Safety culture in Australian intensive care units: Establishing a baseline for quality improvement

Publication Type:
Journal Article
American Journal of Critical Care, 2013, 22 (2), pp. 93 - 102
Issue Date:
Filename Description Size
2012005156OK.pdf382.38 kB
Adobe PDF
Full metadata record
Background: Workplace safety culture is a crucial ingredient in patients' outcomes and is increasingly being explored as a guide for quality improvement efforts. Objectives: To establish a baseline understanding of the safety culture in Australian intensive care units. Methods: In a nationwide study of physicians and nurses in 10 Australian intensive care units, the Safety Attitudes Questionnaire intensive care unit version was used to measure safety culture. Descriptive statistics were used to summarize the mean scores for the 6 subscales of the questionnaire, and generalized-estimation-equations models were used to test the hypotheses that safety culture differed between physicians and nurses and between nurse leaders and bedside nurses. Results: A total of 672 responses (50.6% response rate) were received: 513 (76.3%) from nurses, 89 (13.2%) from physicians, and 70 (10.4%) from respondents who did not specify their professional group. Ratings were highest for teamwork climate and lowest for perceptions of hospital management and working conditions. Four subscales, job satisfaction, teamwork climate, safety climate, and working conditions, were rated significantly higher by physicians than by nurses. Two subscales, working conditions and perceptions of hospital management, were rated significantly lower by nurse leaders than by bedside nurses. Conclusions: Measuring the baseline safety culture of an intensive care unit allows leaders to implement targeted strategies to improve specific dimensions of safety culture. These strategies ultimately may improve the working conditions of staff and the care that patients receive. © 2013 American Association of Critical-Care Nurses.
Please use this identifier to cite or link to this item: