The relationship between nursing skill mix, nurse sensitive patient outcomes and patient satisfaction

Abstract

<p><strong>Literature Review</strong>: Hospitals are the largest components of the United States healthcare system, and are encountering pressure to provide better quality care and reduce cost. The Patient Protection and Affordable Care Act (ACA) directed the Centers for Medicare and Medicaid Services (CMS) to initiate a system to financially incentivize healthcare quality. Nurses are the largest component of the hospital workforce and have the most patient contact hours. Due to the expansion of the Hospital Value Based Program (HVBP) in 2014 to include the nursing-sensitive value-based program (NSVBP), nurses have the highest potential to alter patient outcomes and hospital costs. The impact of nursing skill mix on patient outcomes has been examined, but current literature suggests the need to correlate skill mix with nursing-sensitive indicators and patient satisfaction over longer periods of time. Previous studies in smaller, rural hospital settings have not been reported to our knowledge.</p> <p><strong>Aim</strong>: The goal of this study is to investigate the association between nurse staffing mix, patient safety outcomes, and patient satisfaction in the context of changes in nurse staffing ratios over time at a smaller rural hospital.&nbsp;&nbsp;&nbsp;</p> <p><strong>Methods</strong>: UPMC Horizon converted LPN positions to RN positions on three of its general medical units, providing an opportunity for a natural two-group study design, and resulting in savings to the hospital in fewer total staff hours. We analyzed incident reporting and medical record coding on six nursing sensitive indicators, and eleven domains of the patient satisfaction obtained from hospital consumer assessment of healthcare providers and systems (HCAHP) from July 2011 to December 2015 and then from July 2011 to June 2015.</p> <p><strong>Statistical analysis</strong>. Descriptive statistics were computed and Independent Samples t Tests were used to evaluate the impact of the results at six and 12 month intervals. We hypothesized no difference in mean patient satisfaction scores and adverse patient outcome scores before and after the implementation date at 95% significance level with alpha equal to 5% and a critical value of t at 1.67.</p> <p><strong>Results:</strong> The trend data indicated significant change in a number of patient outcomes and patient satisfaction ratings at six months and at 12 months.</p> <p>Overall rating of hospital (OVERALL) had significantly improved at six months but not at 12 months. Communication with nurses (NURSES) did not improve at six months but had not significantly improved at 12 months. Nurses treat you with courtesy/respect (COURTESY) improved significantly at six months and at 12 months. Nurses listen carefully to you (LISTEN) improved significantly at six and at 12 months. Nurses explain in a way you understand (EXPLAIN) did not show a significant improvement either at six or twelve months. Pain management (PAIN) declined but the drop was not significant. Pain well controlled (CONTROLLED) dropped significantly both at six and 12 months. Staff does everything to help with pain (HELP) dropped at six and at 12 months but the drop was not significant. Communication about medicines (MEDICINE) improved significantly at six and 12 months. Staff tell you what a new medicine is for (TELL YOU) improved significantly at six and at 12 months. Staff describes medicine side effects (SIDE EFFECTS) did not improve at six months but did improve significantly at 12 months. Effect sizes for these results were moderate in the two to three percent range. Medication errors (MED ERROR) increased slightly and significantly at six months but at one year there was no significant change. Patient scanning (SCAN_PATIENT) declined slightly but significantly at six months but at one year there was no change. Fall rates (FALLS), hospital acquired pressure ulcers (HAPU) and medication scanning (SCAN_MED) did not change significantly at six or at 12 months. Failure to detect a difference may be due to limited observations.</p> <p><strong>Conclusions</strong>: The change in nurse staffing did not significantly impact patient outcomes.&nbsp; In a number of areas the change produced a moderate and significant improvement in patient satisfaction with the exception of pain management, which declined slightly but significantly. It may be possible to change nurse-staffing ratios to reduce costs while maintaining outcomes and improving patient satisfaction. Future research should focus on the relationship between factors such as bachelor&rsquo;s level training and continued policy and procedure education, work environment, and patient to nurse ratios.</p>