Maughan MP, Yu J, Jung HY
Objectives: To analyze patient and facility characteristics associated with smaller versus larger nursing home (NH) chains.
Design: This study used a cross-sectional study design.
Setting and participants: NHs affiliated with multi-facility chains in the United States.
Methods: Using nationally representative data from LTCFocus 2021 Facility-Level File and the June 2023 Centers for Medicare and Medicaid Services (CMS) Affiliated Entity Performance Measures, we assessed differences in facility and patient characteristics among categories of NH chains size based on the number of certified beds (smallest to largest quintile of NH chain size) using one-way analysis of variance testing. We conducted linear regression analyses to examine the association between the quintile of chain size and staffing outcomes.
Results: Among the 9348 NHs associated with 610 chains in our sample, the smallest quintiles of NH chains had the lowest percentage of for-profit facilities, a higher percentage of patients with Alzheimer's and dementia-related diseases, and higher percentages of long-stay residents needing assistance with activities of daily living (ADLs). The largest chain quintile was associated with fewer staffing hours per resident day (HPRD) for all total nurse staff -0.69 (95% CI, -0.86 to -0.52; P < .001), registered nurses (-0.25 HPRD; 95% CI, -0.32 to -0.17; P < .001), certified nursing assistants (CNAs) (-0.37 HPRD; 95% CI, -0.48 to -0.26; P < .001), and weekend total nurse staff (-0.57 HPRD; 95% CI, -0.71 to -0.42; P < .001]). The CMS staffing rating was also lower in larger NH chains (-0.73 for quintile 5; 95% CI, -0.98 to -0.48; P < .001).
Conclusions and implications: Larger NH chains tended to treat less clinically complex patients and were associated with lower staffing ratios. Given increased attention and stricter rules regarding staffing by government agencies, increased monitoring of staffing in NHs affiliated with large chains by policymakers, antitrust agencies, and regulators is warranted.