Cholesterol medication after 70: what are the benefits and limits?

After age 70, is it worth starting a statin to prevent a first cardiovascular event? The answer depends on each person’s health and life: the conditions they already have, the medications they take, their independence, their life expectancy and what they consider a priority.

The STAREE trial followed 9,971 people aged 70 or older and found a significant reduction in major cardiovascular events among those who received atorvastatin. But the treatment was not shown to extend life free of dementia or persistent physical disability. To assess this so-called disability-free survival, the researchers looked at the occurrence of death from any cause, dementia or persistent physical disability.

The results were published in the New England Journal of Medicine and presented at ESC Congress 2026, held by the European Society of Cardiology (ESC).

The study helps answer the question of prevention after 70. The decision to start treatment, however, remains an individual one.

How the STAREE trial evaluated atorvastatin

Conducted in Australia, STAREE was a randomized, double-blind, placebo-controlled trial. This means participants were assigned to groups by chance and that neither participants nor researchers knew who was receiving the drug or the placebo.

Participants were community-dwelling adults aged at least 70 without clinical cardiovascular disease, diabetes or dementia. They were randomly assigned to receive atorvastatin 40 mg a day or placebo.

The median follow-up was 5.9 years: half of the participants were followed for less time, and half for longer.

The study investigated primary prevention, in people without known clinical cardiovascular disease. Participants were generally independent and relatively healthy; the sample did not represent every profile of older adults.

What the reduction in cardiovascular events means

To assess the cardiovascular benefit, the researchers tracked a set of events: cardiovascular death, nonfatal heart attack, nonfatal stroke or coronary revascularization — a procedure to restore blood flow in the heart’s arteries. The analysis counted the first of these events in each participant. This is what researchers call a composite outcome.

The result favored atorvastatin. During follow-up, 297 participants in that group had a composite event, compared with 412 in the placebo group.

The rates were 10.9 and 15.5 events per 1,000 person-years, respectively. This measure takes into account both the number of participants and how long they were followed.

Among the 4,984 participants assigned to atorvastatin and the 4,987 assigned to placebo, the proportions with this event were approximately 6.0% and 8.3%. The crude difference was about 2.3 percentage points. These proportions show what happened during the study. Because follow-up time may vary among participants, they are not equivalent to a risk estimate adjusted for the same duration.

The analysis that considers the time to the first event found a hazard ratio (HR) of 0.70, with a 95% confidence interval of 0.61 to 0.82 and P<0.001. This corresponds to approximately a 30% relative reduction in the hazard — the instantaneous rate of the outcome among participants who have not yet had the event — over the follow-up period. NEJM.

The 30% refers to the combined set of events. It does not represent a 30% absolute reduction, nor the same reduction for each component.

When the components are examined separately, the favorable signal appears mainly in myocardial infarction and coronary revascularization. The study did not show a statistically significant reduction in stroke or cardiovascular death on their own. The results for each component, however, come from secondary analyses. Their confidence intervals were not adjusted for multiple comparisons; they therefore call for more caution than the main composite result.

What about life without dementia or disability?

STAREE also investigated whether the treatment would allow people to live longer without dementia or persistent physical disability. For this outcome, which included death from any cause, there was no statistically significant benefit.

The outcome occurred in 637 participants in the atorvastatin group and 676 in the placebo group: HR 0.94; 95% CI 0.84–1.05; P=0.25. The difference did not reach statistical significance. This does not definitively prove the absence of any effect, but neither does it allow anyone to promise greater longevity or dementia prevention. NEJM.

For cardiologist Elena Arbelo, the expert designated by the ESC to comment on the study, the two outcomes answer different questions. The first measures the occurrence of cardiovascular events; the second investigates a broader result involving survival, cognition and physical function.

“These findings are not contradictory,” says Arbelo, in remarks originally made in English.

Cardiologist Raul Dias, a member of the Brazilian Society of Cardiology (SBC) and former president of its Atherosclerosis Department, also notes that physical function in old age depends on multiple factors. In his view, lowering cholesterol will not necessarily produce a detectable difference in this broader outcome.

Independent geriatrician Paola Teruya points out that, at this stage of life, many causes can impair the ability to carry out everyday activities. Paola also raises the hypothesis that the follow-up may have been too short to show an effect on this outcome. It is a possibility, not a proven explanation: STAREE did not show a statistically significant improvement in disability-free survival.

Responding to Prime Health Report, Sophia Zoungas, a STAREE investigator, emphasized the need to consider each patient’s context.

“We believe it is important for older adults to discuss these findings with their doctor or health professional and make an informed decision based on their individual circumstances and priorities.”

According to Zoungas, although the study did not show an improvement in disability-free survival, it also found no evidence of harm, which she considers reassuring for older patients and their doctors.

Zoungas’s statement needs to be read alongside the safety data. It does not mean the treatment is free of adverse effects or that its safety is, overall, equivalent to that of placebo.

Adverse effects of atorvastatin in STAREE

Among participants who received at least one dose, serious adverse events considered by the study physician to be treatment-related occurred in 131 of 4,876 people on atorvastatin and 129 of 4,847 on placebo: 2.7% in both groups. These numbers refer to events classified as treatment-related, not to all health problems that occurred during follow-up.

In the same assessment of events considered treatment-related, the study recorded a higher frequency of musculoskeletal, hepatobiliary and diabetes-related events among participants who received atorvastatin. NEJM.

Discontinuation of treatment due to intolerance associated with adverse events was also slightly more frequent with the drug: approximately 7.2%, compared with 6.1% with placebo.

Over the years, there was also considerable discontinuation of the study medication, and participants originally assigned to placebo began using statins outside the protocol.

As a result, the difference between the treatments the groups actually used changed over time. This matters for interpreting the results, but it does not allow anyone to claim that the benefit of atorvastatin would have been greater if everyone had stayed on their assigned treatment.

Why the cardiovascular outcome changed

The originally planned cardiovascular outcome combined cardiovascular death, nonfatal heart attack and nonfatal stroke. During the trial, the researchers added coronary revascularization to the primary cardiovascular outcome.

The change was made before the database was locked and while the investigators remained blinded to how the results were distributed between the atorvastatin and placebo groups. According to the study documentation, the overall event rate was lower than expected. The change aimed to preserve the trial’s statistical power, that is, its ability to detect a difference between the groups.

Asked by PHR, Zoungas said the decision was made by the study’s steering committee and that the expanded cardiovascular outcome was chosen as primary well before the database was locked, because it was considered clinically relevant for older adults.

The original cardiovascular composite, without revascularization, also favored atorvastatin. This composite showed an HR of 0.73 (95% CI 0.62–0.87). It is, however, a secondary outcome, with an interval not adjusted for multiple comparisons. Its result reinforces the consistency of the cardiovascular signal; it is not equivalent to an additional primary confirmation.

Who do the STAREE results apply to?

To understand who the results apply to, it is important to go back to the profile of the participants. The mean age was 74.7 years, and people of very advanced age made up a smaller share of the sample. The trial also excluded people with clinical cardiovascular disease, diabetes or dementia.

Asked specifically about the possibility of extrapolating the results, Zoungas acknowledged the limitation:

“We highlight in the paper that these findings may not be generalizable to all population groups, and this should be considered when interpreting the results.”

STAREE, therefore, does not allow this balance of benefits and risks to be automatically extended to an 85- or 90-year-old with frailty, multiple conditions and several medications.

Paola illustrates this difference with two clinical scenarios. On one side, a robust, independent 85-year-old with few conditions, for whom prevention may make sense. On the other, someone of the same age with frailty, multiple conditions, dependence in daily activities and polypharmacy — the use of multiple medications. In the geriatrician’s view, STAREE’s design does not allow the results to be applied directly to this second profile. The comparison is a clinical example of individualized care; it does not show that the trial proved a specific benefit for 85-year-olds.

“In geriatrics, chronological age alone should not determine prescribing.”

She considers four dimensions: individual cardiovascular risk; physical function and the presence of frailty or pre-frailty; life expectancy; and the patient’s priorities. The greater the frailty, the number of conditions and the number of medications, she explains, the more important it is to assess whether the expected benefit makes sense given that person’s estimated life span and goals.

Could STAREE change statin recommendations?

STAREE adds randomized evidence on a population with limited representation in cardiovascular trials. Its incorporation into recommendations, however, depends on evaluation by medical societies.

Arbelo believes the results have the potential to strengthen the evidence considered in future European recommendations. She argues that any such incorporation should preserve the assessment of cardiovascular risk, frailty, coexisting conditions, life expectancy and patient preferences. Her assessment is neither an announcement of a guideline change nor a formal institutional position of the ESC.

For Brazilian practice, there is another question: how should the results of an Australian study be applied to patients treated under different conditions?

For Dias, the study adds relevant evidence for cardiovascular prevention and may prompt reassessments of the recommendations of the SBC and other medical societies. His answers were sent through the SBC’s press office, with attribution to the cardiologist authorized. The expectation he expressed is not equivalent to a revision already approved by the organization.

Age alone does not decide

In adults aged 70 or older similar to the trial participants, atorvastatin reduced major cardiovascular events.

But the study did not show that this benefit translates, during the follow-up period, into a statistically significant extension of life free of dementia or persistent physical disability.

Nor did it show that everyone above a certain age should receive the drug.

The question left for the doctor’s visit is: for this person over 70, does the expected cardiovascular benefit justify starting and staying on long-term preventive treatment?

Paola argues that this conversation should include the patient: the expected benefit needs to be weighed against the burden of treatment in daily life and the person’s life goals. STAREE offers more information for this shared decision.

_____________

Funding and transparency

STAREE was funded by the National Health and Medical Research Council, the Heart Foundation of Australia and Monash University. According to the paper and appendix, the trial had no commercial or industry involvement. Zoungas disclosed consulting work for pharmaceutical companies; the disclosure form reports payments to Monash University for four of these relationships and to The George Institute for the relationship with Sanofi. The disclosure regarding AstraZeneca does not specify the recipient of the payment. Trial funding and individual relationships are separate pieces of information; these ties, by themselves, do not demonstrate any interference in the results. NEJM paper and supplementary documents.

References and sources

1. Main paper: Zoungas S, Wolfe R, Moran C, et al.; STAREE Investigators. Atorvastatin, Cardiovascular Events, and Disability-free Survival in Older Adults. New England Journal of Medicine. Published online August 29, 2026. DOI: 10.1056/NEJMoa2607314. PubMed abstract.

2. Supplementary documentation for the paper: supplementary appendix, protocol and statistical analysis plan, conflict-of-interest disclosures and data-sharing statement. Materials associated with the NEJM paper, consulted in reporting this story.

3. Protocol published in 2023: Zoungas S, Curtis A, Spark S, et al. Statins for extension of disability-free survival and primary prevention of cardiovascular events among older people: protocol for a randomised controlled trial in primary care (STAREE trial). BMJ Open. 2023;13:e069915. DOI: 10.1136/bmjopen-2022-069915. This document describes the plan prior to the outcome changes discussed in this story; the final version was checked against the NEJM paper’s documentation.

Sources interviewed by PHR: Sophia Zoungas, STAREE investigator; Elena Arbelo, expert designated by the ESC; Raul Dias, member of the SBC and former president of its Atherosclerosis Department; and Paola Teruya, independent geriatrician. The comments were provided to Prime Health Report for this story. The assessments by Arbelo and Dias are attributed to the experts and do not, in themselves, constitute institutional positions of the ESC or the SBC.