Examining Health Utilization in Later-Life Using Count Models

Andy Sharma, University of North Carolina at Chapel Hill

This paper examines health disparities between Black/White near-elderly individuals using the Medical Expenditure Panel Survey (MEPS). The analysis is conducted at three levels: (1) descriptive statistics showing disparities for outcomes such as doctor visits, outpatient provider visits, and dental visits, (2) geographical analysis for the above three outcomes based on census regions, (3) and three negative binomial count models predicting visits after controlling for age, race, education, income, region, marital status, insurance, health, and attitudes regarding the need for insurance and the need for medical help. Results indicate disparities continue to exist for elderly Blacks. I conduct this analysis for 2003, 2004, and 2005 pooled cross sectional datasets, as well as the 2004-2005 merged panel dataset. The panel data allows one to use fixed/random effects due to over dispersion. This research compels further investigation and I conclude by offering explanations related to 1) social environment and 2) health care access.