Whether or not children with ADHD develop
serious conduct problems - e.g., lying, fighting, bullying, and stealing
- is perhaps the single most important determinant of negative outcomes such
as delinquency, substance use, risky sexual behavior, and serious driving
accidents during adolescence and young adulthood. For this reason,
understanding the development of conduct problems in children with ADHD is
extremely important, and could contribute to the development of interventions
that prevent their emergence.
Although substantial evidence indicates that the onset of ADHD is largely
determined by genetic factors, and that ADHD is not "caused" by parenting,
parent mental health and parenting behavior do influence the development
of significant conduct problems in children. Thus, prior research has
documented that depression in mothers, antisocial behavior in fathers, and
the quality of parent-child interactions are linked to the development and
escalation of conduct problems in children.
These same factors have been hypothesized to influence the emergence and
course of conduct problems in children with ADHD, but this has never been
studied empirically. Recently, however, a study published in Developmental Psychology provided the
first set of data on this important issue [Chronis et al. (2007). Maternal
depression and early positive parenting predict future conduct problems in
young children with Attention-Deficit/Hyperactivity Disorder. Developmental Psychology, 43, 70-82.]
Participants in this study were 108 children diagnosed with ADHD between
the ages of 4 and 7 and the mothers of these children. Approximately
70% of the children were Caucasian and just over 80% were males. Like many
studies examining children's development, fathers were not directly involved
in the study, which may reflect the difficulty recruiting fathers to participate.
During the initial assessment, and during each assessment conducted annually
over the next 8 years, a structured diagnostic interview was administered
to children's mothers to obtain information about the child's functioning
during the prior year. In this interview, particular attention was
given to the presence of Conduct Disorder (CD) symptoms during the prior
year. Information about CD symptoms was also obtained from the majority
of children's teachers at each assessment.
The essential feature of Conduct Disorder is "...a repetitive and persistent
pattern of behavior in which the basic rights of others or age appropriate
social norms or rules are violated." These behaviors fall into 4 main groupings:
1. Aggressive behavior that causes or threatens
to cause harm;
Examples: initiating fights; cruelty to people or animals;
2. Non-aggressive conduct that causes property
loss or damage;
Examples: fire setting with intent to cause damage; deliberate destruction
3. Deceitfulness or theft;
Examples: shoplifting; breaking into someone's house; frequent lying to obtain
goods or avoid obligations;
4. Serious violation of rules;
Examples: truancy from school; running away from home; staying out at night
prior to age 13;
As noted above, children with ADHD who develop symptoms of Conduct Disorder
are at substantially greater risk for a range of negative outcomes during
adolescence and young adulthood.
In addition to establishing the ADHD diagnosis at year and obtaining baseline
data on the number of CD symptoms, psychiatric interviews were also conducted
with mothers during the initial assessment so that the presence of depression
during mothers' lifetime as well as other psychiatric disorders in mothers
could be determined. Mothers and their child also completed in both structured
and unstructured activities so that the quality of mother-child interactions
in different contexts could be observed.
The mother-child interaction task began with 10 minutes of free play activity
in a room stocked with age appropriate toys and activities. During
a subsequent 15-minute structured task, the mother was instructed to have
her child: 1) help with cleaning up the toys and other materials in the playroom;
2) complete a counting activity with blocks; 3) help dust the tables and
chairs; and, 4) play quietly while the mother took a 1-minute telephone call
from the experimenter. It was anticipated that these tasks would place
greater strain on the mother-child interaction because they required the
child to follow a series of directives and commands from the mother.
These interactions were videotaped and later coded so that the nature of
mother-child interaction could be quantified. The parenting behaviors
of particular interest were the amount of positive parenting displayed (praise,
positive affect, and physical affection) and the amount of negative parenting
displayed (negative commands, critical statements, and any type of
mild physical discipline). In addition, the researchers coded the amount
deviant behavior (whining, crying, yelling, refusing to comply with commands)
that children displayed.
Although approximately 20% of children were on medication at the initial
assessment, all children had been off their meds for 1-2 days prior to the
interaction task. During subsequent annual assessments, when upwards
of 50% of children were on meds, mothers and teachers were asked to rate
children's behavior based on periods when the child was not medicated.
- Results -
The question of primary interest to the researchers was whether maternal
depression and positive parenting were related to the emergence of conduct
disorder symptoms in young children with ADHD.
To examine this question, the researchers conducted analyses in which they
controlled for a number of other factors that might be related to the development
of CD symptoms including gender, race, family income, number of ADHD and
CD symptoms at baseline, child deviant behavior observed at baseline, and
parent and teacher ratings of child impairment at baseline. Because
these variables were controlled for when predicting the emergence of CD symptoms,
any effects found for maternal depression and mother-child interaction represent
the contribution of these factors that are above and beyond these other important
Not surprisingly the strongest predictor of CD symptoms during waves 2-8
was the number of CD symptoms present at wave 1. Interestingly, the
number of ADHD symptoms present at wave 1 was not a significant predictor
of CD symptoms during subsequent waves.
As predicted, maternal depression at wave 1 was a significant predictor of
CD symptoms at subsequent waves. Although children whose mothers were
depressed at wave 1, or who had a prior episode of depression, did not show
more CD symptoms at the initial assessment, they had higher rates of CD symptoms
at each subsequent wave. On average, it looked to be between .5 and
1 additional symptom at each wave. Although this may not seem like
a large difference, it was both statistically significant and present at
ALL 7 waves. In addition, because the symptoms of conduct disorder
are so problematic (e.g., often bullies, threatens, or intimidates others,
often initiates physical fights, fire setting, truancy) even a single additional
symptom is quite meaningful.
Results for the mother-child interaction task indicated that mothers high
in positive parenting during the structured interaction task had children
who developed fewer CD symptoms. Compared to mothers with positive
parenting scores in the bottom 25%, those with scores in the top 25% had
children who displayed about 1 fewer CD symptom at each assessment point.
Results for children whose mothers were in the middle 50% on positive parenting
were more mixed, but they showed fewer CD symptoms than those in the bottom
25% and more than those in the top 25%.
Contrary to expectations, the amount of negative parenting behavior did not
predict subsequent CD symptoms in children.
and Implications -
Although ADHD is not caused by 'poor' parenting, results from this study
highlight that maternal characteristics influence the development of CD symptoms
in children with ADHD. Because the development of significant conduct
problems in children with ADHD is perhaps the single most powerful predictor
of negative outcomes during adolescence and young adulthood, these findings
may have important implications for prevention efforts.
As reported above, even after controlling for other factors that would be
expected to predict the development of CD symptoms in children with ADHD
- most importantly, the number of CD symptoms at baseline - mothers with
current or prior depression had children who developed more CD symptoms over
time. And, mothers who displayed high rates of positive parenting behavior
during tasks requiring them to obtain their child's compliance, had children
who developed fewer CD symptoms. These results were consistently present
at all data collection points, and each accounted for an average difference
of roughly .5-1 CD symptom at each wave. Thus, a child whose mother
had a history of depression and showed low levels of positive parenting would,
on average, show between 1 and 2 additional CD symptoms at subsequent time
points. Given the problems represented by even a single CD symptom,
these are important findings.
As with most all studies, there are important limitations to this work that
the authors acknowledge. First, reports of CD symptoms were based solely
on maternal and teacher report, and no information on conduct problems was
obtained directly from children. As children grew older, and some begin
engaging in covert antisocial behavior that parents and teachers may be unaware
of, relying on adults only to assess CD symptoms is problematic.
The authors also note that they did not gather data on the timing of prior
maternal depressive episodes. For example, they did not determine whether
the depressive episode occurred during the child's lifetime, at what age
or developmental stage the child was exposed, or the duration of the exposure.
While these are an important omissions, it would most likely serve to attenuate
the relationship between maternal depression and the development of CD symptoms
in children, rather than artificially inflate it.
Another limitation is that the authors did not control for medication treatment
that some children received, so it is not possible to draw any conclusions
about the impact of such treatment. As noted above, about half the
children were receiving medication treatment at each wave after the first
one, and it would be important to whether this predicted the emergence of
CD symptoms and whether medication treatment interacted with the other predictor
Finally, the absence of fathers is an important limitation in this work.
The authors did collect information on psychiatric symptoms in fathers, but
this was gathered from mothers rather than from fathers directly. And,
there was no observational data gathered on father-child interactions.
Making sure that fathers are represented in studies such as this is an important
challenge for the entire field.
In conclusion, results from this study highlight that maternal depression
is a risk factor and positive parenting is a protective factor in the development
of children with ADHD. The authors certainly do not intend for these
results to be cited as a reason to 'blame' mothers for the development of
conduct problems in their child with ADHD. Rather, they suggest that
their results are a source of hope in that they suggest clear targets for
early intervention in families with a child who has ADHD - treating depression
when present in mothers and helping mothers develop their positive parenting
skills. Based on their findings, they hypothesize that such efforts
may help prevent the development or progression of later conduct problems.
It will be both interesting and important to test this hypothesis in new
presented in Attention Research Update is for informational purposes only,
and is not a substitute for professional medical advice.