Abstract:
Background:
In previous studies, patient safety culture has often been considered separately from occupational safety culture. In addition, there are few studies that examine both kinds of safety culture in the context of working conditions. The aim of the dissertation was to illustrate the perspectives of nurses and physicians at two German university hospitals on patient safety culture, occupational safety culture and working conditions within four publications. Publications 1-3 were based on survey data from nurses and physicians (n=995), which were gathered in the year 2015 as part of the WorkSafeMed study at two German university hospitals.
Methods:
In study 1, the perceptions of nurses and physicians on working conditions, patient safety climate and occupational safety climate were assessed and examined with regard to occupational group differences. These differences were determined using a t-test for independent samples. In addition to the p-value (significance), the effect size was also calculated to evaluate the relevance of the results. Study 2 examined the impact of working conditions, patient and occupational safety culture on perceived patient and occupational safety culture from the perspective of physicians and nurses, and whether there are shared predictors for both kinds of safety culture. Regression models for patient safety culture and occupational safety culture were developed and tested. In study 3, results of the survey on working conditions of nurses and physicians from the WorkSafeMed study were compared with the respective reference data (hospital nurses, hospital physicians, general population) from the COPSOQ database (period 2012-2017). For the comparison, the G-COPSOQ scales from the WorkSafeMed study (G-COPSOQ II) were converted to the G-COPSOQ III scales and tested in several statistical analyses. For the comparison with reference data from the COPSOQ database, a one-way analysis of variance (ANOVA) was performed and additionally the effect size was calculated as a measure of relevance. In study 4, an integrative literature review was carried out in addition with the aim of creating an overview of determinants of an occupational safety culture for the hospital setting. In addition to the hospital setting, industrial work areas were integrated to enable a comparison of different settings. A systematic literature search was conducted in four databases in March 2019 considering the PRISMA statement. The search was updated again in April 2020.
Results:
In study 1, a statistically significant difference with a strong effect size dCohen>.50 was found with regard to the following scales. With respect to patient safety climate, physicians gave a more positive rating than nurses for “staffing” (2.8 vs. 2.4), “management support for patient safety” (3.0 vs. 2.6), and “overall perception of patient safety” (3.3 vs. 2.9). There were less relevant differences with regard to occupational safety climate; nurses rated the index “subjective assessment of occupational safety measures initiated by the employer, related to own safety” more positively than physicians (1.7 vs. 2.0). Nurses rated the following working conditions worse than physicians: “Degree of freedom at work” (36.0 vs. 46.2), “possibilities for development” (71.6 vs. 79.6) and “workplace commitment” (48.4 vs. 61.3). In addition, nurses reported poorer scores for “patient-related burnout” (36.5 vs. 28.0) and “job satisfaction” (67.5 vs. 73.4) compared to physicians.
In study 2, “management support for patient safety” (ß=0.24, p≤.001), “staffing” (ß=0.21, p≤.001) and “supervisor support for patient safety” (ß=0.18, p≤.001) were identified by nurses and physicians as significant predictors of perceived patient safety culture. Important predictors for perceived occupational safety culture were “job satisfaction” (ß=0.26, p≤.001), “work-privacy conflict” (ß=-0.19, p≤.001) and “patient-related burnout” (ß=-0.20, p≤.001). The patient safety culture model achieved a high model fit of R2=0.64, while the occupational safety culture model revealed a more moderate model fit of R2=0.27. “Job satisfaction” and leadership (via the two variables “management support for patient safety” and “supervisor support for patient safety”) were identified as overarching shared predictors in both models.
In study 3, the following converted G-COPSOQ scales showed a statistically significant difference with a strong effect size dCohen>.50. Nurses from the WorkSafeMed sample rated “social relations” poorer than hospital nurses from the COPSOQ database (39.5 vs. 52.9), but reported a higher “job satisfaction” (66.7 vs. 57.8). Physicians in the WorkSafeMed sample also indicated higher “job satisfaction” compared to hospital physicians in the COPSOQ database (72.7 vs. 62.4).
In study 4, 44 studies were included in the integrative literature review. The studies in the hospital sector and in other sectors were classified into seven clusters using a theoretical framework. Determinants of an occupational safety culture in the hospital sector were less represented in the studies compared to other sectors.
Discussion:
Overall, it was found that nurses rated patient safety climate and working conditions worse than physicians. For both occupational groups, predictors were identified which are important for perceived patient and occupational safety culture. The comparison with reference data enabled a more in-depth and subsequent classification of the results on working conditions in the sense of a benchmark. Furthermore, it became clear that previous studies have not yet comprehensively presented determinants as influencing factors of an occupational safety culture for the hospital sector. Further studies are needed to capture all facets of an occupational safety culture in hospitals. The comprehensive assessment of an occupational safety culture in hospitals can presumably provide implications on how patient and occupational safety culture in hospitals can be jointly developed.