Prediabetes Interventions: What the Evidence Shows About Metformin vs Lifestyle
Synthesis of 10 studies including the landmark DPP/DPPOS trial and 2 recent meta-analyses — evaluating who benefits most from pharmacologic versus behavioural approaches to diabetes prevention. Moderate-to-strong evidence base.
This is an original evidence synthesis. See References for source studies.
One-Minute Takeaway
One-Minute Takeaway
- Strong evidence from the DPP and multiple RCTs supports intensive lifestyle intervention as the most effective prediabetes intervention, reducing diabetes incidence by 58% over three years compared with 31% for metformin — a benefit that persists at 22 years of follow-up.
- Metformin approaches lifestyle intervention efficacy in specific subgroups: adults under 60 with BMI ≥35 kg/m², those with higher baseline fasting glucose (≥110 mg/dL), and women with prior gestational diabetes.
- A 2024 meta-analysis of 12 RCTs found that adding metformin to lifestyle interventions modestly improved glycaemic markers but did not significantly reduce diabetes incidence beyond lifestyle alone.
- Key evidence gaps remain, including the lack of head-to-head trials in ethnically diverse populations, limited data on combining metformin with newer GLP-1 receptor agonists for prevention, and insufficient long-term cardiovascular outcome data from prevention trials.
Why Prediabetes Interventions Matter
An estimated 98 million adults in the United States — roughly one in three — meet criteria for prediabetes, yet fewer than 20% are aware of their status. Without intervention, up to 70% of individuals with prediabetes will eventually develop type 2 diabetes, with annual conversion rates of 5–10% depending on risk profile. The economic burden is substantial: healthcare costs attributable to prediabetes and diabetes exceeded $400 billion annually in the US by 2022.
For family medicine clinicians, the central question is not whether to intervene — the evidence clearly favours action — but how to match the right intervention intensity to each patient. The ADA Standards of Care (2026) recommend intensive lifestyle programmes as first-line therapy, with metformin considered for higher-risk subgroups. Understanding the comparative evidence helps clinicians move beyond a one-size-fits-all approach.
What Evidence Was Reviewed
This synthesis draws on 10 key studies spanning from the original DPP trial (2002) through two 2024–2025 meta-analyses. The evidence base is anchored by the DPP/DPPOS — the largest and longest US prevention trial with 22 years of follow-up — supplemented by international RCTs and recent pooled analyses that address the specific question of combined versus standalone approaches.
| Study | Design | n | Population | Key Finding | Quality |
|---|---|---|---|---|---|
| DPP (Knowler 2002) | RCT | 3234 | High-risk adults with IGT, BMI ≥24 | Lifestyle reduced diabetes by 58%, metformin by 31% vs placebo at 2.8 yr | High |
| DPPOS 15-yr (DPP Group 2015) | RCT follow-up | 2776 | Original DPP cohort | Lifestyle 27% and metformin 18% reduction persisting at 15 yr; differences attenuating | High |
| DPPOS 22-yr (DPP Group 2020) | RCT follow-up | 2779 | Original DPP cohort | 25% lifestyle and 18% metformin reduction persist at 22 yr; no CVD or cancer benefit | High |
| Crandall 2019 (DPP subgroup) | Subgroup analysis | 2155 | Metformin vs placebo arms | Greatest metformin benefit in higher FPG, higher HbA1c, and women with prior GDM | Moderate |
| Madsen 2019 (Cochrane) | SR/MA | NR | Adults with prediabetes | Metformin vs intensive lifestyle: no significant difference (RR 0.80, 95% CI 0.47–1.37) | High |
| Amer 2024 | MA of 12 RCTs | 2720 | Adults with prediabetes | Metformin + lifestyle reduced HbA1c (SMD −0.10) and FPG vs lifestyle alone; diabetes incidence trended lower | Moderate |
| Karagiannis 2025 | SR/MA | NR | Metformin-naïve high-risk adults | Metformin reduces T2D incidence by 23% overall (OR 0.77, 95% CI 0.67–0.88); not superior to intensive lifestyle | High |
| Finnish DPS (Tuomilehto 2001) | RCT | 522 | Overweight adults with IGT | Lifestyle reduced diabetes by 58% vs control | High |
| Da Qing (Pan 1997; Li 2008) | RCT + follow-up | 577 | Chinese adults with IGT | Diet and exercise reduced diabetes by 31–46%; 30-yr follow-up shows reduced mortality | Moderate |
| Petrie 2024 (Narrative review) | Review | NR | Broad (metformin indications) | Summarised metformin prevention data; NNT 6.9 for lifestyle vs 13.9 for metformin over 3 yr | Moderate |
Key Findings on Prediabetes Interventions
Lifestyle Intervention Remains the Strongest Preventive Strategy
Across the three landmark prevention trials — DPP, Finnish DPS, and Da Qing — intensive lifestyle programmes targeting moderate weight loss (5–7% of body weight) and 150 minutes per week of physical activity consistently reduced diabetes incidence by 42–58% over 3–6 years. The DPP demonstrated an NNT of approximately 7 over three years, meaning that for every seven participants enrolled in the structured programme, one case of diabetes was prevented. These benefits have proven durable: the DPPOS documented a 25% sustained reduction at 22 years of follow-up, even as the initial weight loss advantage partially attenuated over time.
The 2025 meta-analysis by Karagiannis and colleagues confirmed that when metformin is compared directly to intensive lifestyle modification, no significant difference favouring metformin emerges. This finding is consistent with the original Cochrane review (Madsen 2019), which pooled seven RCTs and found no significant advantage for metformin over intensive behavioural approaches (RR 0.80, 95% CI 0.47–1.37).
Metformin Shows Selective Benefit in Higher-Risk Subgroups
While metformin’s overall effect was smaller than lifestyle intervention in the DPP (31% vs 58% risk reduction at 3 years), subgroup analyses consistently identify populations where metformin approaches or matches lifestyle efficacy. Adults with BMI ≥35 kg/m² and younger participants aged 25–44 showed comparable benefit from either approach. Women with a history of gestational diabetes experienced an equivalent 50% risk reduction with either intervention, a finding that persisted through 10-year follow-up.
The 15-year DPPOS subgroup analysis further refined the profile of metformin responders: individuals with baseline fasting glucose ≥110 mg/dL derived the greatest benefit, with an NNT of just 5 in the highest-risk quartile and an absolute risk reduction exceeding 21% over three years. Those with HbA1c 6.0–6.4% also demonstrated more robust responses. These findings suggest that metformin’s value increases as metabolic risk intensifies.
The DPP enrolled 350 women with a history of gestational diabetes. Both metformin and intensive lifestyle intervention reduced progression to type 2 diabetes by approximately 50% in this subgroup — a notably stronger effect than seen in the overall cohort. This equivalence persisted through the 10-year DPPOS follow-up. The ADA Standards of Care (2026) specifically recommend considering metformin for women with prior GDM who have prediabetes, recognising that this population carries a lifetime diabetes risk exceeding 50% and may face practical barriers to sustained intensive lifestyle programmes during child-rearing years.
In the DPP, participants with severe obesity (BMI ≥35 kg/m²) achieved similar diabetes risk reductions whether assigned to metformin or intensive lifestyle intervention. This subgroup represented approximately 20% of the DPP cohort. The likely explanation is that metformin’s weight-neutral to modestly weight-reducing effect provides proportionally greater metabolic benefit in individuals with more pronounced insulin resistance. However, it is important to note that lifestyle intervention still offered broader cardiometabolic benefits including improvements in blood pressure, lipids, and physical fitness that metformin alone did not replicate.
Age modifies the relative benefit of metformin versus lifestyle. In adults aged 25–44, metformin performed comparably to the intensive lifestyle programme in the DPP. By contrast, participants aged 60 and older derived minimal benefit from metformin, with lifestyle intervention remaining the clearly superior strategy. The ADA now recommends considering metformin specifically for those aged 25–59, reflecting this age-dependent gradient. For younger adults with prediabetes who face decades of potential metabolic exposure, this finding has particular relevance for long-term prevention planning.
The 2024 meta-analysis by Amer and colleagues specifically examined whether adding metformin to lifestyle interventions offers incremental benefit. Across 12 RCTs (n = 2720), the combination produced modest additional reductions in HbA1c (SMD −0.10, 95% CI −0.19 to −0.01) and fasting glucose compared with lifestyle alone. However, diabetes incidence, body weight, blood pressure, lipid profiles, and insulin resistance measures did not differ significantly between groups. The 2025 analysis by Karagiannis et al. reached a similar conclusion, noting that metformin combined with lifestyle was not superior to lifestyle alone for preventing diabetes. These findings suggest that the glycaemic benefits of combination therapy are real but modest, and may require longer treatment durations to translate into meaningful diabetes prevention.
Long-Term Durability and Attenuation of Effects
A critical finding from the DPPOS is that the gap between lifestyle and metformin narrows substantially over time. While lifestyle intervention was nearly twice as effective as metformin at three years, by the 15-year follow-up the difference between groups was no longer statistically significant. This convergence reflects two trends: weight regain in the lifestyle group after the initial intensive phase, and sustained modest weight loss with continued metformin use. By 15 years, cumulative diabetes incidence was 55% in the lifestyle group, 56% in the metformin group, and 62% in the placebo group — illustrating that while both interventions delay progression, neither fully prevents it in a high-risk cohort.
Notably, the 22-year DPPOS data showed no significant reduction in major cardiovascular events or cancer incidence in either intervention group, despite sustained diabetes prevention. Participants who avoided diabetes development, however, had a 28% lower prevalence of microvascular complications regardless of treatment assignment. In women specifically, lifestyle intervention was associated with a 21–22% reduction in aggregate microvascular disease compared with both metformin and placebo groups.
Quality & Consistency of Evidence
What the Evidence Does Not Show
Despite decades of research, several clinically important questions remain unresolved.
Optimal duration of combined therapy. The 2024 Amer meta-analysis noted that the combination of metformin plus lifestyle may require longer treatment periods to affect diabetes incidence. No RCT has established an evidence-based duration threshold for combination therapy, nor defined criteria for stepping down from combined to single-modality approaches.
Comparative efficacy with GLP-1 receptor agonists. While semaglutide and tirzepatide show potent weight reduction and diabetes prevention effects, no completed trial directly compares them head-to-head with metformin or structured lifestyle programmes for prediabetes prevention. This gap is particularly relevant as these agents enter broader clinical use.
Ethnic and geographic generalisability. The DPP was conducted in a diverse US cohort, but most subsequent meta-analysed RCTs come from single-ethnic populations. Whether the subgroup-specific metformin benefits identified in the DPP (e.g., BMI ≥35, prior GDM) apply equally in South Asian, Middle Eastern, and African populations — where prediabetes prevalence is rising fastest — is not established.
Effect on cardiovascular endpoints. Twenty-two years of DPP/DPPOS follow-up showed no reduction in major cardiovascular events with either intervention. Whether this reflects the open-label design after DPP conclusion, contamination between groups (37% of placebo participants were on metformin by year 15), or a genuine absence of cardiovascular benefit remains unclear.
Practical Implications
Based on the current evidence, the following considerations may inform clinical decision-making for adults with prediabetes. These reflect the synthesised findings and current ADA recommendations — they are not directive prescribing guidance.
Evidence Grade & Bottom Line
Overall Evidence Grade: Strong (Lifestyle) / Moderate (Metformin Subgroup Selection)
The evidence supporting lifestyle intervention for diabetes prevention is robust, consistent across populations, and durable over decades. The evidence supporting metformin — while strong versus placebo — is moderate when it comes to identifying which specific patients benefit most, as these recommendations rely primarily on subgroup analyses from a single large trial.
A “strong” grade for lifestyle intervention means the core finding — that structured programmes prevent or delay diabetes — is unlikely to change with future research. The “moderate” grade for metformin subgroup selection reflects that while metformin clearly works for prevention, the precision of patient selection criteria may evolve as new data emerge, particularly regarding combination strategies and newer pharmacologic alternatives.
- Intensive lifestyle intervention reduces diabetes incidence by 42–58% and should be offered to all adults with prediabetes.
- Metformin reduces diabetes incidence by 23–31% versus placebo, with particular benefit in adults under 60 with BMI ≥35, elevated fasting glucose, or prior gestational diabetes.
- The gap between lifestyle and metformin efficacy narrows over 15+ years of follow-up, largely due to weight regain in lifestyle groups.
- Adding metformin to lifestyle may provide modest additional glycaemic benefit but has not been shown to significantly reduce diabetes incidence beyond lifestyle alone.
- Neither intervention has demonstrated reduction in major cardiovascular events over 22 years of follow-up, despite sustained diabetes prevention — a finding that warrants further study.
Article Information
For Educational Purposes Only. This is an original evidence synthesis informed by the studies listed below. It does not replace clinical judgement. Drug dosages should be verified against current prescribing information.
References
- Amer BE, Abdelgalil MS, Hamad AA, et al. Metformin plus lifestyle interventions versus lifestyle interventions alone for the delay or prevention of type 2 diabetes in individuals with prediabetes: a meta-analysis of randomized controlled trials. Diabetol Metab Syndr. 2024;16:273. DOI: 10.1186/s13098-024-01504-8
- Karagiannis T, et al. Metformin’s overall effectiveness and combined action with lifestyle interventions in preventing type-2 diabetes mellitus in high-risk metformin-naïve patients: an updated systematic review and meta-analysis of published RCTs. J Clin Med. 2025;14(14):4947. DOI: 10.3390/jcm14144947
- Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002;346(6):393–403. DOI: 10.1056/NEJMoa012512
- Diabetes Prevention Program Research Group. Long-term effects of lifestyle intervention or metformin on diabetes development and microvascular complications over 15-year follow-up: the Diabetes Prevention Program Outcomes Study. Lancet Diabetes Endocrinol. 2015;3(11):866–875. DOI: 10.1016/S2213-8587(15)00291-0
- Crandall JP, Knowler WC, Kahn SE, et al. Long-term effects of metformin on diabetes prevention: identification of subgroups that benefited most in the Diabetes Prevention Program and Diabetes Prevention Program Outcomes Study. Diabetes Care. 2019;42(4):601–608. DOI: 10.2337/dc18-1970
- Madsen KS, Chi Y, Metzendorf MI, Richter B, Hemmingsen B. Metformin for prevention or delay of type 2 diabetes mellitus and its associated complications in persons at increased risk for the development of type 2 diabetes mellitus. Cochrane Database Syst Rev. 2019;12(12):CD008558. DOI: 10.1002/14651858.CD008558.pub2
- Tuomilehto J, Lindström J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med. 2001;344(18):1343–1350. DOI: 10.1056/NEJM200105033441801
- Pan XR, Li GW, Hu YH, et al. Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance: the Da Qing IGT and Diabetes Study. Diabetes Care. 1997;20(4):537–544. DOI: 10.2337/diacare.20.4.537
- Petrie JR. Metformin beyond type 2 diabetes: emerging and potential new indications. Diabetes Obes Metab. 2024;26(Suppl 3):31–41. DOI: 10.1111/dom.15756
- American Diabetes Association Professional Practice Committee. Prevention or delay of diabetes and associated comorbidities: Standards of Care in Diabetes—2026. Diabetes Care. 2026;49(Suppl 1):S50–S66. DOI: 10.2337/dc26-S003
- Echouffo-Tcheugui JB, Perreault L, Ji L, Dagogo-Jack S. Diagnosis and management of prediabetes: a review. JAMA. 2023;329(14):1206–1216. DOI: 10.1001/jama.2023.4063
- DPP/DPPOS Research Group. Analysis of long-term follow-up of a randomized clinical trial with departures from assigned treatments: estimation of metformin effects on diabetes and its complications in the Diabetes Prevention Program Outcomes Study. Diabetes Care. 2025;48(10):1668–1676. DOI: 10.2337/dc25-0584