Statins After 70: What the STAREE Trial Changes

After 70, a statin can prevent a first cardiovascular event in some people; age alone does not make the decision. The STAREE trial provides the randomised evidence that was missing. Atorvastatin 40 mg daily reduced cardiovascular events in independent older adults without clinical cardiovascular disease, diabetes or dementia. It did not, however, prolong disability-free survival.

Both findings matter. They support a real benefit without turning it into a promise of living longer or remaining independent. They also require a clear distinction between primary prevention—preventing a first event—and secondary prevention after a heart attack, stroke, arterial disease or revascularisation, where the treatment case is different.

What STAREE actually studied

STAREE was a double-blind, randomised, placebo-controlled trial based in Australian general practices. It assigned 4,984 participants to atorvastatin and 4,987 to placebo. Women made up 51.9% of the cohort, and mean baseline LDL-C was 126 mg/dL (3.27 mmol/L).

The intervention was specific: atorvastatin 40 mg daily. The trial did not compare different statins, lower doses, ezetimibe or strategies guided by ApoB or coronary calcium. It also excluded people with diabetes, dementia or previous clinical cardiovascular disease. Those boundaries belong in every interpretation of “statins after 70.”

The trial was supported mainly by Australia’s National Health and Medical Research Council, the Heart Foundation of Australia and Monash University. The project states that it did not accept pharmaceutical-company funding; investigators’ external relationships were disclosed separately.

The result in numbers: relative and absolute benefit

The primary cardiovascular outcome combined cardiovascular death, nonfatal myocardial infarction, stroke or coronary revascularisation. It occurred in 297 people assigned atorvastatin and 412 assigned placebo: 10.9 versus 15.5 events per 1,000 person-years, HR 0.70 (95% CI, 0.61–0.82; p<0.001).

The 30% figure is a relative reduction. Dividing events by participants provides an approximate clinical translation: 6.0% versus 8.3%, a crude difference of 2.3 percentage points over a median 5.9 years. That is around 23 fewer events per 1,000 people treated, or a number needed to treat close to 44. This is an unadjusted approximation; an individual’s absolute benefit may be larger or smaller depending on baseline risk and actual treatment duration.

Outcome	Atorvastatin	Placebo	Practical reading
Primary cardiovascular event	297; 10.9/1,000 person-years	412; 15.5/1,000 person-years	HR 0.70; clear cardiovascular benefit in the studied population
Death, dementia or persistent disability	637; 21.6/1,000 person-years	676; 23.0/1,000 person-years	HR 0.94; no statistically significant difference
Serious adverse events	131 (2.7%)	129 (2.7%)	Similar overall frequency; specific harms still need monitoring

What STAREE did not show

The other primary outcome—death from any cause, dementia or persistent physical disability—occurred in 637 participants assigned atorvastatin and 676 assigned placebo: HR 0.94 (95% CI, 0.84–1.05; p=0.25). This was not a statistically significant difference. About 80% of deaths were non-cardiovascular, so preventing cardiovascular events did not move this broader endpoint during follow-up.

That does not make the cardiovascular result unimportant: preventing a heart attack, stroke or revascularisation may matter greatly. The honest offer is simply more precise: fewer cardiovascular events, not proof of longer independent life.

One methodological limit deserves visibility. The original cardiovascular composite included cardiovascular death, myocardial infarction and stroke. After a blinded review found fewer events than expected, the committee added coronary revascularisation to preserve statistical power; the original three-point outcome became secondary. The change was documented before final analysis, but it makes the components—not only the composite headline—especially important.

Who resembles the trial population—and who does not

Participants lived independently and were healthy enough to join a long trial. STAREE directly informs whether to start treatment in a functional older adult without diabetes, dementia or known cardiovascular disease. It is less direct for:

• marked frailty, functional dependence or complex multimorbidity;
• limited life expectancy or priorities focused on reducing treatment burden;
• diabetes, advanced kidney disease or familial hypercholesterolaemia;
• previous heart attack, stroke, peripheral artery disease or revascularisation;
• established dementia;
• people unable to tolerate atorvastatin 40 mg or with important interactions.

An average result cannot replace baseline risk. A useful assessment brings together blood pressure, smoking, kidney health, LDL-C history, ApoB when particle number adds information, lipoprotein(a) as a largely inherited risk modifier, family history and, when uncertainty remains, validated risk tools or selective coronary calcium scoring.

Decision table: starting, continuing or reviewing a statin

Situation	What STAREE adds	Reasonable next step
Age ≥70, independent, no clinical CVD, diabetes or dementia; meaningful cardiovascular risk	Supports atorvastatin 40 mg to reduce a first event	Estimate absolute benefit, review interactions and decide together
Uncertain risk and reluctance to add medication	Does not mandate treatment by age	Review total risk and modifiers; use imaging only if it can change the decision
Previous MI, stroke, arterial disease or revascularisation	Outside the STAREE population	Use secondary-prevention guidance; do not stop by extrapolating this article
Diabetes, dementia, marked frailty or multimorbidity	Direct evidence is limited or absent	Individualise around goals, function, comorbidity and time to benefit
A statin has been tolerated for years	STAREE studied initiation, not withdrawal	Review indication and response; do not stop automatically because of age
Muscle symptoms or a drug interaction	Does not prove every symptom is statin-caused	Assess cause, CK when appropriate, dose, another statin or an alternative

Harms, interactions and follow-up

Serious adverse events were equal overall: 2.7% in both groups. Musculoskeletal, hepatobiliary and diabetes-related events were somewhat more common with atorvastatin. The absolute differences reported were small, but they can still matter to an individual.

Before treatment, guidance such as NICE recommends checking a lipid profile, diabetes status, kidney function, liver transaminases, blood pressure, alcohol intake, current medicines and previous muscle symptoms; TSH is tested when thyroid disease is suspected. Creatine kinase is not a routine test for someone without symptoms, but it can be useful with unexplained pain, tenderness or weakness. After starting or changing a dose, lipids and tolerability are reviewed at 4–12 weeks—or 2–3 months under some local guidance—and periodically thereafter.

Atorvastatin can interact with certain antibiotics and antifungals, antivirals, immunosuppressants and other medicines; grapefruit may also matter at some intakes. A personal medication review by a pharmacist or prescriber is more useful than memorising a generic ban list.

Starting and stopping are different questions

STAREE tested initiation. It does not show what happens when a statin taken for years is withdrawn. The 2026 French SAGA/SITE trial studied adults aged at least 75 already receiving a statin for primary prevention: stopping was non-inferior for three-year mortality, but its design, population and endpoint differed, so it does not make deprescribing automatic.

A useful review reconstructs the original indication, current risk, harms, medication burden, frailty and personal priorities. Chronological age can start the conversation; it should not finish it.

What still matters beyond a statin decision

A statin does not replace blood-pressure control, stopping smoking, adapted movement, maintaining strength and cardiorespiratory capacity, adequate sleep or a realistic Mediterranean-style eating pattern. Nor does it replace follow-up: value comes from measuring, interpreting, acting and checking again.

At Progevita, a cardiovascular assessment can integrate history, function, medication, blood pressure, metabolism and biomarkers without turning one number or birthday into an automatic prescription. The aim is to identify which decision offers value now and how to verify that it continues to do so.

Sources

1. Zoungas S et al. Atorvastatin, Cardiovascular Events, and Disability-free Survival in Older Adults. New England Journal of Medicine. 2026.
2. Monash University. STAREE design, funding and headline results.
3. European Society of Cardiology. Official STAREE presentation, ESC Congress 2026.
4. STAREE statistical analysis plan and documented cardiovascular-endpoint change.
5. 2025 ESC/EAS Focused Update on dyslipidaemia management.
6. 2026 ACC/AHA multisociety dyslipidaemia guideline.
7. NICE. Statin assessment, safety and follow-up recommendations.
8. 2026 SAGA/SITE trial of statin discontinuation in primary prevention.

Method: narrative review of the primary trial, protocol, statistical plan and current guidelines; Spanish and English search results checked on 23 September 2026. The NNT is an approximation calculated from crude proportions, not an adjusted trial estimate. This article is educational and does not replace medical assessment.