Metformin is a diabetes drug. Its best human evidence comes from the Diabetes Prevention Program, a trial in adults at high risk of diabetes, not in adults with obesity alone. This article covers what the trials show for diabetes, for weight and for safety, and what they do not show.
Preventing diabetes: the DPP
The Diabetes Prevention Program randomized adults at high risk of diabetes to metformin, intensive lifestyle change or placebo. Over 2.8 years, metformin lowered the risk of diabetes by 31% against placebo; lifestyle change lowered it by 58% (DPP, NEJM 2002).
Ten years of follow-up
In the DPP Outcomes Study, diabetes incidence over 10 years was 18% lower in the original metformin group than in the original placebo group (95% CI 7 to 28%), and 34% lower in the lifestyle group. The metformin group continued the drug, and the lifestyle group was offered extra support (DPP Outcomes Study, Lancet 2009).
Weight
Weight loss on metformin was modest. During the DPP, mean weight change was -2.06% on metformin and -0.02% on placebo. Over the unblinded follow-up the gap was 2.0% against 0.2%. The loss was durable for at least 10 years and tracked how much of the drug people took (DPP Outcomes Study, Diabetes Care 2012).
How much of the effect is the drug
When the DPP stopped metformin for a washout period, the study estimated that 26% of the 31% effect was a pharmacological effect that did not persist after the drug was stopped (DPP washout study, Diabetes Care 2003). Part of the benefit therefore depends on continuing the drug.
Safety
- Gastrointestinal symptoms were more common on metformin than on placebo and declined over time (DPP Outcomes Study, Diabetes Care 2012).
- Hemoglobin and hematocrit were slightly lower on metformin in the first year, with no further change afterwards (same study).
- The long-term extension identified no significant safety issues (same study).
What this does not say
- The DPP studied people at high risk of diabetes, not people with obesity alone. Metformin is not approved for weight loss.
- The weight figures are small averages, and trials of weight loss in people without diabetes are small and do not agree.
- The results come from the standard tablet in trials; compounded versions were not studied.
How we use this
Our physicians consider metformin when blood sugar or the risk of diabetes is part of the picture. We do not sell it as a weight-loss protocol.
If you are on it for weight, ask what it is expected to do for you and when we will check that it is doing it.
Explore weight loss treatmentsSources
- 1.Knowler WC et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin (Diabetes Prevention Program). N Engl J Med 2002. NEJM
- 2.Diabetes Prevention Program Research Group. 10-year follow-up of diabetes incidence and weight loss in the DPP Outcomes Study. Lancet 2009. The Lancet
- 3.Diabetes Prevention Program Research Group. Long-term safety, tolerability, and weight loss associated with metformin in the DPP Outcomes Study. Diabetes Care 2012. Diabetes Care
- 4.Effects of withdrawal from metformin on the development of diabetes in the Diabetes Prevention Program. Diabetes Care 2003. Diabetes Care