The relationship between staffing and adverse events in Washington State hospitals: a cross-sectional study using linked hospital data

Abstract

Abstract Objective: To quantify the relationship between staffing characteristics and patient outcomes in acute care hospitals in Washington state. Methods: Retrospective cross-sectional time-series study of linked data from six sources on staffing and outcomes for Washington state hospitals. Key stakeholders provided input on data sources, measures, and outcomes in a four-phase participatory process. After data cleaning and linkage, we used a random effects Poisson regression model to examine the relationship between staffing levels or characteristics and adverse outcomes. Results: The study included 263 hospital-years from 80 distinct hospitals, with 162 hospital-years from general acute care hospitals (n=46) and 101 hospital-years from critical access hospitals (n=34). In general acute care hospitals, a higher ratio of patients to care team staff is associated with a higher number of adverse events (adjusted RR, 1.36 per one SD increase; 95% UI 1.13–1.63), and a lower proportion of RNs on the care team staff is likely associated with a higher number of adverse events (adjusted RR, 1.16 per one SD increase; 95% UI, 0.97–1.39). In critical access hospitals, a lower proportion of RNs on the care team is associated with a higher number of adverse events (adjusted RR, 3.28 per one SD increase; 95% UI, 1.20–7.75). A counterfactual analysis indicated that if all general acute care hospitals had no more than the median staffing ratio of 1.2 patient hours per staff hour, the number of adverse events would be reduced by 10% (95% UI 2.7–16.8). Conclusion: RN staffing is an indisputable component of safe, high quality patient care, and other factors such as availability of care team staff, hospital features, and patient characteristics also impact patient outcomes. This study highlights the utility of merging diverse data sources to provide a comprehensive analysis of the relationships between staffing and patient outcomes.