UCSF

Serious quality of care problems in nursing homes has been a persistent problem in the United States for over 40 years. Staffing shortages and inadequate staff expertise were major contributors to many chronic and recurring quality problems.

The purpose of this study was to examine multidimensional associations among organizational characteristics, nurse staffing levels, and comprehensive aspects of quality of care considering resources (payer mix), residents (case mix), and market characteristics. The quality of care was assessed by nurse staffing levels, processes and outcomes of care, and facility deficiencies.

From various theories: Donabedian's structure-process-outcome model, organizational theory, and economic theory, a conceptual framework and key factors of nursing home and market characteristics were formulated with the support of previous literature.

This study used secondary data from Online Survey Certification and Reporting (OSCAR) data, Minimum Data Set (MDS) 2.0, quarterly staffing data from state inspections, and Area Resource File (ARF). The population of 195 nursing homes out of 199 Colorado freestanding homes in 2000 was used for data analyses. Depending on outcomes, various types of multivariate regression models were used.

The major findings of this study were: (1) quality of care in Colorado nursing homes was highly dependent on resources from the external environment, (2) Medicaid reimbursement rates and proportion of Medicare residents were important resources for nurse staffing, and (3) deficiencies in nursing home care were significantly higher in for-profit and chain-affiliated nursing homes. A useful conceptual framework was confirmed with strong support of theoretical background, literature review, and empirical findings.

Considering different staffing decisions, regulatory systems need to be designed carefully and systematically to improve quality of care. Given the findings, this study suggests that higher nurse staffing levels can improve quality of care in nursing homes.