On Income Inequality and Population Health

Orin Kerr, in response to a comment of mine on Concurring Opinions, asks:

Just to be clear, when you refer to the correlation between income inequality and health, you mean that societies with low inequality tend to have better physical health than societies with high inequality, right? If so, is the idea that they correlate with something else, or that inequality determines physical health? (And if the latter, how?)

As is my wont, I began to ramble, so I figured it would be better to hog this blog space rather than the CoOp comments:

You ask a number of pertinent questions, and I hope it's clear that there remains a great deal about the correlation that has yet to be unpacked.  But the basic idea is that, as Daniels, Kennedy, and Kawachi put in the title of their 1999 essay (turned into a 2000 book), "justice is good for your health."  Greater distributive justice -- lower socioeconomic disparities -- is not just good for the health of the less affluent, but has a demonstrable effect in improving population health even for those who sit on top of the social gradient.

Perhaps the most interesting aspect of this gradient, most directly attributable to Marmot's work, is that it is not simply the case that the wealthy are healthier than the poor.  No one doubts this.  But an impressively linear relationship tends to exist across the social gradient.  Thus the really well-off are healthier than the slightly less well-off, who in turn are healthier than the slightly slightly less well-off, and so on and so forth.

This relationship has been demonstrated in local, state/provinicial, national, and global levels, in widely different societies and cultures.  There are a number of scholars working on explaining the impressive consistency of these social gradients, which gets to your second question -- that inequality is a prime determinant of health.

Really, this idea is not altogether novel.  As I note here, Ramazzini observed the relationship between socieconomic conditions and health as long ago as 1713, and it was a prime motivation for the sanitarians' public health reforms of the late 19th-early 20th century.

But as for the specific mechanism, there are a number of theories.  In my view -- and I want to be careful to avoid privileging reductionist, scientistic explanations here -- some of the most interesting and compelling accounts tie poor health outcomes to inequality in context of the neuroendocrine system.  Basically, it turns out we're not supposed to have the fight-or-flight response turned on all the time.  Where we do -- b/c of socioeconomic conditions, for example -- we maintain very high levels of stress hormones like cortisol.  There is excellent evidence that sustained high levels of cortisol causes all manner of health problems, and is linked to just about every major chronic disease (except for many cancers) you'd care to name (CAD, CVD, diabetes, mental health problems, to name but a few).

It's particularly harmful to developing brains, which suggests a molecular reason for the evidence that really early childhood development is absolutely crucial to flourishing across the lifespan.  (i.e., lower socioeconomic status in childhood seems to produce -- and the causation here is incredibly complex and admittedly poorly understood -- what some refer to as a "cascade" of health problems, each building on the other to increase morbidity and lower life expectancy, along with a number of other poor health measures).

I guess one of your implicit questions is how wider socioeconomic disparities causes poor health even for those on top of the social gradient.  This is a really complex question, and I'm not sure there are a lot of really developed answers out there -- though I am no epidemiologist or sociologist, both of whom tend to dominate the technical work in these fields -- but one example is that even African-Americans who sit atop the social gradient by any measure (education, income, neighborhood, you name it), suffer more disease and have lower life expectancy than Caucasians in the same position on the gradient.

Nancy Krieger, among others, has advanced a theory, with (IMO) credible data that racism in and of itself, after controlling for virtually every conceivable confounder, seems to have an inimical effect on health.  Thinking about this in terms of the neuroendocrine system mentioned above may help process this idea, which, on a purely anecdotal basis, I've seen seems to generate widespread disbelief (though I'm not really sure why -- if it isn't hard to understand why systematic discrimination and prejudice might significantly contribute to higher levels of depression and anxiety, why it is to hard to believe that consistently higher levels of such problems contribute significantly to poor health outcomes over the lifespan?).

Finally, Richard Wilkinson, among others, has produced a large body of evidence showing that societies with significant income inequalities report much higher levels of violence, and that the relationship is proportional.  Also, such violence occurs within positions on the socioeconomic gradient, not just between levels.  (So domestic violence -- presumably between families and communities of similar SES is significantly higher in societies with high income inequality, in addition to what we might expect, that violence between social positions is more common in such societies).

Anyway, none of this is to suggest that all of this has been figured out and is widely understood.  As I tried to indicate in my initial comment, that so much of the relationship remains somewhat cloudy is at least partially why the correlation remains an object of such intense study across so many different disciplines.  (For example, though I'm no methodologist, I do a lot of ethics and public health policy work, and bringing this evidence to bear and thinking about the moral implications of this evidence is vitally important, IMO).  But I also don't want to suggest we are totally in the dark.  The relationship exists.  It is robust, it is persistent, and we have some interesting and evidence-based accounts of how income inequalities produce lower population health.  It's very difficult, in my view, to claim that wide income inequalities are productive of "good" population health.

Of course, none of this exhausts the normative inquiries.  What constitutes good health? How much inequality should we tolerate? How much should we encourage? If not all inequality is bad-- and I agree that it is not -- where do we draw the line?

These are obviously all viable questions.  But it's odd to me that so many seem either to reject the mountains of data suggesting, by virtually any health indicia you might care to use, the U.S. has very poor health relative to its health expenditures, or to express puzzlement at the reasons why this might be so.  This is the basic subject of Kawachi and Kennedy's book on the health of nations.

Hope this helps.

Comments

[Patrick S. O'Donnell]

Perhaps it goes without saying, but this is essential information and I'm glad you took the time to post about it.

I've put together here a short (hence manageable) list that should help persuade others of the salience of equality/inequality issues when it comes to achieving public health goals:

Anand, Sudhir, Fabienne Peter, and Amartya Sen, eds. Public Health, Ethics, and Equity (New York: Oxford University Press, 2004).

Daniels, Norman. Just Health: Meeting Health Needs Fairly (Cambridge, UK: Cambridge University Press, 2008).

Davis, Marion, Carolyn Clancy and Larry R. Churchill, eds. Ethical Dimensions of Health Policy (New York: Oxford University Press, 2002).

Farmer, Paul. Infections and Inequalities: The Modern Plagues (Berkeley, CA: University of California Press, 1999).

Farmer, Paul. Pathologies of Power: Health, Human Rights, and the New War on the Poor (Berkeley, CA: University of California Press, 2003).

Henderson, Gail E., et al., eds. The Social Medicine Reader, 3 Vols. (Durham, NC: Duke University Press, 2005).

Marmot, Michael and Richard G. Wilkinson, eds. Social Determinants of Health (New York: Oxford University Press, 2006 ed.).

Smith, Richard, et al., eds. Global Public Goods for Health (New York: Oxford University Press, 2003).

[mcd7m]

Hello! I am a second year medical student at the University of Virginia. I am doing a research project this summer on blogs maintained by medical students and physicians and I have really enjoyed reading yours! I wanted to let you know about an online journal at UVA called Hospital Drive, which can be found at http://hospitaldrive.med.virginia.edu/. I think you and your readers would really enjoy reading and perhaps even submitting material to the journal. Happy Reading!