Randomized controlled trials (RCTs) suggest that ADHD medication has beneficial short-term effects on symptoms of ADHD, and some co-occurring disorders. RCTs of ADHD medication have serious limitations, however, including the inability to generalize to individuals with serious comorbid problems, study rare-but-serious outcomes, and adequately examine long-term outcomes. Therefore, there is a critical need to more precisely identify the specific benefits and risks of ADHD medication, particularly with samples and designs that can explore rare outcomes. The presentation will provide examples of how the analyses of large-scale health insurance claims datasets in the United States and national registry data (e.g., in Sweden) can help specify the benefits and risks of ADHD medication when using advanced research designs. In particular, the talk will describe how the use of within-individual comparisons (i.e., using periods in which an individual is not dispensed medication as the comparison for periods in which he/she was dispensed medication) can help elucidate the associations between ADHD medication and numerous outcomes, including serious substance use problems, motor vehicle accidents, depression/suicide attempts, seizures, and accidents/injuries.