Adaption and validation of the Safety Attitudes Questionnaire for the Danish hospital setting

Abstract

Purpose: Measuring and developing a safe culture in health care is a focus point in creating highly reliable organizations being successful in avoiding patient safety incidents where these could normally be expected. Questionnaires can be used to capture a snapshot of an employee's perceptions of patient safety culture. A commonly used instrument to measure safety climate is the Safety Attitudes Questionnaire (SAQ). The purpose of this study was to adapt the SAQ for use in Danish hospitals, assess its construct validity and reliability, and present benchmark data.

Materials and methods: The SAQ was translated and adapted for the Danish setting (SAQ-DK). The SAQ-DK was distributed to 1,263 staff members from 31 in- and outpatient units (clinical areas) across five somatic and one psychiatric hospitals through meeting administration, hand delivery, and mailing. Construct validity and reliability were tested in a cross-sectional study. Goodness-of-fit indices from confirmatory factor analysis were reported along with inter-item correlations, Cronbach's alpha (α), and item and subscale scores.

Results: Participation was 73.2% (N=925) of invited health care workers. Goodness-of-fit indices from the confirmatory factor analysis showed: c(2)=1496.76, P<0.001, CFI 0.901, RMSEA (90% CI) 0.053 (0.050-0056), Probability RMSEA (p close)=0.057. Inter-scale correlations between the factors showed moderate-to-high correlations. The scale stress recognition had significant negative correlations with each of the other scales. Questionnaire reliability was high, (α=0.89), and scale reliability ranged from α=0.70 to α=0.86 for the six scales. Proportions of participants with a positive attitude to each of the six SAQ scales did not differ between the somatic and psychiatric health care staff. Substantial variability at the unit level in all six scale mean scores was found within the somatic and the psychiatric samples.

Conclusion: SAQ-DK showed good construct validity and internal consistency reliability. SAQ-DK is potentially a useful tool for evaluating perceptions of patient safety culture in Danish hospitals.

Keywords: Denmark; patient safety culture; questionnaire; reliability; validity.

Figure 1

Distribution of percent of positive scores (%-positive) per dimension for the 31 clinical areas.^a Notes: ^aAll clinical areas are ranked in ascending order according to %-positive for each dimension, and each clinical area was allocated a letter, and this letter was used in the graphical display for all dimensions, signalizing the position of each clinical areas with each dimension. The pale gray bar represents the average %-positive of all clinical areas. Results in this figure were generated by the use of IBM SPSS version 21.0 (IBM Corporation, Armonk, NY, USA) and graphically displayed by Microsoft Excel 2010.