Revised Hospital Survey on Patient Safety Culture (HSOPSC 2.0): cultural adaptation, validity and reliability of the Malay version

Abstract

Background Surveys on Patient Safety Culture™ Hospital Survey (HSOPSC) developed by the U.S. Agency for Healthcare Research and Quality (AHRQ) has been adopted worldwide. The Hospital Survey on Patient Safety Culture (HSOPSC) version 2.0 was released in 2019, but there have been no publications to date of its translation and validation for use in Malaysia. This study aimed to translate and cross-culturally adapt the revised HSOPSC 2.0 into the Malay language and determine its psychometric properties including the content, face, and construct validity, and reliability analyses. Methods This study was conducted from April – June 2023 and divided into three stages: translation and cultural adaptation; content and face validation; and construct validation using confirmatory factor analysis and reliability testing among 319 healthcare personnel from a public university hospital in Malaysia. Results The translated instrument demonstrated excellent content validity (I-CVI = 0.80 ~ 1.0, SCVI-average = 0.96) and face validity (I-FVI = 0.80 ~ 1.0, SFVI-average = 0.98). Reliability testing was acceptable (Cronbach’s α = 0.60 ~ 0.80) but indicated that reverse-coded items were poorly perceived. Confirmatory factor analysis showed a satisfactory model fit for the translated instrument (RMSEA = 0.08, GFI = 0.80, CFI = 0.80, and χ2/df = 2.96). Six items had very low factor loadings (< 0.40), with two constructs “Staffing and Work Pace” and “Response to Error” having AVE < 0.4, but acceptable CR ≥ 0.6. No items were removed from the questionnaire despite low factor loadings following a consensus from an expert panel. Conclusion The Malay version of the HSOPSC 2.0 containing ten domains and 32 items demonstrated satisfactory psychometric properties following expert consensus, with acceptable reliability and construct validity for measuring patient safety culture. Given factor loadings smaller than 0.40 in six items, broader validation is suggested to support the use of the translated instrument in the Malaysian healthcare setting. Supplementary Information The online version contains supplementary material available at 10.1186/s12913-024-11802-6.