The recent article "How Minnesota got a handle on obesity" (Sept. 4) seemed like rare good news in the obesity crisis. A closer look, however, suggests that we shouldn't pat ourselves on our backs yet; we actually may not be any closer to touching our toes.

Start with the data: For this report, people were surveyed by phone about height, weight, activity level, diabetes, blood pressure, and consumption of fruits and vegetables. This required respondents to recall their health history over the last 30 days and to be truthful answering personal questions from a stranger.

But few people know their weight and height. Few remember their 30-day exercise history without tracking it. Many don't accurately recall the results of a checkup — or block it out. And many fudge about their health. The report noted as much in its disclaimer: "A number of studies have shown that rates of overweight and obesity are probably higher than shown by the data because people tend to underreport their weight and exaggerate their height."

Questionable data aside, there's an even more serious flaw in the report: It excludes overweight people.

If you had to read that again, it's no surprise. Almost two-thirds of American adults are classified as "overweight," "obese" or "severely obese." The largest subset, the "overweight" population, was excluded from the analysis.

When you look at overweight and obese Minnesotans, we don't look nearly so good. There's no statistically significant improvement in the percentage of those overweight and obese.

In other words, rather than good news, it's just as likely this study tells us that a few people who were considered very fat two years ago shifted — perhaps temporarily — into a category of moderately fat people. Assuming the data are good.

A victory lap might not be in order — even if we could make it.

America — including Minnesota — is losing the obesity struggle, because we're using an approach that does more harm than good. For more than 50 years, we've been told that "diet" — eliminating processed carbohydrates and fat in our diets, lowering our cholesterol — "and exercise" lead to good health.

Instead, this has led us to epidemic obesity, diabetes and heart disease.

Replacing one high-carbohydrate food with another — even a supposedly "healthier" one — does very little. There may be some minor differences between an Oreo cookie, a piece of fruit and a low-fat snack chip, but your body breaks them all down to (mostly) sugar.

On fat, emerging evidence strongly suggests that meat and saturated fats are actually healthy and that excessive carbohydrates — including fruit, vegetables and grains — are driving obesity, cardiovascular disease and diabetes.

As for cholesterol, no legitimate scientific institution supports the proposition that total blood cholesterol — alone — meaningfully predicts heart disease (though there is one subfraction of cholesterol that may correlate highly with risk).

Exercise is good — unless you're trying to achieve sustainable weight reduction. People who exercise have increased appetites relative to those who don't, and they also overestimate the calorie burn of exercise by up to 300 percent. This combination often results in high-calorie snacks as "rewards" that offset good behavior.

An increasing body of findings indicates that the low-carbohydrate diets of the early 1900s, when people weren't overweight and heart disease was rare, might be a better approach; these diets result in more weight loss, lower triglycerides (a good thing), higher HDL (the good cholesterol) and lower LDL (the bad).

Even so, winning consensus on change will not come easily. In addition to the government's and the medical community's inertia, the food industry has a bias against changing its production model, and the pharmaceutical industry profitably treats the overweight and ill. Both industries lobby Congress and the executive branch as well as influencers like the American Heart Association in support of their commercial agendas. As noted just recently, they have sponsored a substantial body of academic research in these fields that often ends up supporting the status quo.

To get our epidemic obesity levels down, we need to provide up-to-date, reliable information to the public. As long as we keep eating the wrong things and believe we can exercise ourselves into skinny jeans, we'll continue to be plagued by diabetes, cardiovascular disease and obesity — and the crushing medical costs they cause.

Andrew Grossman is a graduate of the Johns Hopkins School of Medicine and the Stanford Graduate School of Business. He is president of Ambient Consulting, an IT firm in Minneapolis, and CEO of Ignite, a medical wellness start-up.