Rationing of Nursing Care and Its Relationship with Nurse Staffing and Patient Outcomes: The Mediation Effect Tested by Structural Equation Modeling.

: The purpose of this study is to test the mediation effect of rationing of nursing care (RONC) and the relationship this has between nurse staffing and patient outcomes. : The analytic sample included 7802 nurse surveys and 5430 patient surveys. Three patient outcome indicators, nurse staffing, RONC, and confounding factors were considered in the model pathways. : The hypothesized model was shown to be statistically significant. In the model, nurses who were in the units with lower nurse-to-patient ratios reported higher scores on RONC, which meant that an increased level of withheld nursing care or a failure to carry out nursing duties was apparent. Nurses who reported a higher score on RONC, scored poorly on the quality assessment and were more frequently involved in patient adverse events. Nurse staffing influenced quality assessments and patient adverse events through RONC. In units with poorer nurse-reported quality assessments or more frequently patient adverse events, patient-reported dissatisfaction scores were higher. : The results suggest that a lack of nurse staffing leads to RONC, which leads to poorer patient outcomes. These results are seen when considering the evaluations completed by both nurses and patients. The relationship between staffing numbers and patient outcomes explains the mediating role of RONC.