To describe the design and rationale of an investigator-initiated observational study to examine the cardiovascular safety of the following commonly-used medications to treat attention deficit hyperactivity disorder (ADHD): amphetamines, methylphenidate, and atomoxetine.
We are conducting an observational cohort study using data from five large Medicaid programs and the HealthCore Integrated Research Database (HIRDSM), which is derived from administrative data from commercial health plans. Our primary outcomes of interest are (1) sudden death/ventricular arrhythmia, (2) stroke, (3) myocardial infarction, and (4) stroke or myocardial infarction as a composite outcome. These claims diagnoses have been previously validated in adults, and the positive predictive value in children will be examined as part of this study. Secondary outcomes are (1) all-cause death, (2) non-suicide death, and (3) non-accident death. All design decisions have been made to minimize bias toward the null. Based on our pilot data, we expect to have at least 90% power to detect a minimum detectable hazard ratio (HR) of 3.0 in children and adolescents who initiate an ADHD medication for each outcome of interest (except for MI, for which the expected minimum detectable HR is 7.9). The expected minimum detectable HR is 1.7 for each outcome for adult incident ADHD medication users.
Potential limitations to this study include a low expected event rate in children and adolescents, potentially incomplete ascertainment of outcomes, and potential confounding by unmeasured variables. Nevertheless, this study will provide important information about the cardiovascular safety of ADHD medications. Copyright © 2010 John Wiley & Sons, Ltd.