Back in the 1990s, the federal government tried an unusual social experiment: It offered thousands of poor women in big-city public housing a chance to live in more affluent neighborhoods.
A decade later, the women who relocated had lower rates of diabetes and extreme obesity -- differences that are being hailed as compelling evidence that where you live can determine your health.
The experiment was initially aimed at researching whether moving impoverished families to more prosperous areas could improve employment or schooling. But according to a study released Wednesday, the most interesting effect may have been on the women's physical condition.
About 16 percent of the women who moved had diabetes, compared with about 20 percent of women who stayed in public housing. About 14 percent of those who moved were extremely obese, compared with nearly 18 percent of the other women.
The small-but-significant differences offered some of the strongest support yet for the idea that where you live can significantly affect your overall health, especially if your home is in a low-income area with few safe places to exercise, limited food options and meager medical services.
The research, led by Jens Ludwig, a University of Chicago professor of public policy, was published in Wednesday's New England Journal of Medicine.
The experiment started as a $70 million HUD project in Baltimore, Boston, Chicago, Los Angeles and New York. It morphed into a health study after a variety of other government agencies and private foundations pitched in with an additional $17 million.
The study involved women living in public housing in neighborhoods where 40 percent or more of residents were poor -- areas like many of those on the South Side of Chicago or in the Bronx in New York City. The women all had children and were considered heads of households.
From 1994 to 1998, nearly 1,800 of them were offered vouchers to subsidize private housing, but the vouchers were only good in higher-income neighborhoods where fewer than 10 percent of the people were considered poor. They were required to live there at least a year.
The rest of the women were divided into two groups. One group got vouchers they could use in any neighborhood. The other women did not receive vouchers, with the expectation that they would stay put.
The women who moved to richer areas had the lowest rates of extreme obesity and diabetes. The difference suggests that moving to a better neighborhood could help at least 1 in 25 women. Or, in other terms, a person's risk of diabetes or extreme obesity dropped by about 20 percent by moving to a higher-income neighborhood.