News Ticker

Needles in the Obesity Haystack

For some time, various special-interest groups have sought to outlaw expansion of the suburban lifestyle preferred by most Americans, pejoratively called “urban sprawl.” Operating under the self-congratulatory label of “smart growth,” the movement has been strongest in urban areas, such as the San Francisco Bay Area, Portland, and Denver, where anti-growth measures have been adopted.

But the movement has run aground on its own faulty research. For example, it no longer decries the loss of valuable farmland to urbanization, a claim put to rest by President Bill Clinton’s Department of Agriculture. Farmland has surrendered to improved productivity, not urban development. Smart-growth advocates have also have been on the defensive because they didn’t understand how rationing raises prices: Their land-rationing policies have raised the price of housing and denied home ownership as an option to low-income households.

In the wake of these failures, smart-growth advocates opened a new front. University of Maryland researchers, working with anti-suburban lobbyists at Smart Growth America, published research blaming suburban lifestyles for America’s obesity crisis. This research, too, was flawed:

* The increase in obesity occurred two to three decades after the period of greatest suburbanization.

* The researchers included four counties in New York City that are more dense than any other of the more than 400 counties in the study, and more than 10 times as dense as the average American or Canadian urban area. Such “outliers” are properly excluded from statistical analyses because they tend to skew the results in unreliable ways.

* Although past research has shown obesity tends to increase as household income falls, the income data available in the researchers’ dataset was excluded from the an alysis.

Even with these deficiencies, the suburban to city differences in obesity were trivial. The researchers’ model predicted, for example, barely a two-pound difference between average weights in the city of San Francisco and suburban Marin County. And this was not the typical detached academic study. The press conferences and political briefings resembled peddling medicine out of wagons in the western frontier.

Now there is a new installment. Professor Lawrence Frank of the

University of British Columbia (UBC) led researchers on a study of neighborhood obesity in the Atlanta metropolitan area. The results indicated people who drive more (that is, people who live in more suburban settings) tend to be more obese.

Again there was a marketing campaign not typical of dispassionate academic research. There were press conferences and an impressive spread in a special Time magazine issue on obesity. And also again, there are serious flaws in the research.

While Frank and his team managed to collect individual data on weight and income, they never apparently asked about eating habits. There is good reason to have included diet in the study, since caloric intake has increased in recent decades.

Approximately 6 percent of the study participants drove an average of more than five hours per day. By any standard, this is substantial. The Atlanta drivers were behind the wheel for more than five hours at a rate more than four times the U.S. average as indicated by the Nationwide

Household and Transportation Survey. The UBC study cannot predict the relationship between driving and obesity for any of the more than 10,000 people involved, let alone for the rest of the U.S.

Indeed, one of the study’s co-authors has gone out of his way to disclaim the principal thesis of the marketing campaign, that suburban lifestyles cause obesity.

The University of Maryland and University of British Columbia studies tell nothing about the causes of obesity. They do, however, illustrate that if you look hard enough for a needle in a haystack, you may find it. That doesn’t mean, however, that the needle created the haystack.

Wendell Cox
Guest Opinion