Bending Health Disparity Curves

By David A. Kindig, MD, PhD

I recently raised the issue of disparities and noted the many unacceptable health differentials that exist across  U.S. subpopulations. It is very easy to claim that such disparities should be eliminated, but seldom do we set specific quantitative targets for such improvement.  

These days, the phrase “bending the curve” usually applies to reducing rates of health care cost increases. But look below at two other curves -- disparities in US mortality between Blacks and Whites as well as males and females from 1979 to 2007.

The figures show improvement (i.e., declining mortality) for all four groups, but the improvements are occurring at different rates (e.g., average annual rates of improvement over the period are 1.13% for Whites, 1.19% for Blacks, 1.36% for males, and 0.90% for females). The reasons for the differences are interesting and important, but not my point today. It might seem that if these trends were to continue, reduction or elimination of these disparities would happen “naturally” without additional intervention. And of course the disparity depends on both curves: if White or female rates were to improve even faster it would take even higher rates of Black or male improvement to narrow or eliminate the differences. We cannot  assume that these disparities will continue to narrow over time nor should we ignore the ethical and pragmatic considerations of inaction or delay.  

How rapidly should disparities be reduced? Clearly, there is no “right” answer to this question but I believe this is an important issue for communities, states, and nations to discuss explicitly and then act by adopting programs and policies help them achieve their goals. In theory, focusing available effort and resources on less healthy groups will help narrow the gaps and is particularly important in cases where disparities are increasing. For example, the gap in health between more vs. less educated people appears to be getting bigger over time instead of narrowing.

Of course the biggest challenge is what to do to improve health for the less healthy groups in order to bend the disparity curve. Here in Wisconsin, we compiled the What Works for Health database summarizing evidence on what works to improve health. But, despite our best intentions, we were unable to locate as much evidence on what works to reduce disparities. Furthermore, some programs and policies that improve overall health may actually worsen disparities. For example, media campaigns to promote smoking cessation may have the unintended effect of increasing disparities by socioeconomic status. 

So, our collective challenge is to a) figure out how much of our resources we want to direct toward reducing disparities and b) find the most cost-effective ways to use those resources to narrow these gaps. Our population health research agenda must prioritize understanding what the most cost-effective disparity reduction investments are so that they may be put into practice.

David A. Kindig, MD, PhD, is Emeritus Professor of Population Health Sciences and Emeritus Vice-Chancellor for Health Sciences at the University of Wisconsin School of Medicine and Public Health.

*****

For further reading:

Koh HK, Oppenheimer SC, Massin-Short SB, Emmons KM, Geller AC, Viswanath K. (2010). Translating Research Evidence Into Practice to Reduce Health Disparities: A Social Determinants Approach. American Journal of Public Health, 100(S1), S72-S80.

Read JG, Gorman BK. (2011). Gender and health revisited. In BA Pescosolido, JK Martin, JD McLeod, A Rogers (eds.), Handbook of the Sociology of Health, Illness, and Healing: A Blueprint for the 21st Century, Part 5, 411-429.

 

Comments

[Craig Mosbaek]

I agree that what we need to do is improve health for less healthy groups. And it is true that many public health policies and programs actually increase disparities because these policies/programs are more effective in healthier populations. But, we can do better.

We should look at each policy/program and see if it is increasing or decreasing health equity. Where it is decreasing health equity, we should modify the policy/program to address the disparity.

You mention that media programs to promote cessation are often less effective with low SES populations. How can we do better?

First, cessation advertisements can be placed where the ads reach low SES populations. This is fairly easy with radio and TV where there are good data on the demographics of viewers and listeners.

Second, cessation advertisements can be better designed to be effective in low SES populations. We should look at the language, the settings, the actors, etc. used in these ads. Of course, the ad concepts and final ads can be tested with various audiences.

On the policy front, let’s look at smokefree workplace laws. In the past, many smokefree workplace laws exempted bars and restaurants. This meant that these laws were not protecting some low-income workers and certain racial/ethnic groups. Smokefree workplace laws help support people who are trying to quit smoking. And, there are data to show that low-income workers quit at the same rate as higher-income workers when their workplace goes smokefree. Removing the exemption for bars and restaurants helps increase health equity.

I believe we should look at every policy and program with an eye toward decreasing disparities. This focus on solutions can help bend the curve and lead us towards our goal of eliminating health disparities.