Objectives: The Safety Attitude Questionnaire (SAQ) has been used in several countries to measure healthcare workers' attitudes toward patient safety. It has not been validated in Taiwan, however. The objective of this study was to develop and validate a Chinese version of the SAQ. Methods: The original SAQ was translated into Chinese in October 2006. There were 30 core items, measuring six domains of an organization's safety culture: Teamwork Climate, Safety Climate, Job Satisfaction, Stress Recognition, Perception of Management, and Working Conditions. A hospital-wide safety attitude survey was conducted in one medical center in Taipei in April 2007. The reliability and validity of the SAQ were examined by multilevel confirmatory factor analysis. Results: The overall response rate was 71.1%. The internal consistency of each domain was high (composite reliability: 0.77-0.93). The explanatory abilities of two-thirds (21/30) of the core items are considered important for the corresponding domains (squared multiple correlation [SMC]≥0.5). The standardized factor loadings (λ) for most core items were statistically meaningful (λ=0.75-0.85), but the two reverse items were not (λ<0.5). The model fit was satisfactory for the six domains and for the safety culture via structural equation models. Except for Stress Recognition (λ=-0.09, SMC=0.01), all domains were highly correlated to one another (correlation coefficients: 0.67-0.93) and had good factoral loadings (λ=0.69-0.88) and SMC (0.48-0.78) to the hospital safety culture. Conclusions: The Chinese version of the SAQ is reliable and valid in measuring frontline workers' safety attitudes in this healthcare organization in Taiwan.
Objectives: The Safety Attitude Questionnaire (SAQ) has been used in several countries to measure healthcare workers' attitudes toward patient safety. It has not been validated in Taiwan, however. The objective of this study was to develop and validate a Chinese version of the SAQ. Methods: The original SAQ was translated into Chinese in October 2006. There were 30 core items, measuring six domains of an organization's safety culture: Teamwork Climate, Safety Climate, Job Satisfaction, Stress Recognition, Perception of Management, and Working Conditions. A hospital-wide safety attitude survey was conducted in one medical center in Taipei in April 2007. The reliability and validity of the SAQ were examined by multilevel confirmatory factor analysis. Results: The overall response rate was 71.1%. The internal consistency of each domain was high (composite reliability: 0.77-0.93). The explanatory abilities of two-thirds (21/30) of the core items are considered important for the corresponding domains (squared multiple correlation [SMC]≥0.5). The standardized factor loadings (λ) for most core items were statistically meaningful (λ=0.75-0.85), but the two reverse items were not (λ<0.5). The model fit was satisfactory for the six domains and for the safety culture via structural equation models. Except for Stress Recognition (λ=-0.09, SMC=0.01), all domains were highly correlated to one another (correlation coefficients: 0.67-0.93) and had good factoral loadings (λ=0.69-0.88) and SMC (0.48-0.78) to the hospital safety culture. Conclusions: The Chinese version of the SAQ is reliable and valid in measuring frontline workers' safety attitudes in this healthcare organization in Taiwan.