Abstract
OBJECTIVES: A recent investigation between staffing patterns in nursing homes (NH) and quality outcomes has not occurred. Our objective was to examine whether staffing levels and utilization of agency staff are associated with injurious falls in NHs.
DESIGN: Cross-sectional study.
SETTING AND PARTICIPANTS: A total of 11,183 NHs with 1,115,304 long-stay residents enrolled in Medicare fee-for-service were included (July 1, 2019-December 31, 2019).
METHODS: Staffing data and agency utilization by discipline (eg, Certified Nursing Assistants, CNAs) were obtained from Payroll Based Journal reports. Staff minutes were averaged per discipline, and agency staff percentage was calculated weekly as contract hours divided by total hours worked, per discipline. Injurious falls among long-stay residents during the 24-week study period were ascertained through Medicare Provider Analysis and Review claims. Poisson regression models were used to estimate rate of injurious falls with staff levels and agency staffing, respectively, adjusting for NH and patient-level characteristics. Models were stratified according to whether the NH had low certified nursing assistant (CNA) coverage (<2.45 hours per resident per day) vs recommended coverage (≥2.45 h/resident/d) according to the 2024 Centers for Medicare & Medicaid Service Minimum Staffing Standards for Long-Term Care Facilities.
RESULTS: Of the 11,183 NHs analyzed, 3320 (29.7%) met the proposed CNA coverage, whereas 7863 (70.3%) did not. Among NHs that achieved the recommended CNA coverage, increased CNA [incidence rate ratio (IRR), 0.84; 95% CI, 0.74-0.96] and registered nurse (IRR, 0.61; 95% CI, 0.52-0.72) staffing hours were associated with a decrease in injurious falls. In NHs that did not achieve the recommended CNA coverage, increased CNA staffing was associated with an increase in injurious falls (IRR, 1.20; 95% CI, 1.09-1.31). No association was found between agency staff use and injurious falls, except for low registered nurse agency use, defined as >0% but ≤10% of agency staff use, which was modestly associated with more falls in NHs that achieved the suggested CNA coverage (IRR, 1.15; 95% CI, 1.03-1.29).
CONCLUSIONS AND IMPLICATIONS: Although increased CNA staff levels were associated with fewer injurious falls in NHs that were well staffed, increased CNA levels were associated with an increase in injurious fall rates in low-staffed facilities. Utilization of agency staff had little impact on injurious falls. Efforts to improve quality outcomes in NHs through increasing staff hours may have a variable effect depending on an NH's baseline resource availability.