Vol. I · No. 33
Thursday, August 13, 2026
Issue: Summer · 2026
Established · MMXXVI
— The evidence base for longevity medicine —
Indexed by PubMed · CTG · Cochrane
Editorial team · geroevidence.com
Subscription · app.geroevidence.com

The ACCORD trial: what it taught longevity medicine about intensive glucose control

The Action to Control Cardiovascular Risk in Diabetes trial found that intensive glucose lowering increased all-cause mortality in high-risk type 2 diabetes patients. It is one of the most important cautionary data points in metabolic longevity medicine.

By Geroevidence editorial team·Published 19 July 2026·Category Clinical Trials·11 min read
HR 1.22
All-cause mortality — intensive arm
vs standard glycemic control
Stopped early
Intensive arm halted
Feb 2008 · safety concern
10,251
Participants enrolled
T2D + high CV risk · mean HbA1c 8.1%
<6.0%
Target HbA1c — intensive arm
Standard arm target: 7.0–7.9%
§ The hypothesis and what happened

ACCORD was designed to test whether normalizing blood glucose to near-normal levels in high-risk type 2 diabetics would reduce cardiovascular events. The hypothesis was biologically plausible and the clinical community largely expected a positive result.

The intensive arm targeting HbA1c below 6.0% was halted in February 2008 after a median follow-up of 3.5 years due to increased all-cause mortality: 257 deaths in the intensive arm versus 203 in the standard arm (HR 1.22, 95% CI 1.01–1.46). The cause of the excess deaths was not definitively established. Hypoglycemia was increased in the intensive arm but did not fully explain the mortality difference in adjudicated analyses.

The final NEJM publication (2008) and subsequent analyses confirmed the finding. HbA1c was successfully lowered to mean 6.4% in the intensive arm versus 7.5% in the standard arm, confirming protocol adherence. The mortality finding was not a result of failure to achieve the target.

§ Key evidence
RCT
ACCORD — NEJM, 2008
n=10,251
High-risk T2D patients. Intensive HbA1c target <6.0% vs standard 7.0-7.9%. Intensive arm stopped early due to all-cause mortality HR 1.22 (95% CI 1.01-1.46). CV death HR 1.35. Mechanism of excess mortality not established — hypoglycemia increased but did not fully explain the signal. One of the most important negative findings in metabolic medicine.
Extended follow-up
ACCORD 10-year follow-up — NEJM, 2016
n=8,601
Participants followed after trial completion. Excess mortality in the intensive arm attenuated over time — long-term all-cause mortality HR 1.07 (95% CI 0.97-1.18, non-significant). CV outcomes also attenuated. Suggests the mortality harm was concentrated in the intensive period of glucose control, not a permanent legacy effect.
Mechanism analysis
ACCORD mechanism analyses — 2010-2018
n=10,251
Multiple post-hoc analyses attempted to explain excess mortality. Hypoglycemia was associated with mortality in both arms but was more frequent in the intensive arm. However, severe hypoglycemia events did not fully mediate the mortality difference in adjudicated analyses. Weight gain was greater in intensive arm. Polypharmacy required to achieve glycemic targets may have contributed. Definitive mechanism was never established.
§ What ACCORD means for longevity medicine

ACCORD is the most important piece of evidence against the naive version of the glucose-aging hypothesis — that lower blood glucose is always better for longevity. In high-risk diabetic patients, aggressive glucose lowering appears to increase mortality. The lesson is not that glucose control is unimportant but that the relationship is nonlinear and population-dependent.

For longevity medicine practitioners, ACCORD is particularly relevant when considering metformin use in patients at risk for hypoglycemia, or when advising patients on dietary glucose restriction targets. The ACCORD finding applies specifically to high-risk diabetic patients on intensive pharmacological glucose lowering — it does not directly apply to healthy non-diabetic adults. But it is a clear reminder that interventions targeting metabolic pathways can have unexpected direction-of-effect in specific populations.

§ The honest clinical position

ACCORD is cited as a cautionary trial about the dangers of treating biomarkers rather than patients. Achieving target HbA1c in high-risk diabetics increased mortality despite successfully improving the surrogate endpoint. It reinforces a fundamental evidence principle: improving a surrogate marker does not guarantee benefit — and in complex biological systems, aggressive surrogate normalization can cause harm. This principle applies across longevity medicine whenever an intervention claims benefit based solely on improving a biomarker.

This information is provided for educational reference only and does not constitute medical advice or a treatment recommendation.

Full trial database
All indexed trials · daily alerts · evidence updates
Start free trial
14 DAYS FREE · NO CARD REQUIRED
Related trials
TAME trial update
Metformin longevity · 2026 status
EMPA-REG OUTCOME
Empagliflozin · CV death reduction
SELECT trial
Semaglutide · non-diabetic CV outcomes
View all interventions →