Non communicable diseases (NCDs), mainly cardiovascular disease, cancers, diabetes and chronic lung disease, are responsible for roughly two out of every three deaths today. Other NCDs, including mental illness, accidents, renal disease and dental disease, kill or maim millions more. The World Health Organization (WHO) is now putting the final touches on a draft Action Plan NCDs, as well as a list of global targets and indicators on NCDs. These are part of the follow up to the landmark UN High Level Meeting on NCDs, in September 2011.
One of the topics that arises regularly in this context is the effect that social determinants of health – poverty, education, living and work conditions among them – have on the development of chronic illness.
Despite the high-level meeting and the acknowledgment by the WHO’s world conference in October 2011 of the role these factors play in the NCD epidemic, the current draft of the WHO targets document contains no direct measurement of any social determinants. This is important because “what gets measured gets done,” and without a commitment to measuring social determinants that cause illnesses, it is less likely that concrete action will happen.
Why is it that leaders so far have failed to place appropriate emphasis on measuring social determinants of health in connection with NCDs? On the surface, the case seems clear for action in this arena, since it seems intuitively obvious to many that social determinants dramatically affect health outcomes overall, and must have a significant impact on NCDs. The Public Health Institute, as a science-based organization, fully supports the idea that social determinants of health have important impacts on most illnesses, including NCDs. But the issue is complex.
In 2011, Stanford Medical School scientist David Rehkopf and University of California researcher Nancy Adler collaborated on a report for the Public Health Institute analyzing more than a hundred research studies on the connection between social determinants and health. Overall, they found significant evidence that socio-economic status, education level and neighborhood significantly impact health. However, just below the surface lie some very complicated and counter-intuitive findings. These don’t cast doubt on the basic idea, but collectively they do imply that a great deal more research is necessary in defining appropriate solutions.
One example they found: the overwhelming financial advantage to being born in the United States does not necessarily convey a protective benefit from a health point of view. In fact, foreign-born immigrants to the United States have lower rates of disease than those born in the United States, except for stomach cancer and liver disease. One study found that during 1999-2001, immigrants lived 3.4 years longer on average than native-born US residents, which was up from 2.3 years longer in a similar study 20 years earlier. And the gap was even larger among certain groups – immigrant black populations and Latinos. The differences cannot be fully explained by the new immigrants’ healthier lifestyles, their having better health than those who remain in sending countries, or their decisions to leave the US if they become ill.
One of the stronger findings in the field of social determinants is that people who are better educated tend to live longer and healthier lives. But Rehkopf and Adler point out that proving that more education improves health is difficult, in part because education moves in a single direction; it is not possible for individuals to “uneducate” themselves, and so studies that would assess whether changes in “exposure” to education cause changes in health outcomes are impossible. It would also be immoral to randomly “select” who gets a high school or college education, and so “spurious correlations” can occur. Rehkopf and Adler did find one study showing an increase in the years of school was associated with greater longevity, but this was a lonely example. In short, while the evidence points toward more education equaling better health, a great deal more work needs to be done to prove this is true.
The effect of living in a better neighborhood is yet another example from the Rehkopf and Adler meta-analysis of how difficult it is to pin down the role of social determinants in health. It is well documented that neighborhoods in the United States with generally wealthier populations and more expensive homes have healthier residents. But is this because the neighborhoods themselves are healthier, or because healthier people tend to live there? Rehkopf and Adler looked at an experiment in which residents of New York, Boston, Chicago, Baltimore and Los Angeles were offered a choice of moving to a “low poverty neighborhood,” moving to any neighborhood of their choice, or appearing on a waiting list for future housing vouchers. So far, over three years, the experiment seems to show positive health changes associated with moving from a poor neighborhood to a low-poverty neighborhood. But the changes are not simple. For example, a much more positive effect on the health seems to be occurring for children than for adults. And teen boys actually experienced a negative effect – an increase in smoking cigarettes. Why is this, and what results should those trying to frame NCD prevention policies draw? It is difficult to know.
Perhaps the framers of the WHO NCD targets document are aware of these complexities, and for that reason have decided to leave them out of the global accounting. However, experts in the field of social determinants of health would argue that, while there is no doubt of the overall effect that social determinants have on health, more research is absolutely necessary. No one thinks that interventions lifting people out of poverty and poor education need await a “perfect” study proving causation. But exactly what those interventions should be, and how they can best be targeted for the greatest potential positive health outcomes, requires additional public health scrutiny and measurement, including in the NCD targets to be adopted by WHO in 2013.