Validity of the patient safety climate items from the RN4CAST survey compared with the SAQ : Internal consistency, concurrent validity and

Abstract

Background: The Norwegian Knowledge Centre for the Health Services participated in the multinational nursing survey RN4CAST as the Norwegian representatives. The items relating to nurses` perceptions of patient safety climate have not been assessed in the Norwegian setting, which is needed if these items are to be used with certainty regarding their validity and reliability. Objectives: The main objective of this thesis is to assess the concurrent validity and internal consistency of items relating to perceptions of patient safety climate in the Norwegian part of the multinational nursing survey RN4CAST. A secondary objective is to identify the main challenges associated with identifying causal relationships between safety climate and safety outcomes, and discuss these in relation to the survey undertaken in this study. Methods: Nurses from two hospital wards completed both the SAQ and the items relating to patient safety climate from the RN4CAST survey. Two scaled indexes were created from the RN4CAST items, and three analyses were done: 1) Pearson`s correlation coefficient between the SAQ and the RN4CAST; 2) Chronbach`s Alpha for internal consistency within the RN4CAST items; and 3) Bland-Altman analysis for assessment of agreement. Results: The assessment of the two scales created demonstrated high internal consistency (Chronbach’s Alpha of .82 and .80, respectively), moderate inter-item correlations, and good concurrent validity as measured by Pearson’s correlation coefficient (r= .73 and .78, respectively). The results from the Bland-Altman analysis demonstrated reasonably good agreement between the two measures of patient safety climate. Conclusion: The result of the analyses demonstrates that the two scales are internally consistent, there is an association between the scales from RN4CAST and SAQ safety climate dimension, and that the level of agreement between the two methods is good. The two scales are both found to be validated measures of patient safety climate.