← All sources
Long-term Sustainability of Diabetes Prevention Approaches: A Systematic Review and Meta-analysis of Randomized Clinical Trials
Meta-analysis of 43 RCTs (n≈49,000) found lifestyle modification reduced diabetes incidence (RR 0.61) at the end of active treatment and retained a 28% relative reduction after multi-year follow-up, whereas medication effects largely washed out after short pharmacologic washout periods.
Design
RCTs (1990–2015) testing lifestyle modification (LSM) and/or medications >6 months for diabetes prevention in adults at risk; pooled relative risks of diabetes incidence.
Active treatment phase (pooled)
- LSM: RR 0.61 (95% CI 0.54–0.68) → ~39% relative risk reduction (19 trials)
- Medications: RR 0.64 (0.54–0.76) → ~36% RR reduction (19 trials)
- Risk difference: about 4.0 fewer cases per 100 person-years (95% CI 1.8–6.3) vs control (NNT ≈ 25)
Sustainability window
- LSM after mean ~7.2 y follow-up (5.7–9.4 y): RR 0.72 (0.60–0.86) (~28% RR reduction)
- Medications after mean ~17-week washout/follow-up: RR 0.95 (0.79–1.14) — no sustained incidence reduction in pooled medication arms
Publication
Haw JS, Galaviz KI, Straus AN, et al. JAMA Intern Med. 2017 Dec 1;177(12):1808-1817. PMID 29114778.
Outcomes
- End of active intervention: lifestyle modification vs control RR 0.61 (95% CI 0.54–0.68) for diabetes incidence; medications vs control RR 0.64 (0.54–0.76).
- Long-term follow-up: pooled lifestyle modification RR 0.72 (95% CI 0.60–0.86) at mean ~7.2 y; post-washout pooled medication RR 0.95 (0.79–1.14) at mean ~17 weeks—no sustained medication effect.