Online Program

Studying Drug Safety: From RCTs to OCER
Anne E. Libby, University of Colorado Denver 
*Robert Valuck, University of Colorado 


Randomized controlled trials (RCTs) are the gold standard design for establishing the efficacy and safety of drugs for regulatory approval. Despite their strengths, RCTs are limited in their ability to detect drug safety issues. Observational (nonrandomized) designs offer an alternative approach to RCTs for studying the relationships between drugs, diseases, and risks of adverse events. Ecological studies, prospective and retrospective cohort studies, and case-control studies have unique strengths for quantifying the relationships between drug and outcome at the population level and for determining the risk of rare, serious outcomes in “real world” treatment populations. Traditionally, such studies have been conducted using administrative claims data from various sources. These studies have been limited in their own right, due to the nature of administrative data. New federated data networks such as the AHRQ-funded Distributed Ambulatory Research in Therapeutics Network (DARTNet) offer unique opportunities to advance the science of comparative effectiveness and safety research in primary care medical practice, using enhanced clinical data and patient reported outcomes (PROs). DARTNet is a prototype system that links electronic health record (EHR)-enabled practices to facilitate data collection from historical and point of care clinical data at the practice level. The use of DARTNet for conducting postmarketing drug safety studies is illustrated through two case examples: Type II oral diabetes medications and risk of hypoglycemic and hepatic events; and antidepressants and risk of suicidality