Weinstein et al followed more than 38000 women from the Women's Health Study over almost seven years of follow-up, linking BMI and self-reported recreational activity level at baseline to subsequent development of diabetes. In all, 1361 women developed type 2 diabetes, with overweight and obese women being significantly more likely to develop the disease. Likewise, more physically active women were less likely to develop diabetes; however, in combined analysis, overweight and obese women, whether they were active or not, had a significantly elevated risk of developing diabetes.

"Although BMI and physical inactivity are independent predictors of incident diabetes, the magnitude of the association with BMI was greater than with physical activity in combined analyses," the authors write.

Weinstein elaborated to heartwire, "Adiposity plays a large role in diabetes and I think that that may be key in this case."

In the second study, Dr Timothy R Wessel (University of Florida, Gainesville) and colleagues examined similar parameters against the end point of obstructive coronary disease or CV events over almost four years in 936 women.

During this period, 38% of women had a first adverse event and 13% had a major adverse event, with 8% dying. Strikingly, while overweight women were more likely to have CAD risk factors, BMI and abdominal obesity were not significantly associated with CAD or CV events. By contrast, women with lower scores on two separate, well-validated measures of physical activity (with lower scores representing lower activity levels) were significantly more likely to have a major adverse CV event over follow-up.

"These results suggest that fitness may be more important than overweight or obesity for CV risk in women," Wessel and colleagues write.

In an accompanying editorial, Drs Steven N Blair and Tim S Church (The Cooper Institute, Dallas, TX) point out that most studies of obesity and health have not adequately accounted for physical activity, such that the two new studies are welcome additions, even if they have "somewhat divergent results."3

With the results of these studies we may be able to tailor programs to individual patients a little bit better.

Differences between the two studies may stem from the different study populations, methods, and outcomes, they note. In particular, it is possible that the two separate measures, physical activity and function (only one of which was self-reported), used by Wessel et al were "more accurate" than that used by Weinstein and colleagues.

Asked to comment on this hypothesis for heartwire, Weinsten pointed out that the population they studied was much larger and included women from all over the US, self-reporting their physical-activity level. "We don't necessarily have the same measure of physical activity as the other study, but I still think it's a reliable and valid method," she said.

Taken together, the two studies provide "a timely opportunity" to revisit the ongoing debate over the importance of fitness vs fatness, Blair and Church observe. But rather than carping on differences, the editorialists offer a resolution.

"Although the debate may never be fully resolved, the relative contribution of fitness and obesity to overall health and risk actually may be a trivial matter because a common treatment is already available for both low fitness and excess body weight," they note. "In essence, physical activity is the common denominator for the clinical treatment of low fitness and excess weight, making the 'fitness vs fatness' debate largely academic."

To heartwire, Weinstein says she concurs with Blair and Church's "resolution" but believes the different results of the two studies may still have value.

"I agree that physical activity in conjunction with weight loss is the important message here," she says. "But with the results of these studies we may be able to tailor programs to individual patients a little bit better."