First published June 2016, updated March 2022


Physical activity, independent of weight loss, may help to prevent or delay type 2 diabetes in people who are at high risk for diabetes, according to new follow-up data from the landmark Diabetes Prevention Program (DPP) study, presented at the ADA conference.

The DPP was primarily supported by National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), part of National Institutes of Health. Published in 2002, the study found that lifestyle changes, including moderate weight loss and increased physical activity, reduced the chance of developing type 2 diabetes by 58% over three years in overweight people with prediabetes. Study participants were randomised to one of three groups: lifestyle intervention, metformin or placebo. Both lifestyle and metformin substantially reduced the risk of developing type 2 diabetes, with the lifestyle intervention more effective than metformin. Participants in the lifestyle intervention arm of this multi-centre study received an intensive behavioral intervention composed of nutrition and physical activity with the two goals of being physically active for a minimum of 150 minutes per week, and achieving a weight loss of 7%.

Due to the success of the lifestyle intervention, all participants were provided with a modified version of the intervention and asked to participate in the DPP Outcomes Study (DPPOS), also supported by NIDDK. Despite the fact that everyone beginning the DPPOs had been offered the lifestyle intervention, the DPPOS demonstrated that the lifestyle arm maintained a significantly lower cumulative diabetes incidence than the other two groups, which could not be explained by differences in weight loss.

The current study is a 12-year follow-up of the subgroup of 1,793 participants (93% of eligible cohort; n for lifestyle=589; metformin=599; and placebo=605) who were part of a DPP ancillary study in which they were asked to wear an activity monitor for one week. The interviewer-administered Modifiable Activity Questionnaire was used to track physical activity (PA) yearly, and diabetes status was determined by annual oral glucose tolerance and semiannual fasting plasma glucose tests.

The researchers found that in all three treatment groups, diabetes incidence was lower for participants who were more physically active, regardless of changes in weight. Specifically, there was a 2% decrease in diabetes incidence (HR 0.98 [0.97, 0.99].

Lead investigators, Professor Andrea M. Kriska, Department of Epidemiology, University of Pittsburgh Graduate School of Public Health, said: "These current results show that physical activity, over an average of 12 years, decreased the chances of developing diabetes even after considering any changes in weight. This protective effect was greater in those who were less active at baseline. They also suggest that the lower development of diabetes across the entire study in those that took part in the lifestyle arm of the study may be partially explained by improvement in physical activity levels as well as weight loss."

"Until now, the importance of physical activity in preventing diabetes development in the DPP was thought to be due to its role in achieving weight loss and weight maintenance; however, it was not considered a strong key factor alone."