Cross-cultural adaptation and validation of Hospital Survey on Patient Safety Culture in Bengali version (B-HSOPSC 2.0) among nurses in Bangladesh: a cross-sectional study

Background The Hospital Survey on Patient Safety Culture (HSOPSC 2.0) was developed and updated by the Agency for Healthcare Research and Quality in 2019. It has now been widely adopted and translated into different languages worldwide. However, the validity and reliability of the Bengali version of HSOPSC 2.0 (B-HSOPSC 2.0) have not been tested among healthcare professionals. This study aimed to determine the validity and reliability of the B-HSOPSC 2.0 with cross-cultural adaptations, among hospital nurses in Bangladesh. Methods The study was conducted among nurses in eight tertiary-level government medical college hospitals in Bangladesh. A two-step study design was employed, encompassing the translation, cultural adaptation, and psychometric evaluation of B-HSOPSC 2.0. The translation process included forward and backward translation by panel, expert consensus, review, and pretesting. Content validity, reliability, and test-retest reliability were assessed using the content validity index, Cronbach’s alpha, and the interclass correlation coefficient, respectively. Construct validity was evaluated through confirmatory factor analysis (CFA), and convergent validity was examined using average variance extraction and Spearman's correlation coefficient. Results Out of 7,170 eligible nurses, 4,982 completed questionnaires (response rate: 69.5%), and a subset (n = 424) provided retest responses. The words “manager” and “clinical leader” were removed, and the word “units” was replaced with “wards,” as they were deemed inappropriate for the Bangladeshi healthcare system. The content validity index provides strong evidence of effective measures of the instruments (I-CVI = 0.83-1.00, S-CVI = 0.98). B-HSOPSC 2.0 demonstrated a good internal consistency (Cronbach’s α = 0.70–0.76), and test-retest reliability (ICC = 0.65–0.76) showed acceptable reliability. In the CFA model, the indices for the 10 and 9 dimensions were CFI = 0.79, 0.83; NFI = 0.78, 0.82; TLI = 0.74, 0.79; GFI = 0.91, 0.92; RMSEA = 0.06, 0.06; SRMR = 0.07, 0.06, respectively. Conclusions The psychometric properties of the Bengali version showed acceptable reliability and validity characteristics. Findings suggest that it can be used to measure nurses' perceived patient safety culture in hospital settings in Bangladesh. This validated tool will enhance safety culture and benefit nursing practice. Further studies are required, as the psychometric properties of B-HSOPSC 2.0 among other healthcare professionals in Bangladesh remain to be confirmed. Clinical trial number Not applicable.