Abstract
BACKGROUND: The Norwegian National Action Plan for Patient Safety and Quality Improvement recommends implementing electronic patient safety monitors (PSM) in Norwegian hospitals. We hypothesized that PSM implementation would reduce hospital length of stay, with secondary endpoints related to clinical and patient safety outcomes. METHODS: We performed a retrospective single-center, pragmatic, quasi-experimental, pseudo-randomized controlled trial assessing differences in endpoints for patients admitted before (control group: years 2019-2020) and after PSM implementation (intervention group: years 2021-2022) in two orthopedic bays in a Norwegian teaching hospital. To account for temporal changes unrelated to PSM, we assessed the same endpoints for 2019-2020 vs. 2021-2022 in two orthopedic bays that did not implement PSM. Data extraction from the local data warehouse was controlled according to internal standards, and we predefined all statistical analyses. RESULTS: The intervention group (n = 2786) and the control group (n = 2610) were evenly distributed for baseline characteristics, including age (median 73 [quartile 1-3 55-83] vs. 72 [55-83] years, p = 0.29), female sex (76.2% vs. 76.4%, p = 0.96), and Charlson Comorbidity Index (p = 0.24). Median hospital length of stay was 4.4 (2.1-7.7) days after implementation of PSM compared to 4.3 (2.0-7.2) days in the control group (p = 0.046 for difference between the groups). Nutritional screening and fall screening within 24 hours were lower in the intervention group compared to the control group, while medical reconciliation during the hospital stay was higher in the intervention group. We found no differences between the groups in 30-day or 1-year readmission or mortality rates. Hospital length of stay was similar in 2021-2022 vs. 2019-2020 for the bays that did not implement PSM. CONCLUSION: The implementation of PSMs in this study did not improve clinical or patient safety outcomes. Hospital length of stay was statistically longer after PSM implementation, although the difference was not clinically relevant.