Patient safety climate profiles across time: Strength and level of safety climate associated with a quality improvement program in Switzerland-A cross-sectional survey study.

Safety Climate has been acknowledged as an unspecific factor influencing patient safety. However, studies rarely provide in-depth analysis of climate data. As a helpful approach, the concept of "climate strength" has been proposed. In the present study we tested the hypotheses that even if safety climate remains stable on mean-level across time, differences might be evident in strength or shape. The data of two hospitals participating in a large national quality improvement program were analysed for differences in climate profiles at two measurement occasions. We analysed differences on mean-level, differences in percent problematic response, agreement within groups, and frequency histograms in two large hospitals in Switzerland at two measurement occasions (2013 and 2015) applying the Safety Climate Survey. In total, survey responses of 1193 individuals were included in the analyses. Overall, small but significant differences on mean-level of safety climate emerged for some subgroups. Also, although agreement was strong at both time-points within groups, tendencies of divergence or consensus were present in both hospitals. Depending on subgroup and analyses chosen, differences were more or less pronounced. The present study illustrated that taking several measures into account and describing safety climate from different perspectives is necessary in order to fully understand differences and trends within groups and to develop interventions addressing the needs of different groups more precisely.