Objectives: To examine the association between in-hospital mortality and four nurse staffing variables-the ratio of total nursing staff to patients, the proportion of RNs to total nursing staff, the mean years of RN experience, and the percentage of nurses with bachelor of science in nursing degrees.
Background: Studies suggest that nurse staffing changes affect patient and organizational outcomes, but the impact of nurse staffing on patient outcomes has not been studied sufficiently and the results of the previous studies are equivocal. Additionally, the studies of the relationship between nurse staffing and patient outcomes or the impact of nurse staffing on patient outcomes had not been previously examined in Thailand.
Methods: A retrospective, cross-sectional, observational research design was employed to study the research questions. Data of 2531 patients admitted to seven medical units and 10 surgical units of a 2300-bed university hospital in Thailand was used. All data of patients admitted to this hospital with four common groups of principal diagnoses (diseases of the heart, malignant neoplasms [cancer of all forms], hypertension and cerebrovascular diseases, and pneumonia and other diseases of the lung) was extracted from patient charts and discharge summaries in the calendar year 1999. Nurse staffing variables for each nursing unit in 1999 came from nursing service department databases. Multivariate logistic regression was used to determine the relationship between nurse staffing variables and in-hospital mortality.
Results: The findings of this study revealed that the ratio of total nurse staffing to patients was significantly related to in-hospital mortality in both partial and marginal analyses, controlling for patient characteristics. In addition, the ratio of total nursing staff to patients was found to be the best predictor of in-hospital mortality among the four nurse staffing variables, controlling for patient characteristics. The study did not find any significant relationship between in-hospital mortality and three nurse staffing variables (the proportion of RNs to total nursing staff, the mean years of RN experience, and the percentage of bachelor degree prepared nurses) probably due to the low variation of these variables across nursing units or because they may have correlated with other variables.
Conclusions: The findings of this study add to our understanding of the importance of nurse staffing and its relationship to the patient outcome of hospital mortality. Further, the findings also provide information for hospital and nursing administrators to use when restructuring the clinical workforce, revising hospital policies, or making contractual decisions on behalf of nursing and public beneficiaries.