The use of prescription amphetamines for the treatment of attention-deficit hyperactivity disorder (ADHD) has markedly increased in the last fifteen years, with the greatest increase in adolescents and young adults. In 2007, the FDA mandated changes to stimulant labels to warn of increases in psychosis and mania in patients without pre-existing conditions. Our multidisciplinary research team from Brigham and Women's Hospital and McLean Hospital recently published a landmark cohort study demonstrating an increased risk of psychotic episodes in new users of amphetamine compared to methylphenidate in adolescents and young adults with ADHD. Alarmingly, the use of amphetamines increased four-fold over the study period (2004 ? 2015). In addition, our data show that prescription amphetamine use has dramatically increased in patients with pre-existing bipolar disorder. Thus, there is an urgent need to accelerate knowledge about prescriber-level factors, patient-level factors and disease states that potentiate the risk of psychosis and mania with prescription amphetamines. Large-scale studies using real-world data are the only option to study the risk of infrequent and serious adverse outcomes.
In Aim 1, using incident-user cohort study designs, we will utilize data from two national administrative claims databases to identify prescriber dosing strategies that increase the risk of first-episode psychosis in ~220,000 adolescents and adults with ADHD initiating amphetamines. We hypothesize that there will be a dose-dependent increase in the risk of psychosis. The goal of Aim 2 is to identify patient subgroups at heightened risk of psychosis/mania with prescription amphetamine use. To accomplish this goal, we will perform a case control study using electronic medical records (EMR) from McLean Hospital in ~1,650 patients hospitalized for an initial episode of psychosis or mania compared to ~3,300 controls with a first psychiatric hospitalization for reasons other than psychosis or mania. We will apply natural language processing to unstructured narrative notes to capture detailed patient data not available in claims data. We hypothesize that patients with a family history of psychiatric illness will have an increased risk of psychosis/mania with amphetamines compared to patients without a family history. We also hypothesize that patients with concurrent cannabis use will have an increased risk of psychosis/mania with amphetamines compared to patients without cannabis use.
Aim 3 compares the risk of treatment-emergent mania in ~52,000 new users of amphetamine versus methylphenidate in a cohort study of patients with pre-existing bipolar disorder, an urgent issue given the known risk of mania with stimulant use. Combined, these studies will provide actionable evidence to be incorporated into clinical practice guidelines with the goal of mitigating the risk of psychosis and mania with prescription stimulants. The proposed research agenda aligns with NIMH Strategic Objective 3.3 ?Striving for Prevention and Cure? on testing interventions in real-world practice settings to evaluate the impact of patient- and provider-level factors on clinical outcomes.
In recent years, 6.1 million U.S. children and adolescents were diagnosed with attention deficit hyperactivity disorder (ADHD) and 16 million adults reported past-year use of prescription stimulants. The research team recently found an increased risk of psychosis in new users of prescription amphetamines compared to new users of prescription methylphenidate in adolescents and young adults with ADHD, translating to potentially increased risk of psychosis for thousands of U.S. patients. Therefore, identifying patient subgroups and prescribing practices that increase the risk of psychosis and mania with prescription amphetamines has enormous public health significance and will guide providers to avoid amphetamines in patients at highest risk, preventing future cases of psychosis and mania.
