Natural Solutions Radio header image


ADHD and Substance Use

David Rabiner, Ph.D.
Duke University

Attention Research Update

Although it was previously believed that most children with ADHD would "outgrow" the disorder by adolescence, several longitudinal studies have provided clear evidence that ADHD often persists into adolescence and adulthood.  Prior research has also established that childhood ADHD is associated with a number of negative outcomes during adolescence, with school-related difficulties being especially prominent.

A particular concern of many parents and professionals is the extent to which childhood ADHD is a risk factor for substance use during adolescence.  Some researchers have reported higher rates of cigarette use for individuals previously diagnosed with ADHD, but not higher rates of alcohol use or other drugs.  Others have reported higher rates of drug use but not alcohol use, while still others have not found any elevated substance use among adolescents previously diagnosed with ADHD.

There are several possible explanations for these discrepant findings.  First, prior studies of this issue did not always include comprehensive assessments of adolescent substance use.  Second, because many adolescents experiment with different substances it is important to examine the frequency and quantity of use, including heavy use, rather than whether or not any substance use has occurred.  Finally, age of first substance use is a well-established predictor of later problematic use, and this variable has often not been sufficiently included in prior studies.

Another unresolved issue is the extent to which childhood ADHD symptoms per se, as opposed to the conduct problems that often accompany ADHD, are most directly associated with the development of substance use problems.  Some studies suggest that childhood aggression and other Conduct Disorder symptoms predict substance use in adolescence, while core ADHD symptoms - i.e., inattention and hyperactivity-impulsivity - do not.  (Note: For a listing of inattentive and hyperactive-impulsive symptoms go to  To view symptoms of ODD and CD go to ).

A limitation of these prior studies is that they vary in the adequacy with which childhood ADHD and Conduct Disorder were assessed, and they did not separately consider the contribution of attention problems and hyperactivity-impulsivity to later substance use.  These two groups of symptoms are associated with different difficulties, however (e.g., inattention is reliably associated with academic difficulties while hyperactivity-impulsivity is more closely linked to behavior problems), and the failure to consider these dimensions separately may have obscured associations between ADHD symptoms in childhood and substance use during adolescence.

A study published in a recent issue of the Journal of Abnormal Psychology [Molina & Pelham (2003). Childhood predictors of adolescent substance use in a longitudinal study of children with ADHD, 112, 497-507] provides new data that clarifies several of these prior inconsistencies.  Participants were 142 adolescents with childhood ADHD and 100 demographically similar adolescents without childhood ADHD.  Participants with ADHD had been diagnosed between the ages of 5 and 12 and received services at that time, although information on the nature and extent of treatment services received is not provided.  Comparison children without ADHD were recruited at that time from similar neighborhoods.

When participants reached adolescence, their parents were recontacted to participate in the study.  At this time, participants with and without ADHD were given an extensive interview pertaining to their history of substance use.  In addition, adolescents and parents were interviewed to assess the extent of current ADHD symptoms as well as any co-occurring psychiatric difficulties.  An average of 5.26 years had elapsed between the childhood diagnosis of ADHD and this assessment during adolescence.  


The authors first examined lifetime differences in use of any substances between the two groups.  Compared to adolescents without a childhood diagnosis of ADHD, those with a childhood diagnosis were not significantly more likely to have used alcohol, cigarettes, or marijuana.  They were, however, more then 3 times as likely to have used at least one illicit drug beside marijuana - i.e. 20.4% of adolescents with childhood ADHD reported this compared to only 7% of comparison subjects.  Although this is concerning, it is important to note that nearly 80% of the adolescents with childhood ADHD did not report illicit drug use.  


In addition to lifetime substance use history, the researchers considered group differences in the level of substance use.  Adolescents with a childhood history of ADHD were:

- about 3 times more likely to be a daily smoker (30.4% vs. 12%);

-  had consumed more cigarettes in the past 6 months;

- were about twice as likely to have been drunk multiple times during the 6 months
  (23.2% vs. 12%); and,

-  had higher rates of alcohol problems overall (15.5% vs. 85).

On average, adolescents with ADHD began smoking about 13 months earlier than other adolescents and used their first illicit drug about 7 months earlier.  These differences are important because earlier onset of substance use is associated with higher rates of subsequent substance use problems.  


Next, the authors examined predictors of substance use among adolescents with childhood ADHD.  Predictor variables included the severity of inattentive symptoms during childhood, the severity of hyperactive-impulsive symptoms during childhood, and the severity of other behavior problems (i.e. oppositional defiant disorder symptoms and conduct disorder symptoms) during childhood.  A variety of substance use outcomes were considered including those pertaining to cigarette use, alcohol use, marijuana use, and use of other illicit drugs.

In general, inattentive symptoms were a consistent predictor of different adolescent substance use outcomes while hyperactive-impulsive symptoms and ODD-CD symptoms were not.

The main exception to this pertained to illicit drug use (i.e. the use of drugs besides marijuana) where childhood ODD/CD symptoms were significant predictors but childhood inattention symptoms were not.  


Finally, the authors considered whether the persistence of ADHD into adolescence, and the co-occurrence during adolescence of ADHD and Conduct Disorder, was associated with adolescent substance use.  Interestingly, 72% of adolescents who had been diagnosed with ADHD during childhood continued to meet diagnostic criteria for ADHD at follow-up, and about 1/3 of these adolescents also had a diagnosis of Conduct Disorder.

Compared to adolescents without persistent ADHD those with persistent ADHD were more than twice as likely to be smokers.  Rates of other substance use problems, however, were roughly comparable.  For adolescents without persistent ADHD, substance use patterns were not different from those of adolescents never diagnosned with ADHD.

Adolescents with persistent ADHD who had also developed Conduct Disorder had the highest rates of substance use problems of any group, including adolescents with persistent ADHD but who had not developed Conduct Disorder.  


Results from this study provide clear evidence that childhood ADHD is associated with increased risk of use and abuse of alcohol and with earlier and heavier use of tobacoo and other drugs in the teenage years.  Of particular interest is the finding that inattentive symptoms more strongly predict adolescent substance than hyperactive-impulsive symtoms or childhood antisocial behavior.  The only exception to this was that childhood antisocial behavior was a stronger predictor of non-marijuanna illicit drug use.

Although inattentive ADHD symptoms were a key predictor of adolescent substance use, it is important to note that substance use problems were strongest among adolescents who had also developed Conduct Disorder.  Thus, the combination of high levels of inattentive symptoms and the development of serious antisocial behavior is a substantial risk factor for substance use and abuse during adolescence.  This finding suggests that effective treatment of inattentive symptoms, and preventing the development of serious conduct problems, may be especially important in reducing substance use problems during adolescence.

It is important to put these findings within an appropriate context.  Although children with ADHD were more likely than comparison children to smoke, drink, and use other illict drugs as adolescents, most participants did not report significant problems with substance use. For example, even among adolescents with persistent ADHD and Conduct Disorder - the highest risk group - less than 30% reported being drunk multiple times in the prior 6 months or using illict drugs other than marijuanna during the same period.

The authors note that relative to other populations at risk for substance use problems, the risks associated with childhood ADHD are roughly comparable.  As an example, they suggest that ADHD in childhood appears to be an equivalent risk factor for the development of substance use problems as having a positive family history of substance use disorder.

Why might inattentive ADHD symptoms be such an important predictor of adolescent substance use?  The authors note that attention problems are more strongly associated with academic failure than hyperactive-impulsive symptoms.  They suggest that ongoing academic struggles may lead children to drift away from conventional peers oriented towards academic success and toward nonconformist peer groups where substance abuse is tolerated, modelled, and encouraged.  Although this theory was not specifically tested, it highlights the importance of carefully attending to the academic success of students with ADHD and making sure that an appropriate network of educational supports are in place.

The authors also offer an interesting speculation about the relationship between inattentive symptoms and the substantially greater rates of smoking. They note that nicotine has psychostimulant properties that enhance vigilance and attention, and suggest that this may encourage earlier and heavier cigarette use by adolescents as a method of "self-medication".  They reason improvements in cognitive functioning that result from cigarette use may be rewarding for individuals with ADHD, and that this could explain their increased propensity to become heavy smokers.

Overall, results from this study highlight the importance of carefully attending to early substance experimentation and use among adolescents with ADHD.  This is especially true among children with prominent inattentive symptoms and among those who develop the serious behavior problems associated with Conduct Disorder.  Given the clear health consequences of smoking, and the substantially higher rates of smoking among adolescents with a childhood history of ADHD, this may warrant particular attention even though cigarettes are not an illegal substance at older ages.

There are several limitations of this study that should be noted.  First, participants in this study were adolescents who had received treatment at an ADHD specialty clinic during childhood, and the extent to which these results can be generalized to the wider population of individuals with ADHD is not known.

Second, the vast majority of participants in this study were males, and testing gender specific effects was thus not possible.  The extent to which the findings reported would apply to girls specifically is thus also unknown.

Finally, although results highlight that ADHD symptoms - especially attention problems - increase the risk for adolescent substance use, the authors did not examine factors that may have protected adolescents from developing substance use problems.  As noted above, most participants with childhood ADHD did not develop substance use problems, and it would be helpful to know what factors inhibited the development of such difficulties in many of the participants.

One explanation is that those with less severe symptoms were less likely to become substance users.  Even among participants with more severe ADHD symptoms in childhood, however, rates of adolescent substance use were variable.  What factors may have accounted for this?

Given that participants in this study had all received some form of treatment, how did treatment effectiveness related to subsequent substance use?  Were particular treatment(s) more helpful than others in reducing the risk of substance use in adolescence?  Were any partciular family factors associated with reduced risk of later substance use.

These are interesting and important questions that will hopefully be addressed in subsequent efforts to build on this interesting and informative study.

Thanks again for your ongoing interest in the newsletter. I hope you enjoyed the above article and found it to be useful to you.

The next issue of Attention Research Update will be sent out to you in approximately 4 weeks.


David Rabiner, Ph.D.
Senior Research Scientist
Duke University

(c) 2003 David Rabiner, Ph.D.

Copyright Issues?