Evid Based Ment Health. 2014;17(1)
Humphreys KL, Eng T, Lee SS. Stimulant medication and substance use outcomes: a meta-analysis. JAMA Psychiatry 2013; 70:740–9.
Does exposure to stimulant medication for attention deficit hyperactivity disorder (ADHD) have an effect on the lifetime risk of substance use or dependence?
Lifetime substance use or dependence (alcohol, cocaine, marijuana, nicotine and illicit non-specific drug). Method of assessing substance use/dependence is not reported for included studies.
Systematic review and meta-analysis.
PubMed search for key words related to ADHD and substance use, supplemented by a hand search of reference lists of relevant articles, listservs of research organisations and contact with authors who have published longitudinal studies of children with and without ADHD; study inclusion dates January 1980 to February 2012.
Study Selection and Analysis
Longitudinal studies in children with ADHD where pharmacological treatment preceded the measurement of substance use. ORs were calculated for the risk of substance use/dependence among children with ADHD treated with stimulant medication compared with those not treated. Results were pooled using a random-effects model, and heterogeneity was assessed using the Cochran Q test. Publication bias was assessed using the Egger and Begg tests. Additional analyses examined whether the association between stimulant treatment and substance use/dependence was moderated by demographic and methodological factors of: age at follow-up, sex, race, age at initial ADHD assessment, percentage of participants in the medicated group, study population source, diagnostic and statistical manual of mental disorders version used to diagnose ADHD, and length of follow-up.
Fifteen longitudinal studies (n=2565) were included. Pharmacological treatment for ADHD, typically methylphenidate, did not influence the risk of later substance use or dependence. Pharmacological treatment did not influence the risk of alcohol use (OR 0.99, 95% CI 0.61 to 1.62, 4 studies), alcohol abuse/dependence (OR 0.80, 95% CI 0.46 to 1.38, 11 studies), cocaine use (OR 2.21, 95% CI 0.87 to 5.65, 3 studies), cocaine abuse/dependence (OR 1.10, 95% CI 0.51 to 2.38, 7 studies), marijuana use (OR 1.01, 95% CI 0.68 to 1.50, 4 studies), marijuana abuse/dependence (OR 0.97, 95% CI 0.59 to 1.59, 9 studies), nicotine use (OR 1.55, 95% CI 0.73 to 3.30, 4 studies), nicotine abuse/dependence (OR 1.34, 95% CI 0.90 to 1.99, 6 studies), non-specific drug use (OR 1.27, 9% CI 0.88 to 1.82, 3 studies) and non-specific drug abuse/dependence (OR 0.85, 95% CI 0.51 to 1.40, 7 studies). No significant publication bias was detected. Moderator analyses found that a higher proportion of children with ADHD receiving pharmacological treatment reduced the risk of alcohol dependence/abuse, whereas longer follow-up was associated with increased risk of alcohol abuse/dependence. A higher proportion of males in the study population increased the risk of non-specific drug abuse/dependence.
Treatment of ADHD with stimulant medication does not predispose towards or protect against later substance use or abuse.
One of the included studies had a population of children with a reading disorder rather than ADHD.
ADHD has long been recognised as a risk factor for substance abuse and dependence, particularly in association with conduct disorder. Increasing rates of diagnosis and treatment have been paralleled by increasing levels of community concern about whether stimulant medication could increase individual susceptibility to substance use. An earlier meta-analysis by Wilens et al  suggested that stimulant medication had a protective effect for subsequent substance use disorder. Although the larger meta-analysis by Humphrey and colleagues did not replicate this result, it still suggests that, on average, stimulant medication appears not to increase the risk of substance use and addiction across a range of substances.
Meta-analyses are useful because they can derive an overall risk based on large cohorts derived from a series of studies. The main drawback is that much detail is lost. This means that significant confounders may not be evaluated and findings can be difficult to interpret. The individual studies were carried out in a range of settings and it was noted that as the size of the treated group declined relative to the untreated group, the proportion of treated individuals meeting the diagnostic criteria for alcohol abuse and dependence increased. Although this could be indicative of subgroups of individuals with ADHD for whom stimulant medication might increase the risk of alcohol abuse, it could also reflect different prescribing practises.
A meta-analysis is limited by the quality of the data reported in the included studies, and importantly, Humphrey and colleagues were unable to control for baseline symptom severity. They could not therefore determine whether those who were treated were more severely affected and should actually have had a higher risk of substance use.
Although reassuring, this study is unlikely to change clinical practise. The need remains for longitudinal studies that extend well into adulthood and report symptom severity as well as the presence of associated diagnoses and important individual and social factors. Once all of these have been controlled for, it is likely that any residual contribution of stimulant medication, whether protective or otherwise, will be small.
1. 1. Wilens TE, Faraone SV, Biederman J, et al. Does stimulant therapy of attention-deficit/hyperactivity disorder beget later substance abuse? A meta-analytic review of the literature. Pediatrics 2003;111:179–85.
Sources of funding
National Institutes of Health.
National Institutes of Health.
Evid Based Ment Health. 2014;17(1) © 2014 BMJ Publishing Group Ltd, the British Psychological Society and the Royal College of Psychiatrists