. The alpha reliability for this scale was 0.81 for mothers and 0.85 for fathers.

2.2.4. Parents’ Negative Behaviors

Angry Outbursts and Coldness/Rejection. To measure two aspects of negative parenting practices, angry outbursts and coldness/rejection, we used adolescents’ responses to 11 statements about how their parents typically responded to behaviors they did not like. The stem question for all these items was, “What happens if you do something your parent really dislikes?” The response options for both scales ranged from “never” to “most often”. Five items assessed angry outbursts [26]. Some examples are, “Becomes very angry and has an outburst”, “Has a hard time controlling his or her irritation”, “Screams and yells at you”. The alpha reliability for this scale was 0.90 for both mothers and fathers. Parents’ coldness/rejection was assessed with six items [27]. Some examples of the items are, “Is silent and cold towards you”, “Makes you feel guilty for a long time”, “Avoids you”. The alpha reliability for this scale was 0.80 for mothers and 0.82 for fathers.

2.3. Analyses

Moderation analysis was mainly used to examine the interaction effects of parenting variables. Correlational analyses were first conducted to assess the bivariate intercorrelations among study variables. Four separate models were performed with the interaction term. We calculated the same model set on maternal and paternal variables. Model results were summarized using the R-package “stargazer”, version 5.2.3 [28]. Detected significant interactions were further visualized with the R-package “interactions” version 1.2.0 [29]. In our analyses, we deemed p-values less than 0.05 as the statistical significance level. Listwise deletion was used, and sample size for each regression model is noted in the table. All analyses were conducted in R [30].

3. Results

3.1. Descriptive Statistics

Means and standard deviations, as well as correlations for all study variables, are reported in

Table 1. Means, standard deviations, and correlations of study variables.

Variable	M	SD	1	2	3	4	5	6	7	8	9
1. Antisocial behavior	1.15	0.33
2. ADHD	1.83	0.52	0.23 **
3. Attempted understanding (m)	2.42	0.44	−0.11 **	−0.15 **
4. Attempted understanding (f)	2.32	0.50	−0.14 **	−0.15 **	0.66 **
5. Warmth (m)	2.46	0.44	−0.13 **	−0.13 **	0.50 **	0.34 **
6. Warmth (f)	2.34	0.50	−0.13 **	−0.07 **	0.31 **	0.49 **	0.64 **
7. Angry outbursts (m)	1.63	0.57	0.17 **	0.19 **	−0.15 **	−0.10 **	−0.28 **	−0.20 **
8. Angry outbursts (f)	1.62	0.58	0.12 **	0.17 **	−0.06 *	−0.11 **	−0.13 **	−0.28 **	0.47 **
9. Coldness and rejection (m)	1.35	0.45	0.15 **	0.17 **	−0.23 **	−0.19 **	−0.27 **	−0.15 **	0.59 **	0.36 **
10. Coldness and rejection (f)	1.38	0.48	0.16 **	0.12 **	−0.17 **	−0.19 **	−0.18 **	−0.28 **	0.38 **	0.62 **	0.60 **

Note. In the brackets, m = mothers, f = fathers. * p < 0.05. ** p < 0.01.

. Reports of ADHD symptoms (M = 1.83, response options ranging from 1 to 4) and antisocial behaviors (M = 1.15, response options ranging from 0 to 4) were on the lower side, which is according to expectations. Further, means for parenting practices were generally higher for adolescents’ reports of positive parenting practices (warmth and attempted understanding, ranging from 2.32 to 2.46) than for their reports of negative parenting practices (angry outbursts and coldness and rejection, ranging from 1.35 to 1.63). Hence, adolescents perceived their parents to use more positive than negative parenting practices.

Regarding correlations, antisocial behavior showed significant bivariate correlations with most study variables, with the strongest relationships with ADHD symptoms (r = 0.23). Furthermore, the ADHD symptoms and all parenting behavior variables had significant intercorrelations. Specifically, adolescents with more ADHD symptoms perceived that their mothers and fathers showed less attempted understanding and warmth and more angry outbursts and cold behaviors. Strengths of the correlations ranged from −0.07 for fathers’ warmth to 0.19 for mothers’ angry outbursts.

3.2. Mothers’ Positive and Negative Parenting Behaviors as Moderators of the Association between ADHD Symptoms and Antisocial Behavior

Model results from maternal and paternal data are summarized in

Table 2. Regression model results: maternal models.

	Dependent Variable:
	Antisocial Behaviors
	(1)	(2)	(3)	(4)
ADHD symptoms	0.29 **	0.35 **	−0.07	−0.10
	(0.11)	(0.11)	(0.06)	(0.07)
Attempted understanding	0.05
	(0.08)
Attempted understanding × ADHD symptoms	−0.06
	(0.04)
Warmth		0.11
		(0.08)
Warmth × ADHD symptoms		−0.09 *
		(0.04)
Angry outbursts			−0.12
			(0.07)
Angry outbursts × ADHD symptoms			0.11 **
			(0.04)
Coldness and rejection				−0.22 **
				(0.09)
Coldness and rejection × ADHD symptoms				0.17 **
				(0.05)
Observations	768	760	777	780
R2	0.07	0.07	0.09	0.08
Adjusted R2	0.06	0.06	0.08	0.08

Note. * p < 0.05. ** p < 0.01.

and

Table 3. Regression model results: paternal models.

	Dependent Variable:
	Antisocial Behaviors
	(1)	(2)	(3)	(4)
ADHD symptoms	0.41 **	0.31 **	0.06	−0.03
	(0.10)	(0.10)	(0.06)	(0.07)
Attempted understanding	0.15
	(0.08)
Attempted understanding × ADHD symptoms	−0.12 **
	(0.04)
Warmth		0.08
		(0.08)
Warmth × ADHD symptoms		−0.08
		(0.04)
Angry outbursts			−0.04
			(0.07)
Angry outbursts × ADHD symptoms			0.05
			(0.04)
Coldness and rejection				−0.12
				(0.08)
Coldness and rejection × ADHD symptoms				0.11 *
				(0.04)
Observations	748	738	751	753
R2	0.08	0.07	0.07	0.08
Adjusted R2	0.08	0.06	0.07	0.08

Note. * p < 0.05. ** p < 0.01.

, respectively. The tables include four models, in which we used each of the four parenting behaviors as predictors and in the interaction with ADHD symptoms.

In the models of positive maternal behaviors (i.e., warmth, attempted understanding), ADHD symptoms were a significant predictor of the level of antisocial behavior when controlling for adolescents’ reports of these parenting practices. None of the main effects of positive maternal behaviors were significant, but the interaction between adolescents’ reports of maternal warmth and ADHD symptoms was a significant predictor of antisocial behaviors (β  =  −0.09, p  =  0.035), indicating that maternal warmth—but not attempted understanding—attenuated the association between ADHD symptoms and adolescent antisocial behavior.

In the models with negative maternal behaviors, both angry outbursts and coldness and rejection directly predicted antisocial behavior, but ADHD symptoms did not. Both interaction terms were significant, including angry outbursts (β  =  0.11, p  =  0.001) and coldness and rejection (β  =  0.17, p  <  0.001). As shown in Figure 1, mothers who displayed higher levels of negative parenting elevated the association between ADHD symptoms and antisocial behavior.

3.3. Fathers’ Positive and Negative Parenting Behaviors as Moderators of the Link between ADHD Symptoms and Antisocial Behavior

Some patterns in the maternal models were also found in the paternal models. For example, ADHD symptoms significantly predicted antisocial behavior in the models with positive parenting behaviors but not negative parenting behaviors. Moreover, some effects were specific to adolescents’ reports of their fathers. Only one near-significant main effect was found for adolescents’ reports of fathers’ behaviors, namely, attempted understanding (β = −0.15, p = 0.065). Adolescents who reported more paternal attempted understanding tended to report fewer antisocial behaviors. The interaction effect between ADHD and fathers’ attempted understanding was significant, which suggests that adolescents who perceived their fathers as understanding (β  =  −0.12, p  =  0.004) tended to engage less in antisocial behavior. Fathers’ warmth was not a significant moderator (β  =  −0.08, p  =  0.072). Considering fathers’ negative behaviors, one interaction term was significant. The results shown in Figure 2 suggest that fathers’ coldness and rejection moderated the link between ADHD symptoms and antisocial behavior (β  =  0.11, p  =  0.011), enhancing the severity of adolescent antisocial behavior. No significant interaction effect of fathers’ angry outbursts was observed (β  =  0.05, p  =  0.206).

4. Discussion

In this study, we examined the association between ADHD symptoms and antisocial behaviors in adolescents, using adolescents’ reports of their parents’ positive and negative behaviors as moderators. The results showed similar patterns across mothers and fathers. ADHD was a significant predictor of antisocial behaviors, but only in the positive parenting models. Adolescents’ reports of positive parenting behaviors had an interactive effect on antisocial behaviors, as they tended to buffer the negative effect of ADHD. When considering negative parenting practices, adolescents’ reports of such behaviors tended to be more important for their antisocial behaviors than were their ADHD symptoms. Negative maternal behaviors had both a significant main effect on antisocial behavior and a significant interactive effect in combination with ADHD. Reports of paternal negative behaviors, however, tended to be weaker predictors of antisocial behaviors. No significant main effects of fathers’ negative behaviors were found, and only the interaction including coldness/rejection was significant.

Our results, which suggest that ADHD symptoms seem less important for antisocial behaviors than negative parenting practices, are not surprising. Previous research has consistently demonstrated a robust link between negative parenting practices and antisocial behaviors in adolescents. For example, parental hostility, anger, and control are shown to significantly predict increases in various types of problem behaviors, including aggression and delinquency, over time [14,16,31]. In addition, research has also shown that negative parenting practices affect youths’ self-control [32], which suggests that adolescents scoring higher on ADHD symptoms and who are exposed to more negative parenting behaviors may be at increased risk for antisocial behavior. Clearly, the links between ADHD symptoms, negative parenting practices, and adolescent antisocial behavior need more attention.

Comparing mothers and fathers in how their parenting practices are linked to adolescent antisocial behavior revealed a few differences. First, our result suggests that negative maternal behaviors seem to be more influential (all four tested effects were significant) than negative paternal behaviors (one out of four tested effects was significant) in moderating the link between ADHD symptoms and adolescent antisocial behavior. Second, positive paternal behaviors (two out of four tested effects were significant) seemed somewhat more important than positive maternal behaviors (one out of four tested effects was significant) in buffering the association. This may imply differential effects for maternal and paternal parenting practices and antisocial behavior in adolescence and is in line with some previous research suggesting that mothers are usually more involved in child-rearing than fathers, both in terms of time spent at home with their children and the number of responsibilities in child upbringing [33]. Because of the imbalance in parental involvement, children are thought to be overall more sensitive to mothers’ negative practices than the same paternal behaviors. This would be a speculative explanation for our findings showing that fathers’ positive behaviors seem to buffer the negative association with antisocial behavior. Longitudinal studies are needed to show whether maternal and paternal behaviors contribute differently to antisocial behavior during adolescence.

From a clinical perspective, our results show that negative parenting practices—irrespective of parents’ gender—enhance the positive link between ADHD symptoms and adolescent antisocial behavior. Positive paternal behavior, expressed as warmth and understanding, seems somewhat more important than positive maternal behavior in buffering this link. Therefore, one implication of these findings is that fathers represent a meaningful target for intervention among youths at risk for antisocial behavior and should be effectively involved in treatment programs next to mothers, who tend to be overrepresented in parenting intervention programs.

The results of this study should be interpreted in light of a few limitations. First, we mostly used adolescents’ self-reports of their own and their parents’ behavior. This presents potential issues with one-reporter bias, as adolescents’ reports on one measure might influence their reports on the next measure. It can also inflate the correlations between variables, as these are measured by the same reporter. Additionally, as these are adolescents’ reports of their parents’ behaviors, they do not necessarily represent actual observable behaviors. Still, the use of adolescents’ reports, rather than parents’ reports of parenting behaviors, might be more closely linked to adolescents’ behaviors, as it is likely that adolescents’ perceptions of their parents’ behaviors are what they react to. Second, the measure for assessing ADHD symptoms was not a clinical assessment, although it should be noted that the SNAP-IV has been shown to be a robust indicator of ADHD in normative adolescent samples [22]. Hence, these represent ADHD symptoms and not necessarily clinical levels. It is likely, however, that some adolescents in our sample either had levels that would result in a diagnosis of ADHD or already had a diagnosis. We did not have this information for the current study. Third, as already acknowledged, the cross-sectional design makes it impossible to examine temporal ordering between ADHD symptoms and antisocial behavior and whether parental behavior is a response to youth problem behaviors. Previous research has shown that ADHD and antisocial behavior are bidirectionally associated, and that parents’ negative behaviors might be both a response to and a predictor of youth problem behavior [34,35]. Future studies using longitudinal data may clarify the temporal ordering between ADHD symptoms, antisocial behavior, and parenting.

Some strengths deserve mention. First, given that much of what we know about the relationship between ADHD and problem behaviors is based on studies relying on youths in clinical or referred contexts, examining the association and potential moderators of it among youths in community settings is important. This may have important implications and could inform prevention and which parenting behaviors to focus on. Second, in this study, we make use of a large sample, which increases the power to detect even smaller effects. These effects might have theoretical and clinical implications but might have been undetected with a smaller sample. Third, as adolescents’ perceptions and relationships with their mothers and fathers might have different impacts on their development, it is a strength that we separate these measures in this study. In fact, although some patterns were similar across adolescents’ reports on fathers and mothers, some differences emerged, highlighting the importance of not treating parental behavior as a single entity.

In conclusion, our results show that there are small, yet significant, differences in the effects of mothers’ and fathers’ behaviors on adolescent problematic behaviors. This speaks to the importance of considering the unique contributions of mothers’ and fathers’ behaviors.