Higher risk of psychosis with amphetamines than methylphenidate in treatment of ADHD

Reference

Moran LV, Ongur D, Hsu J, Castro VM, Perlis RH, Schneeweiss S. Psychosis with methylphenidate or amphetamine in patients with ADHD. N Engl J Med 2019; 380(12):1128-1138.

Synopsis

Amphetamine and methylphenidate are both stimulants widely used in the United States for the treatment of ADHD. There have been few head-to-head trials of these drugs, and no evidence from previous network meta-analyses that amphetamines provide a clinically important treatment advantage, if any. They have somewhat different pharmacologic properties, and amphetamines are generally not used in countries outside the United States because of concerns about adverse effects such as psychosis. These authors used 2 large claims databases to identify US patients aged 13 to 25 years who had been newly prescribed one of these drugs between 2004 and 2015. Patients with comorbid conditions that might also cause psychosis were excluded, as were those taking any other stimulants, oral glucocorticoids, mood stabilizers (such as lithium), or antipsychotic drugs. Drugs classified as "amphetamine" included amphetamine-dextroamphetamine, dextroamphetamine, and lisdexamfetamine. Drugs classified as "methylphenidate" included methylphenidate and dexmethylphenidate. In the 2 claims databases, the authors identified 5.26 million patients (out of more than 250 million total patients) who were exposed to a stimulant, of whom approximately 339,919 met inclusion criteria and were included in the final cohort. They identified 119,708 users of methylphenidate and 218,211 users of amphetamine, with a mean age of 17 years; 37% were women. Approximately 17% had comorbid depression, and approximately 80% were cared for by family physicians or pediatricians. A propensity score matching analysis was done to match patients who were otherwise similar except one received amphetamine and the other received methylphenidate. The risk of psychosis was significantly higher with amphetamine (hazard ratio 1.65; 95% CI 1.31 - 2.09). The risk was slightly higher for patients aged 13 to 17 years than for those aged 18 to 25 years, and also for those receiving care from a primary care physician as opposed to a psychiatrist (though there may be some confounding because younger patients may be more likely to see a pediatrician or family physician than a psychiatrist). The absolute risk of psychosis was low: 0.1% in those receiving methylphenidate compared to 0.21% in those receiving amphetamine (number needed to treat to harm = 909). There was no difference between groups in other psychiatric events, which supports causality. A limitation of the study is the lack of information on race, ethnicity, or socioeconomic status in this privately insured population.