Commentary Examines Cross-Domain Interactions and Error Control in Musculoskeletal Platform Trials

A commentary in The Journal of Bone and Joint Surgery examines design and analysis challenges in multi-domain platform trials for musculoskeletal disorders, responding to an article by O’Hara and colleagues. These trials combine separate domains under a shared master protocol and infrastructure, with domain-specific protocols and the ability to add domains over time. The commentary reports no new patient outcomes or sample-size data.

The first concern is that participants receiving interventions across several domains may experience treatment interactions that dilute or inflate observed effects. The author calls for empirical analyses after enrollment and analysis of the initial domains to assess whether adjustment for these interactions is needed and how they affect statistical power and type-I error.

The second concern is whether overall type-I error—the risk of false-positive findings—should be controlled across domains. The commentary questions differing expectations for multi-domain trials and single-domain, multi-arm platforms, particularly when participants are shared across domains and independence is lost. Citing prior methodological work, it notes that strict familywise error control can increase sample-size requirements in single-domain platforms and proposes evaluating the corresponding implications for multi-domain trials. These are methodological questions rather than evidence supporting any particular musculoskeletal intervention.