Medication adherence in clinical trials: missing the forest for the trees

Medication adherence is an important determinant of validity in clinical trials but has been underappreciated and incompletely measured. Nonadherence can bias efficacy and safety results, increase trial costs, and undermine regulatory conclusions. This commentary reviews the effects of nonadherence, addresses shortcomings of the conventional self-report methods, and reviews newer tools like electronic monitoring, serum drug concentrations, and ingestible sensors. Behavioral interventions, digital tools, and trial design changes are emphasized as ways to improve adherence. Practicality, cost-effectiveness, and feasibility differ across patient populations and study environments, highlighting the importance of customized approaches. Frameworks such as EMERGE seek to harmonize reporting on adherence, but additional validation is necessary. Scalable real-time tracking, predictive modeling, and harmonized definitions are areas that future research needs to address to enhance comparability and decision-making. Addressing adherence in a comprehensive manner will enhance the consistency of clinical trials and the generalizability of results to everyday practice.