Developing and validating patient safety sessions in Japanese medical education: a modified Delphi study

This study aimed to (1) develop and validate educationally sound and feasible model session plans for patient safety learning in Japanese undergraduate medical education and (2) identify barriers to their implementation in educational settings.

A convergent mixed-methods design was embedded within a modified Delphi study. Four session plans based on Merrill's First Principles of Instruction were evaluated by 45 patient safety specialists and faculty purposively sampled from Japanese national university hospitals. Participants assessed plans using a 14-item Context-Input-Process-Product checklist. Items were rated on a four-point scale; consensus was defined as a mean score ≥ 3.5 and a standard deviation < 1.0. Simultaneously, open-ended comments on implementation challenges were analyzed using directed content analysis based on Steinert's framework for educational barriers. Quantitative and qualitative data were integrated using a joint display to derive meta-inferences.

All four sessions met consensus criteria in the first round (n = 36; response rate = 80%). Topics covered human error, incident reporting, root cause analysis, and conflict management. Analysis identified 13 subthemes across five domains-teacher, student, knowledge, attitude, and system. Key barriers included limited faculty facilitation experience, insufficient linkage to safety practice, and institutional constraints such as lack of formal educational roles for safety specialists.

The validated sessions demonstrated strong feasibility. However, content readiness alone does not guarantee successful adoption. Addressing identified barriers through targeted faculty development focusing on facilitation skills and organizational alignment is necessary to achieve sustainable implementation of patient safety education in undergraduate programs.