At age 65, a person in the 75th income percentile can expect to live about 5 years longer than a person in the 25th percentile, holding factors like education and reported health constant. Why? Some explanations center on income-based differences in access to health care; lower income households may not see a physician until late in the disease process, and when treated are less likely to receive aggressive medical intervention. Others focus on economic factors; stress related to previous employment, lengthy spells of unemployment, or declining living standards at retirement. A third strand suggests that fundamental differences in personality cause some people to experience both low income and poor health; in this view, income is not causal to but simply correlated with early death. The goal of the research program is to distinguish among these competing (or complementary) hypotheses. To understand the biomedical issues involved, a two-year program of training in epidemiology, biostatistics, Medicare claims data research, medical decision making, cardiology and oncology is proposed. These courses, along with seminars at Dartmouth Medical School and the National Bureau of Economic Research, will provide a long-term basis for health economics research. The second part of the research will focus on individual elements of these hypotheses. Proposed studies in the medical area include documenting income-based differences in health care use, and measuring whether surgical intervention following AMI for high versus low income groups has systematically different effects on outcomes. There will be an emphasis on geographical-based differences in medical care intensity and practice; for areas with high levels of age-adjusted medical utilization, do more resources flow to higher income groups, and if so, how do health outcomes differ by income group? Economic research topics include: testing for whether households save adequately for retirement; if they do not, the extent to which such myopia affects retirement health outcomes. Finally, individual behavioral factors such as smoking or alcohol consumption are crucial to understanding how individual factors might simultaneously affect income, wealth, and health.
| Skinner, Jonathan; Weinstein, James N; Sporer, Scott M et al. (2003) Racial, ethnic, and geographic disparities in rates of knee arthroplasty among Medicare patients. N Engl J Med 349:1350-9 |
