Making the Statins Mess Clear, Once and For All

The science behind statins is not as clear-cut as many would like you to think.

(NB: I am grateful for Dr. Brownstein’s continuous objectivity on the true status of the science on statins. He raises the bar above rhetoric. Here, he joins me in an earnest attempt to describe precisely why doctors who think statins are generally useful or even useful in most cases of cardiovascular disease think the way they do based on the available studies.-JLW)

This article does not argue that statins lack efficacy in high-risk secondary prevention. It argues that the magnitude of benefit is modest in absolute terms, that safety claims must be comparator-stratified, and that extrapolation into low-risk populations exceeds the evidence base. We stand by these estimates of harm based on RCT-level evidence:

• New-onset diabetes: NNH ~100–200 over 4–5 years (lower with high intensity).

• Mild muscle symptoms: NNH ~100 in RCT data.

• Severe myopathy/rhabdomyolysis: NNH >10,000.

• Liver enzyme elevation: NNH ~100–200.

As always, this is not medical advice. Check with your doctor before starting or stopping any medication. Again, do not start or stop any prescribed medication without consulting with the responsible physician.

-Dr. James Lyons-Weiler (PhD) and David Brownstein, MD

Author: The Statin Disaster, www.drbrownstein.com, Blog: Dr Brownstein’s Holistic Medicine

The claim that statins are safe and broadly life-saving rests on a foundational set of clinical trials and meta-analyses conducted in highly selected, high-risk populations. What followed, however, was a widespread extrapolation of these results to low-risk, diet-optimized, and drug-naïve individuals, in whom no outcome trial has ever been conducted. This article, exclusive to Popular Rationalism, reconstructs what was truly shown in the original trials, exposes how comparators were misused and misrepresented in subsequent analyses, identifies the limits of generalizability to the broader population, and calls for a recalibration of risk-benefit claims based on actual evidence.

What Was Found: The Original Placebo-Controlled Trials

The 4S (Scandinavian Simvastatin Survival Study, Lancet 1994;344:1383-1389) enrolled 4,444 patients aged 35–70 with coronary heart disease (CHD) and baseline total cholesterol 5.5–8.0 mmol/L(213mg/dl-309 mg/dl). Participants were randomized to simvastatin 20–40 mg/day vs placebo. After a median follow-up of 5.4 years, the simvastatin group showed a 30% relative reduction in all-cause mortality (RR 0.70, 95% CI 0.58–0.85), with 256 deaths in the placebo group vs 182 in the treatment group. The absolute risk reduction (ARR) was 1.67% per year.

(NB: UPDATE from original): Absolute risk reduction (ARR) refers to the difference in cumulative event rates between treatment and control groups. It differs from annualized hazard rates sometimes reported in secondary analyses. For decision-making, cumulative ARR over the study duration is the relevant measure. Over a 5.4 year follow-up period, the placebo group mortality rate = 256 deaths / 2,223 patients ≈ 11.5% The Simvastatin group mortality rate = 182 deaths / 2,221 patients ≈ 8.2%. The ARR over that time period was 3.3% over 5.4 years. This was a secondary prevention population with established coronary heart disease and elevated cholesterol levels in a pre-modern background treatment era. The 3.3% ARR applies to that population over 5.4 years. It does not apply to low-risk individuals or to contemporary polypharmacy environments without direct evidence. Thank you, ZH).

WOSCOPS (West of Scotland Coronary Prevention Study, N Engl J Med 1995;333:1301-1307) randomized 6,595 men (45–64 years) with hypercholesterolemia (LDL >155 mg/dL) and no prior MI to pravastatin 40 mg or placebo. After 4.9 years, pravastatin reduced the risk of definite MI or CHD death by 31% (relative risk reduction 0.69, 95% CI 0.54–0.87), with an absolute risk reduction of 0.9%/year. Notably, this was a high-LDL, moderate-risk primary prevention group.

The Heart Protection Study (HPS, Lancet 2002;360:7-22) randomized 20,536 high-risk individuals (CHD, other vascular disease, or diabetes) to simvastatin 40 mg vs placebo for 5 years. Statin use reduced major vascular events by 24% (relative risk reduction 0.76, 95% CI 0.72–0.81), and an absolute risk reduction of 5.4% and all-cause mortality reduced by 14% (relative risk) and an absolute risk reduction of 1.8%. This study had consistent effects across subgroups. VERY Importantly, these were not healthy individuals: 16% had prior MI, 33% had diabetes, and 65% were on antihypertensives.

These trials established that statins reduce cardiovascular events in high-risk patients. However, they were conducted in populations with substantial background pharmacotherapy, baseline dyslipidemia, and elevated absolute risk. Furthermore, there was a vast difference between relative risk reduction and absolute risk reduction in these statin studies as well as every other statin study conducted to date.

What They Did NOT Show

None of the foundational trials—4S, WOSCOPS, HPS, or ASCOT—tested statins in individuals with low baseline cardiovascular risk, optimized diet, minimal comorbidities, or absence of other medications. There were no arms comparing statins to intensive lifestyle change, structured low-carbohydrate or low-sugar nutrition, or exercise-only protocols. The populations were not representative of integrative-care patients or healthy middle-aged adults considering statins for marginal LDL elevation. These trials also did not track long-term subclinical harms such as mitochondrial dysfunction, nutrient depletion, coronary calcification, or progressive metabolic disruption. Nor did they stratify outcomes by pre-existing nutritional status, polypharmacy burden, or fitness baseline. Thus, while the trials showed that statins reduce cardiovascular events in high-risk, pharmacologically managed patients, ARR was miniscule from 1-3% in almost every statin study meaning 99-97% arguably received no discernable benefit. Absolute benefit is concentrated in those with the highest baseline risk; in low-risk populations, NNT becomes very large. They also provide no evidentiary basis for prescribing statins to low-risk, health-oriented populations, or people who moved into the low-risk, health-oriented populations—yet it is precisely in this population that prescription rates continue to rise. The result is a dangerous illusion of universality drawn from data that never supported it.

What Comparisons Were Made, and What They Actually Show

Trials comparing statins to placebo in high-risk populations (e.g., 4S, HPS) establish efficacy in those groups. But later studies shifted to different comparators:

● Active-comparator trials (e.g., statin vs niacin, statin vs fibrate) only tell us about relative efficacy and safety between drugs — not absolute safety.

● Usual care comparisons introduce uncontrolled background therapy, which often includes antihypertensives, aspirin, and diabetes medications. Comparing “statin + usual care” to “usual care” dilutes attribution of harm.

● Non-inferiority trials between statins (e.g., high- vs low-dose atorvastatin) cannot reveal absolute risks or benefits versus placebo.

Despite this, meta-analyses often pooled these results together, inflating N sizes while obfuscating which populations and comparators were used. The 2016 CTT Collaboration meta-analysis (Lancet 2016;388:2532–60) combined 26 RCTs but mixed placebo and active comparators without stratification. To date, the CTT Collaboration has refused to allow access to de-identified data for verification of their findings. Withholding de-identified data for independent review should disqualify a study from not only publication but also from setting guidelines based on that data.

This practice of mixing comparators destroys interpretive clarity. Pooling results from studies that use different populations also blunts and confuses (prevents) actionable clinical knowledge. Statin vs placebo trials allow for causality inference regarding both benefit and harm. Statin vs drug comparisons can only reveal whether one active treatment is better than another. Statin vs usual care embeds multiple confounders. Pooling these together as if they provide the same evidentiary weight is methodologically indefensible and unsound.

Over time, the control conditions in statin trials have shifted in ways that materially affect the apparent magnitude of benefit. Early landmark trials such as 4S (1994), WOSCOPS (1995), and HPS (2002) compared statins to true placebo in high-risk populations with relatively limited background lipid therapy. As statins became standard of care, later trials increasingly compared one statin to another, high-intensity to moderate-intensity dosing, or statins added to already optimized background therapy. Each step narrows the incremental margin available to demonstrate benefit: when baseline risk falls, when background therapies improve, and when control groups receive partial lipid-lowering treatment, absolute risk reduction inevitably shrinks. This is not evidence that statins stop working; it is arithmetic. As event rates decline due to better overall care and broader inclusion of lower-risk participants, the difference between treated and control groups compresses, and the net benefit becomes smaller in absolute terms—even if relative risk reductions appear similar. Over decades, shifting inclusion criteria and comparator design have therefore produced a progressive narrowing of measurable benefit, especially in primary prevention cohorts where baseline event rates are already low.

How Safety Claims Were Inflated by Comparator Collapse

Many recent claims that statins cause “no muscle pain” or “no cognitive side effects” rely on meta-analyses that include non-placebo-controlled studies. For example, a 2022 meta-analysis published in The Lancet (PMID: 36130978) argued that the incidence of muscle symptoms was nearly identical in statin and placebo groups. However, many included trials were not placebo-controlled and/or involved high-risk populations.

These studies typically compared statins to either other lipid-lowering agents (e.g., fibrates, niacin, ezetimibe) or to “usual care” in polypharmacy, comorbid patients. As a result, the side effect rates attributed to statins were interpreted as relative to alternative pharmacotherapies—not to a true physiological baseline. In such contexts, shared toxicities are indistinguishable, and additive harms are masked. For instance, in active-comparator trials, both arms may include medications that impair mitochondrial function or coenzyme Q10 synthesis, making it impossible to isolate statin-specific adverse effects.

Worse, meta-analyses like the 2022 Lancet study aggregated these active-controlled trials with a handful of placebo-controlled trials without stratifying by comparator type. This effectively neutralized the detection of true statin-attributable adverse events, because any signal was diluted by unrelated background drug effects or obscured by confounded baseline risks.

Furthermore, trials often excluded patients who developed symptoms during run-in phases or early withdrawal periods, a practice known as “healthy adherer bias.” By selecting participants who tolerate medication well in the first weeks—and removing those that have adverse effects-- the resulting study populations underrepresent the real-world prevalence of adverse effects. This selection process, when compounded with inadequate comparator framing, creates a misleading picture: side effects seem rare not because they are, but because the study architecture has systematically filtered out sensitive individuals and obscured the baseline for comparison.

Thus, when meta-analyses claim “statins do not increase the risk of muscle pain, fatigue, cognitive issues, neurological problems, or diabetes,” those claims rest on a scaffold of comparator collapse, selective exclusion, and inappropriate generalization—none of which support the safety profile implied for broad, low-risk primary prevention use. Physicians who have proper knowledge of statistical manipulation can identify these issues. Unfortunately, most physicians are relatively innumerate when it comes to statistical analysis. The widespread reporting of relative risk results instead of absolute risk outcomes are a perfect example of this.

Moreover, trials like the 2013 Mikus et al. study (JACC 2013;62:709-714) demonstrated that simvastatin 40 mg blunted VO2 max improvement and skeletal muscle mitochondrial enzyme adaptation to exercise training. In this RCT of 37 sedentary overweight adults, those in the statin + exercise group saw only a 1.5% increase in VO2peak vs 10% in exercise-only.

Biochemically, statins inhibit HMG-CoA reductase, reducing downstream synthesis of CoQ10, heme A, and dolichols. These compounds are essential for mitochondrial respiration and membrane stability. Vitamin K2 inhibition (used to activate matrix Gla-protein) is also impaired, leading to calcification of arterial walls Zhelyazkova‑Savova et al., Kaohsiung J Med Sci 2021;37(2):157–164). Other downstream (from HMG-CoA reductase) essential hormones are also inhibited include testosterone, DHEA, pregnenolone and progesterone.

Expected clinical consequences of these biochemical disruptions include reduced mitochondrial ATP production manifesting as fatigue, exercise intolerance, muscle weakness or myopathy, impaired cardiac diastolic function, and progressive cardiomyocyte dysfunction over long durations of exposure as well as hormonal imbalances. Concurrent inhibition of vitamin K2–dependent matrix Gla protein activation would be expected to promote vascular stiffening and coronary artery calcification, increasing long‑term atherosclerotic burden despite short‑term reductions in lipid-rich plaque. Indeed, statin studies have been shown to increase coronary artery calcium scores (Ngamdu et al, 2023; Henein, M, etal. 2015).

These mechanistic harms were never tracked in pivotal trials.

Generalization Without Evidence: The Push to Low-Risk Populations

After initial success, public health authorities and institutions such as Oxford Population Health began modeling statin effects in low-risk groups. Their claim: for every 10,000 low-risk adults treated, 500 cardiovascular events could be prevented over 5 years. But these were model-derived projections, not outcomes from randomized trials.

No statin trial has tested statin efficacy in truly low-risk individuals (10-year ASCVD risk <5%) with optimized lifestyle, normal metabolic markers, and no medication burden. The assumption of benefit in this group is unsupported.

The risks, however, are not theoretical.

What Was Never Tested: Diet and Exercise vs Statins

No head-to-head trial has ever compared:

● Statin + structured exercise vs

● Low-carbohydrate, zero-sugar, high-protein diet + exercise

Yet the latter has been shown, in meta-analyses of RCTs (Mansoor et al., Br J Nutr 2016;115:466-479; PMID:26768850), to improve triglycerides (~30 mg/dL reduction), HDL (~1.7 mg/dL increase), weight (~7 kg loss), and fasting glucose over 6–24 months. In STRRIDE-PD (Bennett et al., Front Physiol 2023;14:1199763), combined diet and exercise outperformed exercise alone in reversing metabolic syndrome.

Thus, a clinically plausible, non-pharmacological alternative has never been formally tested against statin therapy in an RCT — yet the latter is presumed superior.

Just like insulin resistance, doctors should encourage patients to reverse their condition with diet and exercise first, and reserve prescriptions with side effects for patients who truly cannot help themselves.

Where Risk Exceeds Benefit: Inappropriate Generalization of Benefit-Risk Ratios

Absolute risk reduction (ARR) with statins scales with baseline cardiovascular risk. In high-risk patients (10-year risk >20%), ARR ~2–5% over 5 years. In other words, from 95% to 98% of high-risk patients receive no benefit from taking statins over five years. In low-risk patients (10-year risk <5%), ARR is <0.5% over 5 years. In the case of low-risk patients, 99.5% of patients prescribed a statin will receive no benefit over five years of use. Those miniscule benefits may be acceptable if there were no adverse effects from statins. However, adverse effects (e.g., muscle pain, fatigue, insulin resistance, diabetes acceleration) do not scale down. Innumerate physicians do not understand simple statistics and this is why statins are so widely prescribed.

Before prescribing any drug, a physician needs to weigh the number-needed-to-treat (NNT) with the number-needed-to-harm (NNH). Without this data, patients do not have the proper information to decide if the therapy (a statin drug) is going to be a positive or negative for their condition. A lower NNT between 1-4 is an example of an effective medication. A lower NNH indicates more people will be harmed by the drug.

Number Needed to Treat (NNT) in low-risk populations ranges from 500–1000, while Number Needed to Harm (NNH) is often <100. The benefit-risk ratio collapses in favor of avoiding pharmacologic intervention. Who would take medication if they know the harms far outweigh the risks?

Global Epidemiology Contradicts the Universal Statin Narrative

Japan has among the lowest CHD mortality rates in the world despite higher average LDL levels and lower statin use. The French paradox shows similar patterns. Cross-country analyses demonstrate that diet, lifestyle, and metabolic syndrome prevalence are stronger drivers of CHD mortality than statin penetration.

The pattern in Japan alone draws the Oxford suggestion into question.

What an Ethical Statin Consent Conversation Should Include

An ethical statin consent process by any physician should include frank and detailed discussion of

● True ARR and NNT by risk category

● Real-world incidence of fatigue, muscle symptoms, sexual dysfunction, and cognitive complaints

● The untested nature of statin use in diet-optimized, non-polypharmic and healthy individuals

● Acknowledgment that pleiotropic effects (e.g., CRP lowering) may drive benefit, not LDL alone

What Needs to Happen: Study Design Repair

A 2x2 factorial trial should be conducted:

- Arm 1: Statin + usual care

- Arm 2: Statin + structured aerobic exercise

- Arm 3: Low-carb, no-sugar, high-protein diet + exercise

- Arm 4: Lifestyle + placebo

The trial must pre-specify censoring rules and conduct sensitivity analyses for right- and left-censoring, with fragility index reporting. Importantly, consistency (robustness) of results to censoring or fragility matter - no cherry-picking supportive results.

This study should be run on multiple demographic populations separately, and reproduced, to truly identify the relative benefits and risks of the two approaches.

The study should be large enough to do powered stratification by BMI, insulin resistance, and baseline polypharmacy. Measure mitochondrial function, CoQ10 levels, arterial calcium, inflammatory cytokines, and long-term cardiac endpoints.

The low-carb, no-sugar, high-protein diet + exercise long-term outcome must also protect against survivor bias.

Conclusion: The Statin Mess Was Constructed, Not Discovered

Statins slightly (ARR 1% to 3% in most studies) reduce events in high-risk patients. That is the core truth. But the vast expansion of use into populations that were never studied, through conflated comparators and misrepresented safety profiles, and the fact that safety was ascertained relative to other pharmaceuticals constitutes one of the most consequential distortions in evidence-based medicine. The comparator collapse, the modeling overreach, and the suppression of biochemical side effects have left clinicians and patients in the dark. The path forward must begin with reintroducing scientific clarity, ethical consent, and properly designed trials.

Oxford recently estimated that up to 2M people in the UK could benefit from statin use based on the flimsiest of assumptions - assumptions derived from modeled projections rather than new outcome trials.

Oxford’s Recent Bid to Open Markets

Oxford recently estimated that up to 2 million people in the UK could benefit from statin use based on the flimsiest of assumptions.

In coverage of a major systematic review published in The Lancet, researchers at Oxford Population Health and co-authors including Christina Reith and Sir Rory Collins argued that statins are “life-saving drugs” whose benefits outweigh most listed side effects and raised the possibility that many more at elevated cardiovascular risk should be receiving them, noting that current uptake leaves a sizeable gap between those treated and those eligible under modelled risk thresholds. UK reporting highlighted that although around 7–8 million people already take statins, experts — and models adopted from population-wide risk calculators — suggest that statin therapy could be beneficial for millions more who are currently untreated, a projection that assumes broad extrapolation of relative risk reductions without new outcome data in lower-risk groups. These estimates have been widely interpreted in the press as meaning up to 2 million additional Britons could benefit from starting statins, despite the fact that the underlying evidence is largely drawn from high-risk trials and risk projection models rather than direct outcome trials in genuinely low-risk, lifestyle-optimized populations.

Statins Are Not Without Risk

Despite their established benefits in secondary prevention for high-risk individuals, statins are not benign agents. A growing body of mechanistic, observational, and clinical research reveals plausible and increasingly quantified pathways through which chronic statin use may cause harm—particularly when used in low-risk, metabolically healthy individuals or for durations beyond the scope of pivotal trials.

Coronary Artery Calcification (CAC) Progression

A 2023 retrospective study of 1,181 U.S. veterans by Ngamdu et al. demonstrated a clear, dose-duration relationship between statin exposure and severe coronary artery calcification (CACS ≥ 400 Agatston units). While Ngamdu et al. is an observational trial, both temporal association and dose dependence are votes in terms of causality.

Compared to statin non-users, those using statins for >10 years had an adjusted odds ratio of 4.48 (95% CI: 2.7–7.43; p<0.001) for severe CAC, even after accounting for baseline ASCVD risk, BMI, and chronic kidney disease. This magnitude of association suggests that statins may accelerate atherosclerotic calcification in high-risk smokers—paradoxically elevating the very metric (CAC) often used to justify their use.

Mitochondrial Dysfunction and Nutrient Depletion

Statins inhibit the mevalonate pathway, reducing biosynthesis of CoQ10, heme A, and vitamin K2—critical cofactors for mitochondrial ATP production and vascular protection. Okuyama et al. (2015) synthesized pharmacological and clinical evidence indicating that chronic ATP depletion in cardiac myocytes, arising from CoQ10 and heme A suppression, may directly contribute to the development of statin-induced cardiomyopathy and congestive heart failure. Selenium-dependent glutathione peroxidase activity, likewise suppressed by statins, is also implicated in oxidative damage and cardiac dysfunction.

Accelerated Arterial Calcification via Vitamin K2 Inhibition

Statins interfere with the conversion of vitamin K3 to K2, inhibiting γ-carboxylation of matrix Gla protein—a process essential to preventing vascular calcification. Okuyama’s 2015 mechanistic review, paired with human cohort data, outlines how long-term statin use could mimic or potentiate warfarin-like calcification, elevating coronary artery stiffness and atherosclerotic burden.

Diabetes and Metabolic Disruption

Large-scale studies, including the Women’s Health Initiative (Culver et al., 2012) and the Veterans Affairs Diabetes Trial (VADT) (Saremi et al., 2012), confirm that long-term statin use increases the risk of new-onset diabetes mellitus. This effect is dose-dependent and plausibly linked to statin-induced interference with glucose transporter signaling, insulin receptor glycosylation, and mitochondrial regulation of insulin sensitivity. Diabetes is a known risk factor for the development of atherosclerosis.

Myopathy and Muscle Weakness

Beyond the well-known risk of rhabdomyolysis, statin users frequently report muscle pain, fatigue, and reduced exercise capacity. The Mikus et al. study (JACC 2013) showed that simvastatin 40 mg/day blunted the physiological gains from aerobic training, including VO₂ max and mitochondrial citrate synthase activity, compared to exercise alone. Emerging evidence suggests this may be due to toxic calcium leakage from sarcoplasmic reticulum, impairing myocyte contraction and resilience.

Clinical Blind Spots and Misattribution

One of the most insidious risks is diagnostic misattribution. As the Okuyama review notes, physicians routinely diagnose heart failure in statin-treated patients but rarely consider statins as causal agents. Delayed onset, non-specific symptoms, and the normalization of long-term therapy obscure causal inference, making cumulative toxicity harder to detect without deliberate post-market surveillance.

It is unacceptable that Oxford Population Health, with the weight of institutional authority, continues to promote the idea that statins carry “virtually no risk” while relying on meta-analyses that collapse placebo-controlled and active-comparator trials without distinction. This not only violates the basic principles of evidence stratification—it misleads the public into believing that adverse effects reported by thousands of real-world patients are imaginary. By uncritically amplifying these claims to millions of readers, The Telegraph has become a conduit for pharmaceutical reassurance masquerading as scientific consensus. There is no excuse for headline-level assertions of “no risk” when the underlying literature explicitly documents mitochondrial toxicity, vitamin K2 suppression, elevated coronary artery calcification, and increased diabetes incidence—none of which were adequately tracked in the trials used to justify the expansion of statins into low-risk populations. This is not a difference in interpretation; it is a failure of scientific responsibility. When institutions abandon the obligation to match conclusions to comparators, and when the media uncritically repackage those distortions, the result is not public health—it is narrative management.

A New Oxford-Led Study Quietly Confirms Risk—Then Publicly Denies It

In February 2026, the Lancet published a comprehensive meta-analysis by Reith, Blackwell, Emberson, and colleagues from the Cholesterol Treatment Trialists’ (CTT) Collaboration—the same Oxford-led group whose modeling underpins many of the most aggressive public health recommendations on statins. The study, which analyzed adverse effects from double-blind randomized controlled trials, confirms six categories of statistically significant harm, most of which are dose-dependent. And yet, in press coverage just days later, Oxford researchers publicly claimed statins carry “virtually no risk” and criticized product labeling for overstating side effects. The contradiction is not subtle—it is public messaging inconsistent with published findings, and that’s not right.

The study explicitly reports a moderate, dose-dependent increase in new-onset diabetes, especially in patients with prediabetes. High-intensity statins carried higher risk; low-intensity therapy less so. This aligns with earlier findings from the Women’s Health Initiative and Veterans Affairs Diabetes Trial. Additionally, a ~1% absolute excess in muscle-related symptoms was confirmed, primarily within the first year—ranging from mild myalgia to rare but real cases of myopathy and rhabdomyolysis (~1 case per 10,000 person-years).

The study also confirms:

● Elevations in liver transaminases and other liver function test abnormalities;

● Urinary abnormalities, including proteinuria and albuminuria;

● A small but real increase in peripheral edema.

Each of these effects met thresholds of statistical significance.

None of this is compatible with the phrase “virtually no risk.” But that phrase appears in The Telegraph’s February 6, 2026 article quoting Oxford researchers—an article that effectively advised the UK public that the adverse effects listed on official drug labels were baseless relics. In truth, Oxford’s own meta-analysis supports much of what those labels say—and contradicts the public messaging that followed.

This episode reveals the growing separation between what Oxford researchers publish under peer review and what they claim through media channels. It also illustrates how “safe” becomes a public narrative not through evidence, but through the selective suppression of it. As one journalist echoes Oxford’s public relations line, the fine print in the Lancet tells another story entirely. The public, and their physicians, deserve to read both.

Conclusion

The prevailing narrative of statins as safe for all obscures critical nuance. For high-risk patients, statins slightly (1% to 3%) reduce cardiovascular events. But in low-risk, long-duration, or metabolically optimized individuals, the risk-to-benefit ratio may invert—particularly given underappreciated mechanisms involving mitochondrial, vascular, and metabolic dysfunction. As new evidence accumulates, continued use without proper informed consent or outcome-specific monitoring is not risk-neutral—it is epistemologically and ethically indefensible.

References

Bennett, J. A., Slentz, C. A., Houmard, J. A., Kraus, W. E., & Duscha, B. D. (2023). Effects of combined diet and exercise on metabolic syndrome severity: STRRIDE-PD analysis. Frontiers in Physiology, 14, 1199763. https://doi.org/10.3389/fphys.2023.1199763

Cholesterol Treatment Trialists’ (CTT) Collaboration. (2016). Efficacy and safety of LDL-lowering therapy among 90,000 people in 26 randomised trials. The Lancet, 388(10059), 2532–2561. https://doi.org/10.1016/S0140-6736(16)31357-5

Henein, M, et al. 2015. High dose and long-term statin therapy accelerate coronary artery calcification. Int. J. of Cardiology. Vol. 184: p581-586April 01, 2015

Knapton, S. (2026, February 6). Millions more people ‘should be taking life-saving’ statins. The Telegraph. https://www.telegraph.co.uk/news/2026/02/06/millions-more-people-should-be-taking-life-saving-statins/

Mansoor, N., Vinknes, K. J., Veierød, M. B., & Retterstøl, K. (2016). Effects of low-carbohydrate diets vs. low-fat diets on body weight and cardiovascular risk factors: A meta-analysis of randomised controlled trials. British Journal of Nutrition, 115(3), 466–479. https://doi.org/10.1017/S0007114515004699

Mikus, C. R., Boyle, L. J., Borengasser, S. J., et al. (2013). Simvastatin impairs exercise training adaptations. Journal of the American College of Cardiology, 62(8), 709–714. https://doi.org/10.1016/j.jacc.2013.02.074

Ngamdu, K. S., Ghosalkar, D. S., Chung, H. E., et al. (2023). Long-term statin therapy is associated with severe coronary artery calcification. PLOS ONE, 18(7), e0289111. https://doi.org/10.1371/journal.pone.0289111

Okuyama, H., Langsjoen, P. H., Hamazaki, T., et al. (2015). Statins stimulate atherosclerosis and heart failure: Pharmacological mechanisms. Expert Review of Clinical Pharmacology, 8(2), 189–199. https://doi.org/10.1586/17512433.2015.1011125

Reith, C., Blackwell, L., Emberson, J., Preiss, D., Spata, E., Davies, K., et al. (2026). Assessment of adverse effects attributed to statin therapy in product labels: A meta-analysis of double-blind randomised controlled trials. The Lancet. Advance online publication. https://doi.org/10.1016/S0140-6736(25)01578-8

Scandinavian Simvastatin Survival Study Group. (1994). Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: The Scandinavian Simvastatin Survival Study (4S). The Lancet, 344(8934), 1383–1389. https://doi.org/10.1016/S0140-6736(94)90566-5

Shepherd, J., Cobbe, S. M., Ford, I., et al. (1995). Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. New England Journal of Medicine, 333(20), 1301–1307. https://doi.org/10.1056/NEJM199511163332001

Zhelyazkova-Savova, M. D., Marinov, M. R., & Dimitrova, K. R. (2021). Statins, vascular calcification, and vitamin K-dependent proteins: Is there a relation? Kaohsiung Journal of Medical Sciences, 37(2), 157–164. https://doi.org/10.1002/kjm2.12373