Results from multivariate regression analysis of the 12 safety culture dimensions of the Hospital Survey On Patient Safety Culture in 920* responses to a longitudinal follow-up study after a stepped wedge cluster RCT on surgical safety checklists, in Haukeland university Hospital, Bergen, Norway, 2009–2017
| HSOPSC safety culture dimensions | 2009 (n=330) | 2010 versus 2009 (n=139 and 137 vs 330) | 2017 versus 2009 (n=265 vs 330) | 2017 versus 2010 (n=265 vs 137) | ||||||||
| SSC | b0 | 95% CI | b1 | 95% CI | P value† | b2 | 95% CI | P value† | b2 | 95% CI | P value‡ | |
| Overall patient safety | No | 3.49 | (3.42 to 3.57) | 0.05 | (−0.09 to 0.18) | 0.470 | n.a. | – | – | n.a. | – | – |
| Yes | n.a. | – | 0.13 | (−0.01 to 0.26) | 0.067 | 0.09 | (−0.02 to 0.20) | 0.123 | −0.04 | (−0.18 to 0.10) | 0.574 | |
| Frequency of events | No | 2.80 | (2.72 to 2.88) | 0.01 | (−0.14 to 0.16) | 0.948 | n.a. | – | – | n.a. | – | – |
| Yes | n.a. | – | −0.02 | (−0.17 to 0.13) | 0.819 | 0.09 | (−0.03 to 0.22) | 0.134 | 0.12 | (−0.04 to 0.28) | 0.129 | |
| Unit managers support | No | 3.66 | (3.58 to 3.74) | −0.11 | (−0.26 to 0.04) | 0.153 | n.a. | – | – | n.a. | – | – |
| Yes | n.a. | – | 0.05 | (−0.10 to 0.20) | 0.538 | 0.21 | (0.09 to 0.33) | 0.001 | 0.15 | (−0.00 to 0.31) | 0.053 | |
| Continuous improvement | No | 3.33 | (3.26 to 3.40) | 0.03 | (−0.10 to 0.15) | 0.657 | n.a. | – | – | n.a. | – | – |
| Yes | n.a. | – | 0.19 | (0.06 to 0.31) | 0.003 | 0.19 | (0.09 to 0.30) | <0.001 | 0.03 | (−0.10 to 0.16) | 0.671 | |
| Team work in unit | No | 3.61 | (3.54 to 3.67) | 0.05 | (−0.16 to 0.07) | 0.440 | n.a. | – | – | n.a. | – | – |
| Yes | n.a. | – | 0.12 | (−0.00 to 0.24) | 0.052 | 0.19 | (0.10 to 0.29) | <0.001 | 0.08 | (−0.05 to 0.20) | 0.227 | |
| Open communication | No | 3.60 | (3.53 to 3.67) | 0.02 | (−0.11 to 0.15) | 0.764 | n.a. | – | – | n.a. | – | – |
| Yes | n.a. | – | 0.02 | (−0.11 to 0.15) | 0.745 | 0.09 | (−0.02 to 0.19) | 0.099 | 0.07 | (−0.07 to 0.20) | 0.327 | |
| Error feedback | No | 3.18 | (3.10 to 3.26) | −0.17 | (−0.31 to to 0.02) | 0.025 | n.a. | – | – | n.a. | – | – |
| Yes | n.a. | – | 0.05 | (−0.10 to 0.20) | 0.515 | 0.17 | (0.05 to 0.29) | 0.004 | 0.15 | (−0.01 to 0.30) | 0.058 | |
| Non-punitive | No | 3.82 | (3.75 to 3.88) | −0.09 | (−0.21 to 0.03) | 0.156 | n.a. | – | – | n.a. | – | – |
| Yes | n.a. | – | 0.09 | (−0.04 to 0.21) | 0.165 | 0.15 | (0.05 to 0.25) | 0.004 | 0.08 | (−0.05 to 0.20) | 0.243 | |
| Adequate staffing | No | 3.41 | (3.34 to 3.48) | −0.09 | (−0.22 to 0.04) | 0.180 | n.a. | – | – | n.a. | – | – |
| Yes | n.a. | – | 0.18 | (0.05 to 0.31) | 0.006 | 0.09 | (−0.01 to 0.20) | 0.084 | −0.09 | (−0.22 to 0.05) | 0.210 | |
| Hospital managers support | No | 2.82 | (2.74 to 2.91) | 0.16 | (0.01 to 0.30) | 0.039 | n.a. | – | – | n.a. | – | – |
| Yes | n.a. | – | 0.09 | (−0.06 to 0.24) | 0.237 | 0.33 | (0.21 to 0.45) | <0.001 | 0.23 | (0.08 to 0.39) | 0.003 | |
| Teamwork across units | No | 3.07 | (3.01 to 3.12) | 0.07 | (−0.03 to 0.18) | 0.151 | n.a. | – | – | n.a. | – | – |
| Yes | n.a. | – | −0.03 | (−0.13 to 0.07) | 0.552 | 0.13 | (0.05 to 0.22) | 0.002 | 0.16 | (0.06 to 0.27) | 0.002 | |
| Handoffs and transitions | No | 3.04 | (2.97 to 3.10) | 0.13 | (0.01 to 0.25) | 0.035 | n.a. | – | – | n.a. | – | – |
| Yes | n.a. | – | 0.03 | (−0.10 to 0.15) | 0.674 | 0.17 | (0.08 to 0.27) | 0.001 | 0.15 | (−0.04 to 0.24) | 0.170 | |
*The analysis is based on 871 cases with complete data for all dimensions and surveys.
†P values=from t-tests in the multivariate linear regression model. Multivariate test for all dimensions: F=3.45, df1=36, df2=2527, p< 0.001 (Wilk’s lambda, exact test).
‡Estimates from model with 2010 SSC as reference category.
b0, intercept, represents the estimated mean value of the dimension in 2009 (reference category); b1 and b2, estimated regression coefficients, represent average changes in dimension scores from 2009 to 2010 and 2017, respectively; HSOPSC, Hospital Survey on Patient Safety Culture; n.a., not applicable; RCT, randomised controlled trial; SSC, Surgical Safety Checklist.