Prevention of Clinical Diabetes Mellitus

Diabetes mellitus affects more than 100 million individuals worldwide and, because of known risk factors common to many individuals that can be manipulated, the development of type 2 diabetes may potentially be modifiable. A number of clinical trials have addressed this hypothesis through dietary modification, physical activity, and drug treatment. Although some studies indicate protection against diabetes development, conclusions remain limited for reasons of study design problems with randomization, subject selection, or intervention intensity.

Six prospective studies have examined the predictors of progression from impaired glucose tolerance to type 2 diabetes. Across the six studies (n, 177-693), the incidence rates averaged 57.2 per 1000 patient-years.

Incidence rates were sharply higher for those in the top quartile of fasting glucoses, but increased linearly with increasing 2-h postchallenge glucose levels. Native and Hispanic incidence rates were higher than those of Caucasians. Baseline age showed no consistent pattern. Obesity was positively associated with future diabetes mellitus, no matter how body fatness was measured.

One study in the elderly and two in younger patients suggest that diet and exercise may prevent diabetes mellitus in patients with impaired glucose tolerance. In Da Qing, China, a screening program detected 577 individuals (mean age, 45.0) with impaired glucose tolerance (IGT). Following stratification by body weight, patients were assigned to either active treatment (one of three possibilities: diet, exercise, or a combination of both) or to a control group who received brochures on diet, exercise, and IGT.

At 6 years, more than two-thirds of the control group (67.7%) developed diabetes, compared to less than half in each of the active treatment groups (43.8% of diet group, 41.1% of exercise group, and 46% in the exercise plus diet group; all p < 0.05 compared to control but no significant differences compared to each other). After adjustment for differences in baseline obesity and fasting glucose, reductions in risk of developing diabetes were 31% for diet, 46% for exercise, and 42% for diet plus exercise, each highly significant compared to control.

Another multicenter, partially blind study in Finland identified 522 overweight middle-aged persons with IGT (mean age, 55; persons over age 65 were excluded) who were randomly assigned to usual care or to individualized counseling aimed at reducing weight and total fat intake while increasing fiber intake and exercise.

Mean weight loss at 2 years was significantly different between the two groups; 7.7 pounds in the intervention group and 1.8 pounds for the control group. At 2 years, the cumulative incidence of diabetes was 6% in the intervention group versus 14% in controls (at 4 years, 11% and 23%, respectively), a risk reduction of 58% (95% CI, 0.3-0.7; p < 0.00l).