Psychometric Evaluations of the Hospital Survey on Patient Safety Culture Version 2.0 in Ethiopia (E-HSoPSC 2.0): A...

To evaluate the psychometric properties of the Hospital Survey on Patient Safety Culture (HSoPSC) version 2.0 in Ethiopian public hospitals.

A cross-sectional study.

Five public hospitals in Eastern Ethiopia.

Healthcare professionals (N=582).

An adapted and contextualised version of HSoPSC 2.0 was used to conduct structural validity using exploratory and confirmatory factor analyses (EFA and CFA). Convergent and discriminant validity were evaluated through item loadings and interfactor correlations, respectively. Reliability was measured using McDonald's omega and Cronbach's alpha.

CFA indicated a poor model fit for the original 10-factor, 32-item HSoPSC 2.0 across all statistical indices: relative chi-square (χ²/df=7.71), root mean square error of approximation (RMSEA=0.108), standardised root mean square residual (SRMR=0.088), comparative fit index (CFI=0.814) and Tucker-Lewis's index (TLI=0.780). Consequently, a comprehensive EFA was conducted, which identified a revised model comprising 5-factor, 21-item. This model accounted for 62.8% of the total variance and demonstrated strong construct validity, with excellent fit indices (χ²/df=3.67, RMSEA=0.068, SRMR=0.034, CFI=0.969, TLI=0.945). Internal consistency, assessed via McDonald's omega and Cronbach's alpha, exceeded the acceptable threshold of 0.70 across all dimensions, except for Response to Error (0.66). The convergent and discriminant validity of the new model was confirmed, ensuring an accurate representation of the underlying constructs.

The original HSoPSC 2.0 with 10-factor, 32-item failed to demonstrate structural validity in the Ethiopian healthcare context. In contrast, a revised 5-factor, 21-item model showed strong validity and acceptable reliability. This adapted version provides a culturally and contextually relevant tool for assessing patient safety culture in Ethiopian healthcare settings.