Background: Mandatory hospital nurse staffing ratios are under consideration in a number of states without strong empirical evidence of the optimal ratio.
Objective: To determine whether increases in medical-surgical licensed nurse staffing levels are associated with improvements in patient outcomes for hospitals having different baseline staffing levels.
Research Design: Cross-sectional and fixed-effects regression analyses using a 1993–2001 panel of patient and hospital data from California. Splines define 4 staffing ratios.
Subjects: Adult acute myocardial infarction (AMI) (n = 348,720) and surgical failure to rescue (FTR) (n = 109,066) patients discharged between 1993 and 2001 from 343 California acute care general hospitals.
Measures: Patient outcomes are 30-day AMI mortality and surgical FTR; 4 baseline staffing levels—4 to 7 patients per licensed nurse [registered nurses (RN) and licensed vocational nurses (LVN)].
Results: Significant cross-sectional associations between higher nurse staffing and AMI mortality are reduced in the fixed-effects analyses. Improvements in outcomes were smaller in hospitals with higher baseline staffing: for each RN and RN + LVN increase, respectively, AMI mortality declined by 0.71 (P < 0.05) and by 2.75 percentage points for hospitals with more than 7 patients per nurse compared with 0.19 (P = NS) and 0.28 percentage points (P < 0.05) in hospitals with more than 4 patients per nurse. Significant cross-sectional associations between higher nurse staffing and FTR were not found in the fixed-effects analyses.
Conclusions: Strong diminishing returns to nurse staffing improvements and lack of significant evidence that staffing uniformly increases improve outcomes raise questions about the likely cost-effectiveness of implementing state-wide mandatory nurse staffing ratios.
From the *University of Pennsylvania School of Nursing, Philadelphia, Pennsylvania; †Department of Health Studies, University of Chicago, Chicago, Illinois; ‡Department of Medicine, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania; §Philadelphia Veterans Affairs Medical Center, Philadelphia, Pennsylvania; and ¶Department of Health Care Systems, the Wharton School, University of Pennsylvania, Philadelphia, Pennsylvania.
This study was funded in part by Doris Duke Charitable Foundation.
The funding source had no role in the design, analysis, or interpretation of the study or in the decision to submit the manuscript for publication.
Reprints: Julie Sochalski, PhD, University of Pennsylvania School of Nursing, Claire M. Fagin Hall, 418 Curie Blvd., Philadelphia, PA 19104. E-mail: firstname.lastname@example.org.