• Roybal Center for Health Policy Simulation
• About the Center
• Projects
• Research Products
• Center Staff

RAND Roybal Center for Health Policy Simulation

Nursing Home Workforce Dynamics and Quality of Care

Objectives

This pilot examines the causes and consequences of variation in worker turnover patterns among nursing homes. The specific aims are:

1. Examine a set of turnover measures that distinguish among the turnover of short, medium and long tenure workers and estimate a parametric distribution that fits the observed patterns.
2. Analyze the causes of turnover among short, medium and long tenure workers: investigate the effect of state policies and programs intended to reduce turnover, controlling for labor supply factors such as unskilled labor force growth, for labor demand factors such as elder population growth and for facility factors such as resident characteristics and staffing levels.
3. Investigate the consequences of turnover by workers of various tenures on the quality of resident care.
4. Simulate the impact of various state policies and programs on worker turnover distributions and on quality of care.

Progress to Date

We have collected two years of quarterly data on staffing factors (turnover, stability, agency use and staffing levels) as well as quality outcomes from Nursing Home Compare. We have used this data to write three papers and submit an RO1 to the National Institute on Aging.

We find that previous work that has focused on staffing levels and, to some extent, turnover, has missed important dimensions of the relationship between staffing factors and quality of care. We find that greater use of nurse aides from temporary agencies reduces quality but greater use of registered nurses from temporary agencies improves quality. We also find important interactions among staffing factors. For example, greater agency use of NAs is most detrimental when there is few long tenure NAs (i.e. lower stability).

The addition of new staffing factors to the analysis adds important insights but also adds greatly to the complexity of modeling and communicating the relationships. In particular, estimating non-linear relationships and interactions for the effect of the many staffing factors on quality requires theory-driven judgments about which non-linearities and interactions to incorporate.

We also have developed new measures for summarizing the impact of four staffing measures and graphical methods for communicating these results. Both of these innovations involve simulating quality outcomes from the results of non-linear models.

Grant Proposal

Castle, N.G. & Engberg, J. Staffing Characteristics of Nursing Homes and Quality. Funded by the Agency for Healthcare Research and Quality: R01 HS016808-01.

Engberg, J. & Castle, N.G. A Simulation Model of Nursing Home Staffing Stability and Quality. National Institute on Aging. First submission: 2/5/07

Papers

Castle, N.G.; Engberg, J. The influence of staffing factors on quality of care in nursing homes. Health Services Research, 2007. (in print).

Castle, N.G.; Engberg, J. Turnover and quality in nursing homes. Inquiry, 2006. (in revision).

Castle, N.G.; Engberg, J. Agency staffing and quality of care in nursing homes. Journals of Gerontology: Social Science, 2006. (in revision).

Next Steps

We are sending out a third wave of the survey to obtain an additional year of staffing information from nursing homes. We have are augmenting the general local labor market measures available in ARF with measures of skill-specific local labor market tightness. We are completing our longitudinal database of state-level policies and programs that aim to have an impact on nursing home staffing. When complete, these data will be used to fully address the specific aims.