|
A recent
issue of Child Development contains a very interesting study on the
long-term outcomes for children who are persistently
hyperactive, persistently oppositional, or persistently aggressive (Nagin,
D., & Tremblay, R.E. (1999). Trajectories of boys' physical aggression,
opposition, and hyperactivity on the path to physically violent and nonviolent
juvenile delinquency (1999). Child Development, 70, 1181-1196. This study
highlights the importance of considering separately hyperactivity,
oppositional behavior, and actual aggression - three different aspects of
children's behavior that often are incorrectly lumped together as being
reflective of ADHD.
This is a complex study in which some very arcane statistical techniques are
employed. What follows is my best effort to present these important data in a
simple and straight-forward way.
Here's what the authors did. At the start of the study, behavior ratings on
almost 1200 boys were obtained from their kindergarten teachers at the end of
the school year. These ratings were used to classify boys as being high,
moderate, or low on three different types of behavior: hyperactivity (i.e.
symptoms of ADHD such as being fidgety and unable to be still),
oppositionality (e.g. irritable, disobedient, refuses to share), and
aggressive (e.g. bullies others, fights with others, kicks or hits others).
Boys were considered to be high, moderate, or low on these 3 types of problem
behaviors based on how their score on each compared to the overall group
average.
Several years later the authors tracked down over 1000 of these boys, and had
their current teachers complete these same behavioral ratings again. These
behavior ratings were then obtained annually from boys' teachers until they
turned 15. All told, therefore,
ratings for aggression, oppositionality, and hyperactivity were obtained on
the boys a total of 7 different times from 7 different teachers. As best the
authors could tell, there were no major differences between the boys they
were able to track down and the ones who were lost to the study.
Finally, subsequent to these 7 waves of teacher behavior ratings, each boy
was interviewed at 15, 16, and 17 and asked about their involvement in a
variety of delinquent and antisocial acts during the past 12 months. All in
all, this was a monumental data collection effort and an exceptionally rich
longitudinal set.
(Too bad, however, that girls were not included).
Results
The first question the authors were interested in concerns how children's
scores on the 3 types of problem behaviors tended to change over the course
of their development. Using a very complicated set of statistical tests, the
authors first identified the most commonly occurring "pathways" for
each of the 3 problem behaviors. (By "pathway", I simply mean how
children's problems in these 3 areas changed over time - e.g. did they get
worse, stay the same, or get better?)
Four different pathways were identified. These were:
1. Persistently high - children who had high scores compared to their peers
at each assessment;
2. High decliners - children who started out high compared to their peers but
whose scores declined into an average range over time.
3. Moderate decliners - children who stared out moderately high compared to
peers but who also declined over time.
4. Persistently low - children who received low scores compared to their
peers at each assessment.
(Interestingly, there was no group of boys that started out with low ratings
on any of the 3 problem behaviors and then increased over time. This may
certainly have occurred for some boys, but not enough for this to show up as
a common pathway like those noted above. It may also reflect problems in how
the behaviors were measured in this study.)
Each child was then assigned to 1 of these 4 pathways for each behavior.
Thus, it is possible for a child to have been in the persistently high group
for aggression, and in the persistently low group for hyperactivity and
oppositionality. Or, a child could have been in the high group for all 3
behaviors. A number of different combinations are of course possible.
To begin with, it is interesting to note the percentage of the sample that
fell into the different groups (i.e. persistently low, moderate decliners,
high decliners, and persistently high for each of the 3 behaviors. Across the
3 types of behavior, the percentages were as follows:
Persistently high - about 5% of the sample for each behavior;
High decliners - between 20-30%, depending on which behavior is being
considered;
Moderate decliners - about 50% of the sample for each behavior;
Persistently low - between 15-25% of the sample depending on the behavior;
Thus, it was quite unusual for a child to show consistent elevations -
relative to his peers - on teacher ratings of hyperactivity, oppositional
behavior, or physical aggression.
Next, the authors examined the degree of overlap that existed across the
behaviors for children in the different groups. In other words, how likely
was it for a child to be in the same grouping for each type of problem
behavior?
Of particular interest here is the overlap that existed between children who
were in the persistently high group for any of the 3 behaviors. Let's focus
on children who were persistently high for hyperactivity. How likely were
these children to also show persistently elevated levels for either
oppositional behavior or physical aggression?
The answer is more than children who were not persistently high for
hyperactivity, but not nearly so often as you might expect. Only 13% of the
boys who were persistently high on hyperactivity were also persistently high
for physically aggression. Only 23%
were persistently high on oppositional behavior. These data make it clear
that the vast majority of persistently hyperactive boys were not showing
persistent difficulties in either of the other two problem behaviors.
The message here is simple and very important: oppositional behavior and
aggression often develop and persist for reasons that have little or nothing
to do with a child's having ADHD. When a child with ADHD also displays these
other behaviors it should not be understood as being "part of the
child's ADHD".
In the second set of analyses the authors examined how well children's
classification on hyperactivity, oppositional behavior, and aggression (e.g.
were they persistently high or persistently low) predicted their involvement
in antisocial and delinquent behavior at age 17. For children in the high and
low groups for hyperactivity, oppositionality, and physical aggression the
average number of offenses reported during the prior 12 months were as shown
below:
Hyperactivity
Oppositionality Aggression
High 2.34 6.38 7.17
Low .33 .01 .06
An examination of these numbers clearly indicates that persistently
hyperactive boys actually reported far fewer offenses than boys who were
persistently oppositional or persistently aggressive. (Note: It would have
been nice to consider these outcomes for boys who were persistently high on 2
or 3 of the different problem behaviors. This was not done, however,
primarily because even with such a large sample, the number of children
required to do this type of analysis were not sufficient.)
Even more telling are results of analyses in which boys' grouping on all 3
behaviors were used simultaneously to predict their involvement in delinquent
and antisocial behavior at ages 15,
16, and 17. These results are a bit complicated but here is an overall
summary:
* Boys' classification for physical aggression was the only significant
predictor of both self-reported violence and self-reported serious
delinquency. What this means is that boys' classification on either hyperactivity
or oppositional behavior did not really matter when trying to predict these
outcomes - only their classification for aggressive behavior mattered.
* For self-reported theft, only boys' classification on the oppositional
behavior dimension was a significant predictor.
* Boys who show high levels of hyperactivity from kindergarten through high
school are at much less risk of juvenile delinquency than those who show high
levels of physical aggression or oppositional behavior.
Implications
These results have very important implications. The very good news, I think,
is that hyperactivity by itself does not increase a child's risk for the
types of antisocial outcomes considered in this study. Now, it is important
to recognize that the ratings of hyperactivity that were used in this study
were not sufficient to determine whether a child had ADHD, but I think it is
reasonable to extend the conclusion above to make this statement:
"When a male child has ADHD but does not also show persistently high levels
of either oppositional or aggressive behavior, he is not likely to become
involved in any serious antisocial behavior as an adolescent."
I make this statement recognizing that it is going a bit beyond what can be
clearly concluded from the data of this study, but it is a stretch that is
supported by the efforts of other researchers as well.
There are a number of reasons why this is quite important but the one that
really sticks out in my mind concerns just how often parents may confuse
oppositional and/or aggressive behavior with ADHD. Typically, what I have
seen happen is that after a child has been diagnosed with ADHD, these other
types of behavior get explained away as being part of the child's ADHD. This
is incorrect, however. These other behaviors are not symptoms of ADHD, and as
this study clearly indicates, high levels of these other behaviors are often
not even associated with ADHD and lead to very different outcomes than do
ADHD symptoms alone.
The take-home message from this study is that if a child with ADHD is also
displaying high levels of oppositional and/or aggressive behavior, do not
assume that treating the ADHD by whatever means is tantamount to addressing
these other difficulties as well. Instead, it is essential to make sure that these
other problems are being specifically targeted in a child's treatment plan,
and these difficulties need to be treated every bit as aggressively as the
child's primary ADHD symptoms themselves.
|