Abstract
BACKGROUND: Measuring patient safety culture should be a fundamental component in the functioning of long-term care facilities. Several tools have been developed for this purpose, including the Nursing Home Survey on Patient Safety Culture (NHSOPSC), which was created and validated in the USA. The psychometric properties of the NHSOPSC have also been tested in some countries, including Norway and Spain. Such an analysis has not been carried out in Poland so far. This study seeks to establish the measurement properties of the Polish version of the NHSOPSC in a sample of long-term care facility staff. METHODS: In the initial phase, the questionnaire was translated into Polish. This was followed by a cross-sectional study of medical, auxiliary, and administrative staff in inpatient long-term care facilities. Additionally, a second measurement was carried out in select establishments. Descriptive statistics were calculated and the distributions of each NHSOPSC item were checked, followed by confirmatory factor analysis (CFA) by Diagonally Weighted Least Squares (DWLS) method to confirm the 12-factor model. The tool's reliability was assessed by calculating Cronbach's Alpha and interclass correlation coefficient (ICC). Correlation analyses were also conducted. The statistical analyses were performed using the JASP 0.18.3.0 and IBM SPSS Statistics 29. RESULTS: The analyses substantiated the presence of a 12-factor model. However, one item (A16) was eliminated to enhance reliability. As expressed by Cronbach's alpha coefficient, the tool's reliability is moderate (α = 0.60-0.82). Most dimensions demonstrate satisfactory ICC levels, although slightly lower values (below 0.6) were observed for three domains (staffing, feedback, and communication about incidents, and organizational learning). The correlation analyses indicate acceptable validity of the tool. CONCLUSIONS: After removing one item, the Polish version of the NHSOPSC can be used to measure the patient safety culture in long-term care facilities.