Originally posted 4/14/2011
In my March 14 post Bending Health Disparity Curves, I focused exclusively on differences in mortality rates, such as deaths per 100,000 persons. Rates are very useful measures, because they allow comparison across populations of different sizes. But from a population health perspective, rates alone are not enough, because large disparities in very small populations have a different impact than similar disparities in larger populations. Burden refers to the impact of a health problem in a population, combining both the rate and the number of people affected.
Although our disparities focus tends to be on race and ethnicity, disparities also exist in other domains such as geography, socioeconomic status, and gender. The table below shows a surprisingly high male mortality rate, but it is the size of this population (146 million) that transforms the rate into a significant population health burden.
Mortality rate per 100,000* Population Size
Black 1009 39 million
White 780 240 million
Male 945 146 million
Female 672 150 million
*Average rate 2003-2007. Centers for Disease Control and Prevention, National Center for Health Statistics. Compressed Mortality File 1999-2007. CDC WONDER On-line Database, compiled from Compressed Mortality File 1999-2007 Series 20 No. 2M, 2010.
The table above reveals disparities related to race and gender that are far more complex than I can do justice to in this brief post. However, this issue of rate vs. burden applies across disparity domains. In Wisconsin, for example, there is a similarly large mortality gap in education. Mortality rates among the 44% of the working-age population with high school or less education are significantly higher than rates among college graduates.
This does not mean that smaller populations with large rates should be ignored. As Keppel and colleagues point out, “rates among small groups, such as the Asian and American Indian or Alaska Native populations, will seldom be high enough to warrant population-specific interventions based on reduction in total burden alone. An independent commitment to the goal to eliminate disparities would be required to warrant intervention with small racial and ethnic groups.”
However, in a resource limited world, choices will have to be made. As Keppel et al again point out, “sizable reductions in both disparity and total burden can result when the largest group has the worst rate and effective interventions are targeted to that group.” We need to engage in robust discussion about priorities for overall outcomes versus disparity reduction, and then get quickly to identifying resources to achieve these ends. Attention to both rates and burden will be required to make the best decisions in such a process.
P.S. Feel free to comment about issues around rates versus burden, the appropriate balance between improving overall health and reducing inequities, whether you think male mortality rates are disparities or inequities (see below), or about anything else in the blog as well.
Population Health 