Cholesterol in older adults: what the debate is actually about

An earlier version of this article presented one side of a contested question as settled fact, and told readers that statin benefits were “grossly exaggerated.” That was not a fair summary of the evidence, and it has been rewritten.

What the review reported

In 2016 a group led by Uffe Ravnskov published a systematic review in BMJ Open covering 19 cohort studies — about 68,000 older people in total. The authors reported that in most of the cohorts they examined, high LDL cholesterol was not associated with higher all-cause mortality, and in a number of cohorts the association ran the other way: higher LDL, lower mortality.

Why it was criticized

Heavily, and by mainstream bodies. The main objections:

• Method. The Oxford Centre for Evidence-Based Medicine and others argued the review relied on aggregated published summaries rather than individual participant data, an approach prone to bias, and that its selection and analysis were not robust.
• It contradicts trial evidence. Observational associations between cholesterol level and death are not the same thing as testing what happens when you lower it. Randomized trials of cholesterol-lowering therapy consistently show reduced heart attacks and strokes, including in older participants.
• Reverse causation is a known trap here. Serious illness — cancer, frailty, advanced heart failure — lowers cholesterol. In an elderly cohort, low LDL can be a marker of being ill rather than a cause of dying, which can manufacture an apparent inverse association.

The diabetes trade-off is real, and it is on the label

The statin debate is often framed as “do they work or not,” which misses the more useful question of what they cost. One documented cost belongs in any honest discussion, and the first version of this article left it out entirely.

Statins raise the risk of new-onset type 2 diabetes. This is not a contested claim. In February 2012 the FDA added a safety label change to the entire statin class warning of increases in HbA1c and fasting blood glucose. A collaborative meta-analysis of randomized statin trials published in The Lancet in 2010 found roughly a 9 percent increase in incident diabetes, and a 2024 individual-participant-data meta-analysis in The Lancet Diabetes & Endocrinology confirmed effects on both new-onset diabetes and worsening glycemic control.

The risk is not evenly spread. It is higher with intensive-dose therapy and concentrated in people who already sit close to the diabetes threshold — which, in an older population, is a great many of them. Cardiology bodies including the American College of Cardiology hold that for patients at meaningful cardiovascular risk the benefit still outweighs this harm, and that remains the mainstream position. But “outweighs” is a trade-off, not an absence of harm, and a trade-off is something a patient is entitled to know they are making.

Deprescribing is a discipline, not a lapse

The most useful frame for an older adult is not whether statins work in general but whether this medication, at this dose, still earns its place on a list that may run to a dozen drugs. Geriatric medicine has formal tools for exactly that question.

• The AGS Beers Criteria, an expert-consensus list of medications that are often inappropriate in older adults because the harms outweigh the benefits at that age.
• The STOPP/START criteria, now in their third version (2023), covering drug-drug and drug-disease interactions, therapeutic duplication, medications that specifically raise falls risk, and treatments that are wrongly omitted.
• STOPPFrail, aimed at people with limited life expectancy, where the years needed for a preventive medication to pay off may exceed the years available.

None of these tools says “stop your statin.” What they do is make periodic review a normal part of care rather than something that only happens after a fall or an admission. If nobody has revisited a prescription in several years, that is worth raising — and it is a very different conversation from stopping the drug yourself.

What is genuinely unsettled

There is a real and legitimate clinical question underneath the controversy, and it is narrower than the headlines: how much benefit does starting a statin for primary prevention give someone in their late 80s with a limited life expectancy and a long list of other medications? Guidelines are genuinely more cautious there, and deprescribing is an accepted part of good geriatric practice.

That is a very different proposition from stopping a statin in someone who has already had a heart attack. “Is this still the right medication for me?” is a good question to bring to a clinician. “I read online that cholesterol does not matter” is not a plan.

Sources

• BMJ Open — Lack of an association or an inverse association between low-density-lipoprotein cholesterol and mortality in the elderly: a systematic review
• Centre for Evidence-Based Medicine, University of Oxford — Post-publication peer review of the above
• Science Media Centre — Expert reaction to the review
• American Heart Association — LDL and HDL cholesterol
• Sattar N et al., The Lancet 2010 — Statins and risk of incident diabetes: a collaborative meta-analysis of randomised statin trials
• The Lancet Diabetes & Endocrinology 2024 — Effects of statin therapy on new-onset diabetes and worsening glycaemia: an individual participant data meta-analysis
• American College of Cardiology — Statins and diabetes: rethinking the data
• American Geriatrics Society — AGS Beers Criteria for potentially inappropriate medication use in older adults