Heterogeneous Effects of Nursing Staff Wages on Turnover: Evidence from U.S. Nursing Homes Using an Instrumental Variable Approach

Background and Objectives Turnover among nursing staff [registered nurses (RNs), licensed practical nurses (LPNs), and certified nursing assistants (CNAs)] is a long-standing challenge in nursing homes with significant implications for quality of care. This study aimed to examine the relationship between nursing staff wages and turnover. Research Design and Methods We used national data from 2021 to 2023, linking multiple datasets including the Payroll-Based Journal and Medicare Cost Reports (n = 37,254). Turnover was modeled separately for each nursing staff type as a fractional outcome bounded between 0 and 1. The primary predictor variable was facility-level average hourly wage for each staff type. To address potential endogeneity and reverse causality, we used a two-stage residual inclusion instrumental variables approach with county-level average wages (excluding the index facility) as the instrument. Models controlled for facility-level organizational and county-level market characteristics, with fixed effects for county and year and cluster-robust standard errors at the facility level. Results Higher CNA wages were significantly associated with lower turnover (β = −0.35, 95% CI [−0.59, −0.11], p = 0.005), indicating a 0.35 percentage point reduction in turnover per $1 increase in hourly wage. Wages were not significantly associated with turnover among licensed nurses. Discussion and Implications Findings suggest that wage increases may be most effective for CNAs, while retaining licensed nurses likely requires complementary organizational and non-monetary strategies. While nursing homes should strive to offer competitive wages to their staff, targeted reimbursement reforms may be necessary to overcome financial constraints.