Are children who take stimulant medications more likely to abuse drugs or alcohol later in life?

Here's an overview of the major research on this topic:

There are seven widely-cited, well-known, published studies that have followed children with AD/HD into their adolescent or early adulthood years; and have analyzed substance use, alcoholism, and/or drug abuse as a function of prior stimulant treatment.  Six of these studies specifically addressed whether treatment with stimulant medication compared with no stimulant medication treatment predicted later substance use.  Of these six,

• three studies reported lower rates of substance use or substance use disorder;
• one study reported higher rates of nicotine and cocaine dependence by adulthood;
• two studies reported no statistically significant associations (although a small connection to more cocaine use in adulthood was reported but was de-emphasized for statistical reasons). 

One study with few untreated children reported that an earlier age of treatment initiation among the treated children predicted a decreased likelihood of non-alcohol substance use disorder.  As of this writing, there are several new studies whose results have been presented at professional scientific conferences but findings are not yet published. 

The hope is that these studies, some of which have large samples, will be able to address some of the study design challenges inherent in answering this complicated question.  For example, studies contributing to the mixed results have usually been small (e.g., about 100 subjects), resulting in very limited numbers of non-medicated children to compare to much larger numbers of medicated children. 

When few children in the study are not medicated, it becomes almost impossible to determine whether the treatment itself affects later substance abuse or whether characteristics that underlie receipt of treatment (such as severity of problems) cause this outcome.  In addition, studies have varied in their approach to the question, considering presence or absence of treatment in the lifetime, duration of treatment, or age of treatment initiation.  Substance use and abuse has been measured differently across studies, and the follow-up age varies across studies (adolescence versus adulthood). 

Again, the most conservative interpretation of the literature is that we do not yet know whether stimulant treatment predicts later use and abuse of substances.  Presently, the relatively smaller number of published studies finding increased risk of substance use or dependence for medicated children usually leads most practitioners to give more weight to the benefits of medication treatment when it appears to be indicated. Proper monitoring, particularly in the face of other risk factors such as conduct problems or a family history of addictions, is always warranted.

Updated: August 2009