Psychometric properties of the Indonesian version of the nursing home survey on patient safety culture

INTRODUCTION: The development of resident safety culture in nursing homes (NH) represents a major challenge for governments and NH owners, with a requirement for suitable tools to assess safety culture. Indonesia currently lacks suitable safety cultures scales for NH.

OBJECTIVES: To evaluate the psychometric properties of the translated Indonesian version of the Nursing Home Survey on Patient Safety Culture (NHSOPSC-INA).

METHODS: This study was a cross-sectional survey conducted using NHSOPSC-INA. A total of 258 participants from 20 NH in Indonesia were engaged. Participants included NH managers, caregivers, administrative staff, nurses and support staff with at least junior high school education. The SPSS 23.0 was used for descriptive data analysis and internal consistency (Cronbach's alpha) estimation. The AMOS (version 22) was used to perform confirmatory factor analysis (CFA) on the questionnaire's dimensional structure.

RESULTS: The NHSOPSC CFA test originally had 12 dimensions with 42 items and was modified to eight dimensions with 26 items in the Indonesian version. The deleted dimensions were 'Staffing' (4 items), 'Compliance with procedure' (3 items), 'Training and skills' (3 items), 'non-punitive response to mistakes' (4 items) and 'Organisational learning' (2 items). The subsequent analysis revealed an accepted model with 26 NHSOPSC-INA items (root mean square error of approximation = 0.091, comparative fit index = 0.815, Tucker-Lewis index = 0.793, CMIN = 798.488, df = 291, CMIN/Df = 2.74, GFI = 0.782, AGFI = 0.737, p