The goal of this research project is to assess the health status of African Americans in nursing homes. Eight percent of the United States elderly population (65 plus) is African American. The proportion of elderly blacks is expected to increase with gains in both life expectancy and birth rates. This growth will result in an increased demand for long-term care. However, it is unknown how patterns of poorer health among younger blacks than younger whites carry over into old age. This study goes beyond simple black vs. white and male vs. female type comparisons and uses sociodemographic variables such as age, gender, income, and race as central rather than peripheral components in analyzing health status. These sociodemographic variables will be used to describe the differences in prevalence of specified health status outcomes. In addition, logistic regression models will be fit using the sociodemographic variables to explain health status outcomes. The statistical models will be tested for interaction effects and model fit (Hosmer-Lemeshow). This study examines approximately 6 million Minimum Data Set (MDS) [nursing home] assessments from up to 13 states that are housed in the University of Michigan Assessment Archive Project (UMAAP) database. Health status will be measured by various outcomes such as: functional status, cognitive status, resource utilization, chronic conditions (cerebrovascular accident, arteriosclerotic heart disease, congestive heart failure, Alzheimer's, hypertension, diabetes), falls, decubitus ulcers, mood and behavior problems, psychosocial well-being, continence, and advanced directives. This research addresses criticisms leveled at previous studies regarding small sample sizes and statistical power - the smallest sample size of African American nursing home residents in 1995 is 4,303. This research has important policy, practice and theoretical implications for the future management and care provided in nursing homes.
