In the Journals: Obesity and Diabetes

Posted By Jan Blustein

Many of us believe that our neighborhood environment affects our health.  However, proving that this is so — and quantifying the environment-health link — is difficult.  That’s because we in some sense “choose” where we live.   Those who are able to live in healthy environments may have other advantages (such as healthier, better educated parents).  Those who live in less healthy neighborhoods may be otherwise disadvantaged.

An article in the October 20th issue of The New England Journal of Medicine tries to isolate the effect of neighborhood environment from neighborhood “choice.” Authored by economist Jens Ludwig and colleagues, the study draws on the experience of thousands of low-income families that participated in an experiment run by the Department of Housing and Urban Development (HUD) in the early 1990’s.  Called “Moving To Opportunity” [MTO], the experiment allowed families to enter a lottery in which some received housing vouchers that would allow them to move to low-poverty neighborhoods, while others did not receive vouchers.  The families have been followed by a set of research teams for more than a decade.

Ludwig and his colleagues looked at rates of obesity and diabetes – two conditions that are strongly linked with poverty and poor neighborhoods.  Not surprisingly, they found that families that had received vouchers lived in better-off neighborhoods.   However, they also found that adults in those families were somewhat healthier.  While there was no difference in obesity in by voucher status (BMI >= 30; 57.5% for the voucher group and 58.6% for the no voucher group), there was a significant difference in extreme obesity (BMI >= 40; 14.4% for the voucher group and 17.7% for the no voucher group), and diabetes rates (16.3% for the voucher group and 20.0% for the no voucher group).

While this tends to confirm our impression that neighborhood environment matters, it doesn’t tell us how environment affects health. Wealthier neighborhoods may offer healthier food options, contact with healthier peer role models, or less stress.  What is perhaps most striking about the experiment is the relatively small magnitude of the effects.  Rates of obesity and diabetes are high in this low-income sample, regardless of the opportunity afforded by the voucher. Health is a complex and cumulative outcome, and neighborhood is but one contributing factor.

Ludwig J, Sanbonmatsu L, Gennetian L, Adam E, Duncan GJ et al.  Neighborhoods, Obesity and Diabetes — a Randomized Social Experiment.  The New England Journal of Medicine.  2011:365:1509-19.

 Jan Blustein, Professor of Health Policy and Medicine at Wagner, teaches courses in statistics, program evaluation, and research methods. Her own research focuses on the dynamics underlying differences in health and health care among older Americans with chronic illnesses.   She can be reached at


Leave a Reply