What Actually Works: The Evidence
Reviewed by CDE Mehandi Sharma · Updated
Diabetes prevention is unusually well studied, which means you can see the size of the effect rather than taking it on faith.
The Diabetes Prevention Program randomised 3,234 people with prediabetes. Over about 2.8 years, diabetes developed in 11.0 cases per 100 person-years on placebo, 7.8 on metformin, and 4.8 on intensive lifestyle intervention — a 58% risk reduction for lifestyle and 31% for metformin.
Lifestyle beat the drug, substantially.
The Indian Diabetes Prevention Programme tested the same question in Asian Indians with impaired glucose tolerance. Over three years it found relative risk reductions of 28.5% with lifestyle modification, 26.4% with metformin, and 28.2% with both together.
Two things are worth reading carefully in those Indian numbers. The reduction was real and meaningful. And unlike the American trial, lifestyle and metformin performed about equally — combining them added nothing, which is a genuinely useful finding.
Longer-term follow-up elsewhere is the most reassuring part: the Da Qing study found sustained reductions in diabetes incidence ten to thirty years after the intervention ended, with 26 to 35% risk reductions across outcomes.
Drug benefits behave differently. When a medication used in these trials was withdrawn, its effect was not sustained — whereas the lifestyle benefit persisted for decades.