Patient perspectives on physician competence: Validation of the CanMEDS framework.

Competency frameworks such as CanMEDS anchor medical training and accountability, yet most validations privilege educators and seldom test whether patients recognize and prioritize the same architecture. We conducted a cross-sectional online survey (February-June 2025) of Lebanese adults (N = 403) to validate, from the patient perspective, the seven CanMEDS roles and to examine sociodemographic moderators of endorsement and rank priorities. A 103-item instrument combined five-point endorsements and forced ranking. Psychometric evaluation used polychoric exploratory factor analysis and confirmatory factor analysis with diagonally weighted least squares; reliability was estimated with Cronbach's α and McDonald's ω. Robust linear regressions (HC4) modeled domain scores, and multinomial logistic regression analyzed rank priorities. Exploratory analysis supported seven factors explaining 68.6% of variance. Confirmatory analysis showed excellent fit with strong loadings and high internal consistency (α ≥ 0.90; ω ≥ 0.91). Men endorsed the Medical Expert role less than women (β=-2.96, p = 0.032). Higher family income showed graded positive associations with Medical Expert (e.g., > $3,000/month: + 8.66 points, p = 0.008). Lower educational attainment predicted lower priorities for Professionalism, Leadership, and Scholarship. Rural respondents prioritized Medical Expert, Communication, and Leadership more than urban peers, whereas physician age and gender were not significant predictors. Embedding patient-derived signals into competency-based medical education-through curricular emphasis, assessment weights, and multisource feedback-may strengthen social accountability and alignment with community expectations. Future work should test longitudinal stability, cross-cultural measurement invariance, and higher-order or bifactor models to parse shared variance among closely related roles.