August 31, 2026 · Science · Biomarkers
What the STAREE Trial Reported
Atorvastatin cut major cardiovascular events by 30% in adults over 70 who had no diagnosed heart disease. STAREE randomized 9,971 people aged 70 and older to 40 mg of atorvastatin daily or to a placebo, then followed them for a median of 5.9 years. Cardiovascular death, nonfatal heart attack, stroke and coronary revascularization occurred in 6.0% of the atorvastatin group against 8.3% on placebo. The hazard ratio was 0.70, with a 95% confidence interval of 0.61 to 0.82.
The trial's primary endpoint was something else, and it did not move. STAREE was built to measure disability-free survival, meaning survival without death from any cause, dementia, or persistent physical disability. That composite reached 12.8% on atorvastatin and 13.6% on placebo, a hazard ratio of 0.94 with a confidence interval of 0.84 to 1.05 and a p value of 0.25.
Sophia Zoungas of the Monash School of Public Health and Preventive Medicine presented the results at the European Society of Cardiology Congress in Munich on August 29, 2026. They were published in the New England Journal of Medicine the following day. Zoungas said in the congress release that for older people worried about heart attack and stroke, "there is an effective measure for lowering that risk". The trial ran in the community in Australia through more than 3,400 general practitioners, according to MedicalXpress.
The short version
Why the Two Endpoints Disagree
Disability-free survival is a wider net than a cardiovascular event count, and most of what falls into it has nothing to do with cholesterol. Death from cancer, dementia of any cause, and physical decline from arthritis or frailty all register in that composite. A drug acting on one arterial mechanism can lower the cardiovascular share of that total and leave the rest untouched.
The size of each pool shows the dilution. The American College of Cardiology reports 637 disability-free survival events on atorvastatin against 676 on placebo, and 297 cardiovascular events against 412. The cardiovascular difference of 115 events sits inside a composite where the arms differ by 39. The other components either moved the other way or did not move at all.
Composition also explains why the benefit is smaller than the headline sounds. TCTMD reports the cardiovascular reduction was driven mainly by nonfatal events, especially heart attack and coronary revascularization, and that the trial was not powered for mortality on its own. Fewer procedures and fewer nonfatal heart attacks are a real result. They are a different claim from living longer.
What the Trial Recorded on Harms
Three harm figures were reported alongside the benefit. New-onset diabetes appeared in 4.0% of the atorvastatin group against 2.7% on placebo, per TCTMD. The same report puts the share who stopped the trial regimen because of an adverse event at 7.2%. The European Society of Cardiology reports serious adverse events at 2.7% in both groups. It also states that muscle, liver, and diabetes-related events were more common with atorvastatin.
The diabetes figure deserves to sit next to the benefit figure rather than in a footnote. A 1.3 percentage point increase in new diabetes against a 2.3 percentage point reduction in cardiovascular events is the actual trade this trial measured. Neither number is large. Both are the kind of thing a person weighing a daily tablet for the next decade would want in front of them.
Who Was Not in This Trial
STAREE excluded anyone with clinical cardiovascular disease, diabetes, or dementia at enrollment. That design choice is what makes it a primary prevention trial, and it is also the boundary on who the result describes. A 74-year-old with type 2 diabetes, or one who has already had a stent, was ineligible, so nothing here speaks to that person's treatment.
Mean LDL cholesterol at baseline was 127 mg/dL, and atorvastatin lowered it by 48 mg/dL against 16 mg/dL on placebo, per TCTMD. Anyone whose own LDL sits far above or below 127 mg/dL is outside the population that produced these rates. The trial also ran in Australia through general practice, which is a different care setting from a subscription platform ordering panels by mail.
What a Lipid Panel Result Can and Cannot Tell You
In the guides we publish here, the recurring gap is between a marker that moves and an outcome that changes, and STAREE measures both halves in the same population. LDL fell by 48 mg/dL, which is a large move on a panel any testing platform sells. Counted cardiovascular events fell by 2.3 percentage points. Independent survival did not change at all.
That sequence is worth holding onto when a longevity platform reports a cholesterol number back with a risk score attached. A panel can tell someone their LDL is 150 mg/dL. Our page on normal versus optimal ranges covers why the threshold a report uses changes what that number appears to mean. Our ApoB guide covers the particle measure that separates people whose standard panels look alike. Neither measurement, and no dashboard built on them, carries the years of counted events that produced the STAREE hazard ratios.
What Would Change This Read
Longer follow-up would change it first. Disability and dementia accrue slowly, and a cohort followed for a median of 5.9 years has not been observed through the period when those outcomes separate most. An extended analysis on the same participants could show a disability-free survival curve that starts to diverge, or confirm that it does not.
Component-level publication would change how the benefit is weighed. The distinction between fewer revascularization procedures and fewer heart attacks matters to a reader deciding what a daily tablet buys. The summaries released so far describe the mix without quantifying each part. Those figures sit in the New England Journal of Medicine paper, which was not retrievable at the time of writing. No number on this page is taken from it.
Related Coverage
This is the second cardiovascular prevention trial in a week to move a lipid number without moving an outcome. The RADICAL PC 2 referral trial raised statin use from 40% to 63% and lowered cholesterol, and found no difference in cardiovascular events over nearly six years. A Stanford secondary analysis covered here tied LDL increases on low-carb diets to a genetic score, again without outcome data. Anyone deciding what to do with a cholesterol result should take it to the clinician who can see the rest of the record. That clinician can weigh the 2.3 percentage point benefit and the 1.3 percentage point diabetes signal against a person's own risks. The biomarkers hub collects the markers that conversation usually starts from.
Sources
- European Society of Cardiology, "Cholesterol-lowering medication reduces major cardiovascular events by 30 per cent in older people without known cardiovascular disease," press release, ESC Congress 2026, Munich, August 29, 2026, escardio.org (accessed August 31, 2026). Source of the 9,971 randomized, the age and exclusion criteria, the atorvastatin 40 mg dose, the 5.9-year median follow-up, the 12.8% and 13.6% disability-free survival rates with the 0.94 hazard ratio and its interval, the 6.0% and 8.3% cardiovascular rates with the 0.70 hazard ratio and its interval, the 2.7% serious adverse event rate in both groups, and the Zoungas quotation.
- American College of Cardiology, "STAREE: Can Statin Therapy Help Prevent CV Events in Healthy, Community-Dwelling, Older Adults?", August 2026, acc.org (accessed August 31, 2026). Source of the 297 and 412 cardiovascular event counts and the 637 and 676 disability-free survival counts.
- TCTMD, "STAREE: Statins Cut MACE Risk by 30% in Older Adults With No CVD History," August 2026, tctmd.com (accessed August 31, 2026). Source of the 127 mg/dL baseline LDL, the 48 mg/dL and 16 mg/dL reductions, the 4.0% and 2.7% new-diabetes rates, the 7.2% discontinuation rate, the statement that the benefit was driven mainly by nonfatal events, and the statement that the trial was not powered for mortality alone.
- MedicalXpress, "Statins cut first heart attack and stroke risk 30% after age 70," August 30, 2026, medicalxpress.com (accessed August 31, 2026). Source of the Monash affiliation, the more than 3,400 participating general practitioners, and the New England Journal of Medicine publication date.
- New England Journal of Medicine, "Atorvastatin, Cardiovascular Events, and Disability-free Survival in Older Adults," DOI 10.1056/NEJMoa2607314, published August 30, 2026, nejm.org. Named as the publication venue by the sources above. The full text returned an access error at the time of writing, and no figure on this page comes from it.
Frequently Asked Questions
What was the primary endpoint of the STAREE trial?
Disability-free survival, defined as survival free of death from any cause, dementia, or persistent physical disability. It did not differ between the two groups. The European Society of Cardiology reports 12.8% of the atorvastatin group and 13.6% of the placebo group reaching that composite. The hazard ratio was 0.94, with a 95% confidence interval of 0.84 to 1.05 and a p value of 0.25. The 30% figure came from the cardiovascular endpoint, a different measure.
How large was the cardiovascular benefit in absolute terms?
The two published rates differ by 2.3 percentage points, 6.0% against 8.3%, over a median of 5.9 years. The American College of Cardiology reports the underlying counts as 297 events on atorvastatin and 412 on placebo. Dividing 100 by 2.3 gives about 44, so roughly 44 people would take atorvastatin for that period before one of them avoided an event. That last figure is arithmetic on the two published rates, not a number the trial reported.
Who was eligible for the trial?
Adults aged 70 and older with no history of clinical cardiovascular disease, no diabetes, and no dementia. That set of exclusions is what makes STAREE a primary prevention trial. It also means the result does not transfer to a 72-year-old with type 2 diabetes or a prior heart attack. Nobody in either arm had those conditions at enrollment.
What harms were recorded?
New-onset diabetes was recorded in 4.0% of the atorvastatin group against 2.7% on placebo, according to TCTMD. The same report puts the share who stopped the trial regimen because of an adverse event at 7.2%. The European Society of Cardiology reports serious adverse events at 2.7% in both groups, and states that muscle, liver, and diabetes-related events were more frequent with atorvastatin. Both figures belong in the same view as the 2.3 percentage point benefit.
Does this mean everyone over 70 should start a statin?
The trial does not answer that, and it was not designed to. It tested one drug at one dose in people who had none of the conditions that usually prompt a prescription. Its primary measure of staying alive and independent did not move. Age alone is a weak basis for the decision. The relevant inputs are life expectancy, other conditions, current medication load, and what a person wants from treatment. That is a conversation for a prescribing clinician holding the full record.
How much did LDL cholesterol change in the trial?
Mean low-density lipoprotein (LDL) cholesterol started at 127 mg/dL and fell by 48 mg/dL in the atorvastatin group against 16 mg/dL on placebo, as reported by TCTMD. That is a large separation between the arms, which is what makes the null primary endpoint informative rather than ambiguous. A trial that moved the marker only slightly would leave room to argue the dose was too low.