Among young people up to 18 years of age, approximately 5 % world-wide are said to have attention deficit hyperactivity disorder (ADHD), a neurodevelopmental syndrome of impulsiveness, inattention, and overactivity that can result in long-term educational and social disadvantage [1, 2]. Symptoms persist into adulthood for a significant minority [3]. Co-morbid substance use disorders (SUD) involving alcohol and cannabis misuse but also stimulants, depressants, hallucinogens and now novel psychoactive substances (NPS) or ‘legal highs’ emerge in adolescence [4]. Young people with ADHD are significantly more likely to develop SUDs than those without ADHD [5].

Many people with ADHD are prescribed medication for their symptoms, particularly stimulants, (often methylphenidate and dexamphetamine) and atomoxetine (NICE 2006). If an individual is suspected of misusing substances, it is crucial to understand if it safe also to prescribe medication. We contacted the manufacturers who recommended that stimulants and atomoxetine not be coingested with alcohol or illicit substances. They had no specific information on potential interactions.

There is limited empirical evidence to guide whether it is advised to treat ADHD before, simultaneously, or only after remission of SUDs [6, 7]. Current guidelines suggest that cannabis use may not be a contraindication to pharmacological treatment but due to shared neurochemical mechanisms, and while not explicitly mentioned, certain novel psychoactive substances (NPAs), ‘cocaine is likely to be a real hazard’ [8]. In the absence of such information, faced with adolescents misusing substances the tendency is for clinicians to avoid prescribing. As adolescents with untreated compared to treated ADHD may have poorer long-term outcomes, this may leave the most vulnerable without adequate intervention [9].

Methylphenidate and dexamphetamine are sympathomimetic agents thought to act both in the central nervous system (CNS) and peripherally by enhancing dopaminergic and noradrenergic transmission through blockade of relevant transporters [10]. There is theoretical potential and some animal evidence for additive effects between prescribed and non-prescribed sympathomimetic agents [11, 12] potentially resulting in a toxic sympathomimetic syndrome with prominent cardiac and neurological effects [10, 12]. Such a syndrome may underly so-called ‘excited delirium’; agitation, paranoia, hyperthermia and muscle breakdown, symptoms similar to neuroleptic malignant syndrome [13]. However, there is also evidence that different stimulants may compete for receptor binding sites limiting the potential for additive effects [14].

Atomoxetine is a selective noradrenaline reuptake inhibitor metabolised by the hepatic oxidase system [15]. It has a chemical structure that resembles the SSRI fluoxetine, that has been implicated in the emergence of mania or the serotonin syndrome, comprised of mental, autonomic and neurological effects [16]. This suggests that toxic or tolerability effects could arise in alcohol misusing individuals [17], through additive effects with stimulants or with NPS that impact serotonin receptors.

We systematically searched three databases (Medline, EMBASE and PsychINFO) from the past 22-years (1992 – June 2014) using PRISMA guidelines [18]. Only those articles that had abstracts available in the English language were included. We used the key search terms alcohol, substance related disorders, death, and combined these with methylphenidate, atomoxetine, dexamphetamine and lisdexamfetamine in turn.

Titles and abstracts of all articles found were screened by XB and PM. Full text articles were screened by XB and PM. The references of all full text articles included were also hand searched to identify any other potentially relevant articles. Study quality was assessed using the relevant tools from Critical Appraisal Skills Program (CASP) or the STROBE document as appropriate to the study type [19].

We included articles which mentioned the coingestion of ADHD medication (methylphenidate, dexamphetamine, lisdexamfetamine and atomoxetine) with another substance including alcohol, publications relating to animal models, and, due to the paucity of data regarding adolescents, all age groups. Also, to clarify mechanisms that might underlie toxicity, we included certain articles covering the physiological effects of ADHD medication alone. We excluded articles relating to the abuse of ADHD medication alone, case reports, articles relating to exposure of illicit drugs or alcohol in utero, foetal alcohol syndrome, and to cigarette smokers only.

Our main outcome was to identify any potential effects (whether positive or negative) on an individual should they combine their ADHD medication with alcohol or an illicit substance. We acknowledge the dangers of diversion of stimulant medication in the substance misuse population; however our study did not focus on this element.

As diagnoses of ADHD have increased among young people of substance using age, the problem of prescribing for substance users with ADHD has become a common dilemma [31]. This review explores current available information concerning risks of toxicity attached to pharmacological treatment of ADHD in the context of substance misuse.

On the basis of experimental studies, and studies of alcohol dependent individuals, when methylphenidate, dexamphetamine or atomoxetine are ingested orally with alcohol, acute severe side-effects appear uncommon. However, as co-ingestion of stimulants such as methylphenidate or amphetamine with alcohol may enable more ‘late night partying’ and high levels of alcohol consumption, there is enhanced risk of more chronic harm from alcohol toxic effects [29, 31]. We did not identify serious adverse side-effects in those with poly-drug misuse using ADHD medication, even in overdose.

Despite the theoretical possibility of serotonergic, hyperdopaminergic or other hyper stimulated states, on the basis of limited evidence, apart from one study suggesting hypertension following a combination of methylphenidate and MDMA, ADHD medication does not normally add to toxicity of alcohol or illicit substances from the pre-novel psychoactive substance (NPS) era.

Limitations include the relatively small numbers in the studies cited. In part, this may reflect a fundamental problem with the science highlighted by Bauman [41]; banning psychoactive agents seriously impedes research into the effects of substances in widespread use alone and in combination by large numbers of young people. In addition, there is no data concerning combinations of cannabis and ADHD medications. However, as cannabis is ubiquitous in the substance using community, it is likely that many adolescents prescribed medications are also using cannabis apparently without acute mishap. However, we lack data concerning any longer term problems with this practice. The effect of combining ADHD medication and NPS is unknown and the numbers in published studies are too few to exclude unusual reactions. ADHD medication appears less effective or even ineffective among those with SUD and if nevertheless prescribed, its use, efficacy and combination with other substances should be closely monitored in an appropriately cautioned consumer.

With appropriate caution, including explanations of the limits of the science, it appears that a blanket contraindication is not justified when considering treatment of ADHD in selected substance users.