Limitations of Randomized Controlled Trials for Detecting and Characterizing Harms: Methodological Reflections on the PATHWAYS Trial

PATHWAYS is the first randomized controlled trial (RCT) to evaluate puberty suppression in youths with gender incongruence. This article examines whether methodological features of PATHWAYS allow a comprehensive safety assessment. Although RCTs are considered the gold standard for establishing effectiveness, their capacity to detect and characterize harm is limited. We challenge assumptions that evidence certainty frameworks developed for effectiveness generalize to harm. We distinguish the role of randomization in estimating intended benefits versus unintended harm, for which confounding by indication is typically not the primary threat to validity. We argue key design elements including modest sample size, short follow-up, restrictive eligibility criteria, intention-to-treat analysis, and uninformative Bayesian priors, are likely to limit the trial's capacity to detect and characterize harms, especially those that are rare, delayed, developmentally contingent, or unexpected. We conclude that a comprehensive safety assessment requires integration of multiple complementary evidence sources, and should precede and inform judgements about clinical equipoise, risk-benefit balance, and PATHWAYS trial design.