Of the fourteen modifiable dementia risk factors the Lancet Commission names, raised LDL cholesterol is the odd one out: most adults over forty have already had it measured. The number is sitting in a GP record. Almost nobody has been told it has anything to do with their brain.
We have covered the vascular cluster before: blood pressure, the ten factors behind stroke risk, what the Commission's 45% figure does and doesn't mean. This one is narrower and easier to get backwards: the timing. The evidence points at LDL measured in midlife. Measured in late life, the same association weakens, flattens, sometimes reverses. Get that wrong and you conclude either that cholesterol has nothing to do with dementia, or that low cholesterol in an eighty-year-old is a warning sign. Neither survives the papers.
What changed in 2024
LDL was not on the 2020 list, which modelled twelve factors with the potential to prevent an estimated 40% of dementia cases. The 2024 report added two — uncorrected vision impairment and high LDL cholesterol — putting the combined population attributable fraction for all fourteen at 45.3%. The Commission says the earlier evidence "was inconclusive"; what changed was volume, as several very large UK primary-care cohorts arrived in between. Its verdict now: "high quality consistent, biologically plausible, evidence that high LDL cholesterol in midlife is a risk factor for dementia," and a one-line recommendation — detect and treat it from midlife.
We are not printing a per-factor percentage for LDL, because the Commission undercuts its own: it "could not find a worldwide estimate of prevalence of LDL cholesterol" and concedes "the use of an estimate from a single cohort study is not ideal." An attributable fraction is a relative risk times a prevalence, so LDL's slice is the shakiest in the model even where the combined 45.3% holds. The factor-by-factor breakdown is for relative importance, not decimal points.
The midlife signal
The clearest demonstration is also the oldest. A 2017 meta-analysis by Anstey, Ashby-Mitchell and Peters pooled seventeen cohort studies in 23,338 participants. High total cholesterol in midlife carried a relative risk of 2.14 (95% CI 1.33–3.44) for late-life Alzheimer's disease. High total cholesterol in late life was associated with none of it — not Alzheimer's, vascular dementia, any dementia, mild cognitive impairment or cognitive decline. Same exposure, same outcome, opposite conclusions depending on when the blood was taken. The pattern held for LDL too: a 2023 review by Wee and colleagues put each 1 mmol/L rise at an 8% higher incidence of all-cause dementia (1.08, 1.03–1.14).
Iwagami and colleagues' 2021 analysis of UK primary care records is the most informative: 1,853,954 people with a first cholesterol measurement, 953,635 with an LDL value, median 7.4 years of follow-up. Split by age, it separates cleanly. Under 65 at baseline, the rate ratio per standard deviation of LDL (1.01 mmol/L, or 39 mg/dL) was 1.10 (1.04–1.15) for dementia diagnosed within ten years and 1.17 (1.08–1.27) for dementia diagnosed more than ten years later. At 65 or over, both were weaker: 1.03 (1.01–1.05) and 1.07 (1.03–1.13). Note the gradient — the association got stronger with longer follow-up, the wrong shape for reverse causation.
Why late-life readings mislead
Cholesterol falls in the years before a dementia diagnosis, so a snapshot taken in someone's seventies can make low cholesterol look protective when it is an early consequence of the disease. In the Honolulu-Asia Aging Study, 1,027 men had total cholesterol assayed five times between 1965 and 1993. Among the 56 who developed dementia — particularly Alzheimer's — cholesterol had already declined at least fifteen years before diagnosis and stayed lower, a difference that survived adjustment for vascular risk factors, weight change, alcohol and lipid-lowering drugs.
The same logic explains a large null that gets quoted as if it refuted the whole idea. An individual participant meta-analysis by Peters and colleagues combined eight studies and over 21,000 people aged 60 and over, averaging 76, and found no relationship between total, HDL or LDL cholesterol and later cognitive decline or incident dementia — for total cholesterol, an odds ratio of 1.01 (0.89–1.13) for incident dementia, unchanged by statin use or APOE ε4 status. A genuine null, and no contradiction: the authors call for further work "in younger populations to understand the role of cholesterol across the life-course."
What Mendelian randomisation can and can't settle
Mendelian randomisation uses inherited variants as proxies for lifelong exposure, sidestepping confounding and reverse causation in principle. It has not delivered a clean answer here. The MR evidence the Commission cites is a 2020 meta-analysis of 27 studies, 3,136 people with dementia and 3,103 controls, concluding that high total cholesterol and reduced HDL "might be potential risk factors" for Alzheimer's. Its instrument was APOE genotype — and APOE is the largest common genetic risk factor for Alzheimer's in its own right, separately from anything it does to lipids. Textbook horizontal pleiotropy: close to the worst available instrument here.
Lipid-specific instruments are more reassuring and less decisive. In 111,194 Danes, Benn and colleagues used PCSK9 and HMGCR variants that lowered LDL by 9.3%: per lifelong 1 mmol/L lower LDL, risk ratios of 0.57 (0.27–1.17) for Alzheimer's and 0.66 (0.34–1.26) for any dementia — right direction, confidence intervals crossing 1.0, not significant. A broader 380-variant instrument did reach significance, at 0.64 (0.52–0.79).
MR applied to dementia should keep everyone humble. A 2015 study found genetically predicted higher systolic blood pressure associated with lower Alzheimer's risk (OR 0.75, 0.62–0.91), and heavier smoking likewise (0.67, 0.51–0.89). Nobody believes either causally. A 2026 review of 53 MR studies of vascular dementia found causal effects for cholesterol "inconsistent depending on the study," with only 38% of those claiming one having completed the STROBE-MR checklist.
Statins: two honest findings, pointing different ways
Statins started in late life do not protect cognition. A Cochrane review found only two eligible randomised trials — HPS (20,536 people on simvastatin, five years) and PROSPER (5,804 aged 70–82 on pravastatin, 3.2 years). Only HPS reported dementia incidence: 31 cases in each arm, odds ratio 1.00 (0.61–1.65). Across five cognitive tests there was no difference from placebo. The reviewers are blunt: "there is good evidence that statins given in late life to people at risk of vascular disease do not prevent cognitive decline or dementia."
Observational data disagree: a 2022 meta-analysis found statin users at lower risk of dementia (36 studies, OR 0.80, 0.75–0.86) and Alzheimer's (21 studies, 0.68, 0.56–0.81). But heterogeneity was extreme, above 90% on the Commission's reading, and indication bias is the obvious explanation. The Commission's own resolution is that long-term trials of statins for dementia prevention "would be unethical and impractical" — so the midlife question has never been tested, and probably never will be.
The other direction matters for anyone frightened off. The FDA added a cognitive warning to statin labels in 2012, on adverse-event reports and a literature review. A 2015 systematic review then pooled 25 placebo-controlled randomised trials in 46,836 people, with test data from 27,643 participants, and found no adverse effect in cognitively normal people (standardised mean difference 0.01, 95% CI −0.01 to 0.03, p=0.42) or in people with Alzheimer's (−0.05, −0.19 to 0.10, p=0.38). Both are nulls, not benefits: the scare did not hold up, and the drugs did not help cognition. A trial of evolocumab added to statins in 1,204 patients, pushing LDL lower still, found no cognitive difference over a median 19 months.
What to do with this
Know the number, and know what it is. NICE defines a full lipid profile as measuring total cholesterol, HDL and triglycerides and then calculating non-HDL and LDL cholesterol, no fasting required — so "my LDL" is usually a derived figure, and may not be reported when triglycerides are high. The number alone decides nothing either: NICE directs clinicians to estimate ten-year cardiovascular risk with QRISK3 for adults aged 25 to 84 without cardiovascular disease, and to review those estimates for people over 40. Treatment comes out of that whole picture — which is why we are not printing a target.
Using apolipoprotein B is consistent with European guidance: the 2019 ESC/EAS dyslipidaemia guidelines recommend apoB for risk assessment "particularly in people with high TG, diabetes, obesity or metabolic syndrome, or very low LDL-C," and say it can replace LDL cholesterol where available. Note the grading — Class I recommendation, Level C evidence: strong advice resting on consensus, not trial data. ApoB counts particles rather than the cholesterol inside them.
We sell nothing for this, including our fish oil
Proco sells no product that addresses raised LDL cholesterol, and our Omega-3 is not a lipid treatment. The 2020 Cochrane review of omega-3 found little or no effect on blood lipids, with one exception: a dose-dependent fall of roughly 15% in triglycerides. Triglycerides are not LDL. We have covered why two major reviews of omega-3 and heart health disagreed and what doses the trials used; one softgel of ours provides 180mg EPA and 120mg DHA. Reading that as cholesterol management is a category error. And if your worry is memory rather than lipids, cheap blood tests cover the reversible causes — most memory worries are not dementia.
What moves this factor is diet, exercise and, where indicated, prescription treatment — the list that outperforms supplements across brain health. LDL travels with insulin resistance, and as with hearing loss, the observational association is far stronger than the evidence that treating it late changes cognition.
The summary is short. High LDL in midlife is associated with more dementia decades later, consistently and across millions of people. Whether lowering it in midlife prevents dementia has not been tested and probably cannot be; lowering it in late life appears neither to help cognition nor harm it. The most useful line here is the dullest: find out your number, and discuss it with someone qualified to interpret it.
This article is information, not medical advice, and cannot assess individual risk. Decisions about cholesterol testing, targets and lipid-lowering treatment belong with your doctor — do not change medication on the basis of an article. Proco products are not intended to diagnose, treat, cure or prevent any disease.
Sources: Livingston G, Huntley J, Liu KY, et al. Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. Lancet. 2024;404(10452):572–628. DOI 10.1016/S0140-6736(24)01296-0 · Anstey KJ, Ashby-Mitchell K, Peters R. Updating the evidence on the association between serum cholesterol and risk of late-life dementia: review and meta-analysis. Journal of Alzheimer's Disease. 2017;56(1):215–228. DOI 10.3233/JAD-160826 · Wee J, Sukudom S, Bhat S, et al. The relationship between midlife dyslipidemia and lifetime incidence of dementia: a systematic review and meta-analysis of cohort studies. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring. 2023;15(1):e12395. DOI 10.1002/dad2.12395 · Iwagami M, Qizilbash N, Gregson J, et al. Blood cholesterol and risk of dementia in more than 1.8 million people over two decades: a retrospective cohort study. Lancet Healthy Longevity. 2021;2(8):e498–e506. DOI 10.1016/S2666-7568(21)00150-1 · Stewart R, White LR, Xue QL, Launer LJ. Twenty-six-year change in total cholesterol levels and incident dementia: the Honolulu-Asia Aging Study. Archives of Neurology. 2007;64(1):103–107. DOI 10.1001/archneur.64.1.103 · Peters R, Xu Y, Antikainen R, et al. Evaluation of high cholesterol and risk of dementia and cognitive decline in older adults using individual patient meta-analysis. Dementia and Geriatric Cognitive Disorders. 2021;50(4):318–325. DOI 10.1159/000519452 · Zhang X, Tian Q, Liu D, et al. Causal association of circulating cholesterol levels with dementia: a Mendelian randomization meta-analysis. Translational Psychiatry. 2020;10(1):145. DOI 10.1038/s41398-020-0822-x · Benn M, Nordestgaard BG, Frikke-Schmidt R, Tybjærg-Hansen A. Low LDL cholesterol, PCSK9 and HMGCR genetic variation, and risk of Alzheimer's disease and Parkinson's disease: Mendelian randomisation study. BMJ. 2017;357:j1648. DOI 10.1136/bmj.j1648 · Østergaard SD, Mukherjee S, Sharp SJ, et al. Associations between potentially modifiable risk factors and Alzheimer disease: a Mendelian randomization study. PLoS Medicine. 2015;12(6):e1001841. DOI 10.1371/journal.pmed.1001841 · Zuo M, Gagliano Taliun SA. A systematic review of risk factors for vascular dementia using Mendelian randomization: a call for methodologically robust studies to advance knowledge. BMC Neurology. 2026;26:359. DOI 10.1186/s12883-026-04907-4 · McGuinness B, Craig D, Bullock R, Passmore P. Statins for the prevention of dementia. Cochrane Database of Systematic Reviews. 2016;(1):CD003160. DOI 10.1002/14651858.CD003160.pub3 · Olmastroni E, Molari G, De Beni N, et al. Statin use and risk of dementia or Alzheimer's disease: a systematic review and meta-analysis of observational studies. European Journal of Preventive Cardiology. 2022;29(5):804–814. DOI 10.1093/eurjpc/zwab208 · Ott BR, Daiello LA, Dahabreh IJ, et al. Do statins impair cognition? A systematic review and meta-analysis of randomized controlled trials. Journal of General Internal Medicine. 2015;30(3):348–358. DOI 10.1007/s11606-014-3115-3 · Giugliano RP, Mach F, Zavitz K, et al. Cognitive function in a randomized trial of evolocumab (EBBINGHAUS). New England Journal of Medicine. 2017;377(7):633–643. DOI 10.1056/NEJMoa1701131 · Abdelhamid AS, Brown TJ, Brainard JS, et al. Omega-3 fatty acids for the primary and secondary prevention of cardiovascular disease. Cochrane Database of Systematic Reviews. 2020;(3):CD003177. DOI 10.1002/14651858.CD003177.pub5 · Mach F, Baigent C, Catapano AL, et al. 2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. European Heart Journal. 2020;41(1):111–188. DOI 10.1093/eurheartj/ehz455 · National Institute for Health and Care Excellence. Cardiovascular disease: risk assessment and reduction, including lipid modification (NG238). 2023.