Eight Americas - new perspectives on health disparities

Christopher Murray, Majid Ezzati and Sandeep Kulkarni, University of Harvard

[Project number 20076]

Our study examines racial, geographic, and socioeconomic health disparities within the United States. In describing these inequalities, we identify 8 subgroups within the US population that capture the gamut of health experiences. While there is a tremendous wealth of literature concerning disparities in the US, these studies focus largely on race or income alone. Our approach instead uses a combination of several characters to define our subgroups, which reveal differences across and within racial and socioeconomic groups. While much of our work until now has been focused on the US, we are planning to extend our work to the UK social classes in understanding international health disparities.

The study currently has 3 major objectives:

1) Computation of trends in life expectancies - using vital registration and censual data, we estimate trends in life expectancies over the last 20 years for each of these 8 subgroups. We also estimate cause-deleted life expectancies to examine relative impacts of different causes of death upon overall life expectancy.

2) Analysis of death by cause - using information on underlying cause of death, we are calculating cumulative probabilities of death for each cause. This analysis allows an examination of the unique patterns of disease burden across subgroups.

3) Assessment of risk factor exposure and health system coverage - we examine prevalence of selected risk factors for each of our 8 subgroups. These risk factors include smoking, alcohol use, and body mass index among others. In terms of access to care, we are also looking differences in healthcare coverage and utilization. The data for these analyses come from the Behavioral Risk Factor Surveillance System, a nationally conducted survey of US households.

We primarily intend to use the LS data to compute life expectancies by sex and class. By applying the same life table algorithms upon the LS data, we hope to compare the differences across our subgroups against the differences between the British social classes. We are particularly interested in the LS data because disparities in the UK have been well documented and can provide an international frame of reference for health disparities. Furthermore, we plan to look at not only the magnitude but also the trend in disparities to assess whether the UK has been more or less successful in reducing disparities vis-à-vis the US.

Top of page