Publication:

The registered footprint of health-professions education studies across the WHO International Clinical Trials Registry Platform: a cross-registry meta-research study

Background Trial and study registries are an underused meta-research lens on health-professions education (HPE), but keyword retrieval cannot reliably isolate learner-focused studies because education and training terms are dominated by patient-education and clinical-trial vocabulary. We mapped the registered HPE footprint across the WHO International Clinical Trials Registry Platform (ICTRP), testing whether classification rather than filtering recovers the learner-focused signal.

Methods From a frozen ICTRP cross-registry export (10 June 2026), 16 narrow plus four broad education terms yielded 1,390 records after three-tier de-duplication. Because keyword precision was low (narrow-term interventional positive predictive value 0.182, 95% CI 0.130-0.248) and no term subset reached 0.85, we applied a four-value blind protocol to all 1,383 classifiable records, double-coding a random 200-record reliability sample.

Results Classification identified 109 registered HPE studies (7.9%, 95% CI 6.6-9.4%): 60 interventional and 49 observational (HPE-versus-rest Cohen κ 0.906). The footprint was small and pandemic-concentrated: 97 of 109 (89%) were registered in 2020-2023, peaking in 2021, and 99 of 109 (90.8%) were COVID-themed, across 11 registries and 29 countries (India most represented). Of 60 trials, 36 (60%) were randomized (median planned enrolment 100). Prospective registration was documented for 51 of 108 (47.2%) and a results record for only 18 of 109 (16.5%).

Conclusions Registries reveal a small, pandemic-era-concentrated HPE footprint that keyword search misses; recovering it requires manual, double-coded classification of the candidate net. Low retrievability and sparse registry-visible results signal opportunities to strengthen prospective registration and results posting. Because this is a retrieved rather than exhaustive footprint, the counts carry a recall caveat.