Design: Randomised, double-blind, placebo-controlled trial in general practice
Population: 9,971 adults aged ≥70 without cardiovascular disease, diabetes or dementia
Primary outcome: Major cardiovascular events; death, dementia or persistent disability
Effect: MACE HR 0.70 (95% CI 0.61–0.82); disability-free survival HR 0.94 (0.84–1.05)

STAREE, published in NEJM on 29 August 2026, randomised 9,971 community-dwelling Australians aged 70 or older with no cardiovascular disease, diabetes or dementia to atorvastatin 40 mg or placebo, recruited through general practices. Mean age was 75 and half were women.

Over a median 5.9 years, major cardiovascular events fell from 15.5 to 10.9 per 1,000 person-years (HR 0.70). The second primary endpoint, a composite of death, dementia or persistent physical disability, did not differ (HR 0.94). Serious adverse events were equal, though musculoskeletal, liver-related and diabetes-related events were more common with atorvastatin.

Until now, statin advice above 70 in primary prevention leaned on extrapolation and subgroup data. This trial gives a direct answer: statins prevent heart attacks and strokes in this age group, but over six years that did not translate into longer life free of dementia or disability. That makes the decision one of preference: a patient who fears a stroke has good evidence to start; one whose priority is staying independent should know the trial did not show that benefit.

Offer a statin for primary prevention to fit adults over 70 who want to reduce heart attack and stroke risk.
Tell patients the benefit is about five fewer cardiovascular events per 1,000 people each year.
Explain that over six years the statin did not lengthen life free of dementia or disability.
Check glucose periodically, as diabetes-related events were more frequent on atorvastatin.

Why it matters

It replaces extrapolation with trial evidence for the commonest preventive prescription in older adults.

The statistics, in plain English

The hazard ratio of 0.70, with an interval of 0.61 to 0.82, is a robust 30% relative reduction. In absolute terms it is 4.6 fewer events per 1,000 people per year. The disability-free survival result (0.94, interval 0.84 to 1.05) crosses 1.0, so no effect on that outcome was shown.