Study analysis · The Journals of Gerontology Series A: Biological Sciences and Medical Sciences · 2025
This diabetes drug might be the secret to living past 90—here’s the shocking data.
Women with type 2 diabetes who took metformin were 30% less likely to die before age 90 than those who took sulfonylurea.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
In simple terms
This study looked at women who took one medicine (metformin) versus another (sulfonylurea) for diabetes and saw that those on metformin were more likely to live to 90. But it didn't randomly assign them — so we can't be sure the medicine caused it. Maybe the women who got metformin were healthier to begin with.
What’s the bottom line?
Scientists looked at older women with type 2 diabetes who started either metformin or sulfonylurea pills and tracked who lived to 90.
How strong is this study?
The researchers did a really good job trying to make this fair — they matched women who were similar in age, health, and habits so the only big difference was the medicine. But since they didn't randomly pick who got which medicine, we still can't be 100% sure the medicine was the reason for living longer.
40 / 100
- COI disclosure+40/40
- Data availabilitydata not shared
- Code availabilitycode not shared
53 / 100
- Randomizationnot randomized
- Blindingblinding unclear
- Control group+15/15
- Sample size (n=438)+17.8/20
- Follow-up+10/10
100 / 100
77 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registrationnot pre-registered
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 566 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
This design cannot establish causation — the findings describe an association, not a cause. This is an observational study using propensity score matching to emulate a randomized trial, but it lacks randomization and blinding. Residual confounding is possible, and the authors explicitly state that causality cannot be inferred due to the observational design.
No Conflicts
No conflicts of interest identified
No conflicts of interest or funding disclosures were reported in the provided text.
Independent Analysis Safeguards
- Propensity score matching to balance confounders
- Target trial emulation framework to reduce bias
- Use of validated WHI data with physician adjudication of outcomes
- Intention-to-treat analysis
The study is observational and uses a robust methodological framework (target trial emulation) to minimize bias. However, the absence of any declared funding or conflict of interest section limits transparency. No industry ties or author affiliations with pharmaceutical companies are indicated.
Key takeaways
- 01
Women on metformin had 3.7 deaths per 100 people per year; those on sulfonylurea had 5.0 deaths per 100 people per year — a 30% lower risk with metformin.
- 02
This means for every 100 women on metformin, about 3 more women lived to 90 compared to those on sulfonylurea over 15 years — a meaningful difference for aging.
Surprising findings
- Metformin’s survival benefit persisted even after excluding deaths in the first two years.Many assume early deaths skew results—like sicker patients being prescribed metformin. But even after removing those early deaths, the 30% advantage held (HR 0.72), suggesting a true long-term effect.
- The E-value for metformin’s effect was 1.88—meaning unmeasured confounders would need a risk ratio of nearly 2 to fully explain the result.Most observational studies have E-values under 1.5, meaning they’re easily explained by bias. An E-value of 1.88 is unusually robust for an observational study.
Practical takeaways
If you're over 60 and newly diagnosed with type 2 diabetes, ask your doctor if metformin is a better first-line option than sulfonylureas like glipizide.
This study only looked at postmenopausal women—results may not apply to men, younger people, or those with kidney issues.
medium confidenceWhy this study matters
Metformin vs. Sulfonylurea: The Longevity Edge
In a 15.5-year study of 438 postmenopausal women with type 2 diabetes, those who started metformin had 3.7 deaths per 100 person-years—compared to 5.0 for those on sulfonylurea. That’s a 30% lower risk of dying before 90 (HR 0.70, 95% CI: 0.56–0.88).
This isn’t about curing diabetes—it’s about living longer with it. For millions of older women managing type 2, this could mean choosing a drug that doesn’t just control blood sugar but might help them see their grandchildren grow up.
Not All Sulfonylureas Are Equal
Metformin’s survival advantage was much stronger against glipizide (HR 0.63) than glyburide (HR 0.78). This suggests the sulfonylurea class isn’t uniform—some drugs may be riskier than others.
Doctors often treat sulfonylureas as interchangeable. This study hints that prescribing glipizide over glyburide could matter for longevity—something rarely discussed in clinics.
The Income Confounder That Disappeared
When researchers adjusted for income and region—factors known to affect lifespan—the metformin advantage weakened (HR 0.81) and lost statistical significance. But the E-value was still 1.88, meaning unmeasured confounders would need to be very strong to explain it away.
It’s rare for a study to test socioeconomic confounders so directly—and still find a persistent signal. This suggests metformin’s effect might be biological, not just a result of wealthier patients getting better care.
Want the whole report?
Detailed mode opens the full scientific breakdown — every score component, the methodology, conflicts of interest, the evidence analysis behind each claim, and the raw study data.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Scientists looked at older women with type 2 diabetes who started either metformin or sulfonylurea pills and tracked who lived to 90.
Research results
Women on metformin had 3.7 deaths per 100 people per year; those on sulfonylurea had 5.0 deaths per 100 people per year — a 30% lower risk with metformin.
What this means - more context
This means for every 100 women on metformin, about 3 more women lived to 90 compared to those on sulfonylurea over 15 years — a meaningful difference for aging.
This study examines whether metformin initiation, compared to sulfonylurea initiation, is associated with exceptional longevity (survival to age 90) in postmenopausal women with newly diagnosed type 2 diabetes.
Among 438 propensity score-matched postmenopausal women with type 2 diabetes, metformin initiation was associated with a 30% lower adjusted risk of death before age 90 compared to sulfonylurea initiation (HR 0.70, 95% CI: 0.56–0.88), with incidence rates of 3.7 vs. 5.0 deaths per 100 person-years. The association persisted after excluding early deaths and was stronger for glipizide than glyburide, but lost significance when adjusting for income and region.
Methods Used
New-user, active comparator cohort study using target trial emulation with 1:1 propensity score matching on demographics, comorbidities, BMI, lifestyle, and medication use. Intention-to-treat analysis of 438 women ≥60 years with incident type 2 diabetes from the Women’s Health Initiative, followed for a mean of 15.5 years.
Main Finding
Metformin initiation was associated with a 30% lower adjusted risk of death before age 90 compared to sulfonylurea initiation (hazard ratio 0.70, 95% CI: 0.56–0.88), with incidence rates of 3.7 vs. 5.0 deaths per 100 person-years.
Confidence Level
Moderate — findings are robust to sensitivity analyses (e.g., excluding early deaths, differential sulfonylurea effects), but residual confounding remains possible due to observational design and lack of randomization.
Study Flags
Red Flags
- •Observational design — cannot prove causation
- •No data on medication adherence, dosage, or disease severity (e.g., HbA1c)
- •Residual confounding possible despite propensity matching
No biological mechanisms were identified in this study. This may be an epidemiological, observational, or survey-based study that reports associations rather than proposing causal biological pathways.
Surprising Findings
Metformin’s survival benefit persisted even after excluding deaths in the first two years.
Many assume early deaths skew results—like sicker patients being prescribed metformin. But even after removing those early deaths, the 30% advantage held (HR 0.72), suggesting a true long-term effect.
Practical Takeaways
If you're over 60 and newly diagnosed with type 2 diabetes, ask your doctor if metformin is a better first-line option than sulfonylureas like glipizide.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 566 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
Human Cohort Study
Subject
Moderate probability
on the GRADE evidence scale
This study looked at women who took one medicine (metformin) versus another (sulfonylurea) for diabetes and saw that those on metformin were more likely to live to 90. But it didn't randomly assign them — so we can't be sure the medicine caused it. Maybe the women who got metformin were healthier to begin with.
No conflicts of interest were detected in this study. No score impact.
Strengths
- Used target trial emulation framework to mimic a randomized trial
- Employed 1:1 propensity score matching on 15+ confounders
- Used intention-to-treat analysis
Weaknesses
- No randomization — observational design limits causal inference
- Blinding status unknown — potential for detection bias
- Medication initiation date based on self-report at year 3 visit
Methodology
Evidence Keywords
Statistical Reporting
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Scientists looked at older women with type 2 diabetes who started either metformin or sulfonylurea pills and tracked who lived to 90.
Research results
Women on metformin had 3.7 deaths per 100 people per year; those on sulfonylurea had 5.0 deaths per 100 people per year — a 30% lower risk with metformin.
What this means - more context
This means for every 100 women on metformin, about 3 more women lived to 90 compared to those on sulfonylurea over 15 years — a meaningful difference for aging.
This study examines whether metformin initiation, compared to sulfonylurea initiation, is associated with exceptional longevity (survival to age 90) in postmenopausal women with newly diagnosed type 2 diabetes.
Among 438 propensity score-matched postmenopausal women with type 2 diabetes, metformin initiation was associated with a 30% lower adjusted risk of death before age 90 compared to sulfonylurea initiation (HR 0.70, 95% CI: 0.56–0.88), with incidence rates of 3.7 vs. 5.0 deaths per 100 person-years. The association persisted after excluding early deaths and was stronger for glipizide than glyburide, but lost significance when adjusting for income and region.
Methods Used
New-user, active comparator cohort study using target trial emulation with 1:1 propensity score matching on demographics, comorbidities, BMI, lifestyle, and medication use. Intention-to-treat analysis of 438 women ≥60 years with incident type 2 diabetes from the Women’s Health Initiative, followed for a mean of 15.5 years.
Main Finding
Metformin initiation was associated with a 30% lower adjusted risk of death before age 90 compared to sulfonylurea initiation (hazard ratio 0.70, 95% CI: 0.56–0.88), with incidence rates of 3.7 vs. 5.0 deaths per 100 person-years.
Confidence Level
Moderate — findings are robust to sensitivity analyses (e.g., excluding early deaths, differential sulfonylurea effects), but residual confounding remains possible due to observational design and lack of randomization.
Study Flags
Red Flags
- •Observational design — cannot prove causation
- •No data on medication adherence, dosage, or disease severity (e.g., HbA1c)
- •Residual confounding possible despite propensity matching
No biological mechanisms were identified in this study. This may be an epidemiological, observational, or survey-based study that reports associations rather than proposing causal biological pathways.
Surprising Findings
Metformin’s survival benefit persisted even after excluding deaths in the first two years.
Many assume early deaths skew results—like sicker patients being prescribed metformin. But even after removing those early deaths, the 30% advantage held (HR 0.72), suggesting a true long-term effect.
Practical Takeaways
If you're over 60 and newly diagnosed with type 2 diabetes, ask your doctor if metformin is a better first-line option than sulfonylureas like glipizide.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 566 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
Human Cohort Study
Subject
Moderate probability
on the GRADE evidence scale
This study looked at women who took one medicine (metformin) versus another (sulfonylurea) for diabetes and saw that those on metformin were more likely to live to 90. But it didn't randomly assign them — so we can't be sure the medicine caused it. Maybe the women who got metformin were healthier to begin with.
No conflicts of interest were detected in this study. No score impact.
Strengths
- Used target trial emulation framework to mimic a randomized trial
- Employed 1:1 propensity score matching on 15+ confounders
- Used intention-to-treat analysis
Weaknesses
- No randomization — observational design limits causal inference
- Blinding status unknown — potential for detection bias
- Medication initiation date based on self-report at year 3 visit
Methodology
Evidence Keywords
Statistical Reporting
Scoring
How strong is this study?
The researchers did a really good job trying to make this fair — they matched women who were similar in age, health, and habits so the only big difference was the medicine. But since they didn't randomly pick who got which medicine, we still can't be 100% sure the medicine was the reason for living longer.
40 / 100
- COI disclosure+40/40
- Data availabilitydata not shared
- Code availabilitycode not shared
53 / 100
- Randomizationnot randomized
- Blindingblinding unclear
- Control group+15/15
- Sample size (n=438)+17.8/20
- Follow-up+10/10
100 / 100
77 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registrationnot pre-registered
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 566 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
This design cannot establish causation — the findings describe an association, not a cause. This is an observational study using propensity score matching to emulate a randomized trial, but it lacks randomization and blinding. Residual confounding is possible, and the authors explicitly state that causality cannot be inferred due to the observational design.
No Conflicts
No conflicts of interest identified
No conflicts of interest or funding disclosures were reported in the provided text.
Independent Analysis Safeguards
- Propensity score matching to balance confounders
- Target trial emulation framework to reduce bias
- Use of validated WHI data with physician adjudication of outcomes
- Intention-to-treat analysis
The study is observational and uses a robust methodological framework (target trial emulation) to minimize bias. However, the absence of any declared funding or conflict of interest section limits transparency. No industry ties or author affiliations with pharmaceutical companies are indicated.