Are statins safe? Evidence verdict.
Known benefits, small confirmed risks
Most people can take statins without a serious problem. They do cause small increases in muscle symptoms, diabetes diagnoses and abnormal liver blood tests, while serious muscle injury remains rare.
Evidence confidence: high for the small increases in muscle symptoms and diabetes; moderate for several less common outcomes.
The question
“Are statins safe, and how common are serious side effects?”
There is no honest one-word answer. Large blinded trials show that several risks are real but small on average. Other commonly feared problems, including frequent memory impairment and widespread liver damage, are not supported by the best trial evidence. The personal balance depends on why the statin was prescribed, the dose, other medicines and the person’s underlying cardiovascular risk. [1]
How to read these numbers
A person-year means one person followed for one year. An extra 11 events per 1,000 person-years is an average study estimate, not a prediction that the same 11 people will be affected in every group.
Relative and absolute risk also answer different questions. For low or moderate intensity statin treatment, the diabetes rate was 10% higher relative to placebo, but the estimated absolute excess was about 1.2 diagnoses per 1,000 people each year. [2]
What the best evidence says
SupportedStatins cause a small increase in muscle symptoms
Across 19 placebo-controlled trials with 123,940 participants, muscle pain or weakness was reported by 27.1% of people assigned a statin and 26.6% assigned placebo. During the first year, the difference was 11 extra reports per 1,000 person-years. More than 90% of all muscle-symptom reports among statin users were not caused by the statin. That population result cannot determine the cause of one person’s pain. [3]
SupportedA small diabetes risk is confirmed
Low or moderate intensity treatment increased new diabetes diagnoses from 1.2% to 1.3% per year, an estimated excess of 1.2 per 1,000 people each year. The relative increase was larger in the high-intensity trials, but those trials checked blood sugar much more often. About 62% of new diagnoses occurred in people whose starting blood-sugar measurements were already in the highest quarter. [2]
SupportedLiver blood-test abnormalities increase slightly
A 2026 analysis found about 1.3 extra combined liver-test abnormalities per 1,000 people each year. It did not find an increase in clinical liver outcomes such as jaundice, hepatitis, liver damage or liver failure. Blood-test changes and organ failure are different outcomes. Very rare events remain difficult to measure reliably in trials. [1]
Not supportedFrequent memory impairment is not shown in randomised trials
A review of 25 randomised trials involving 46,836 participants found no average cognitive impairment attributable to statins. A newer analysis also found no causal signal for cognitive impairment among the recorded trial outcomes. These findings do not prove that an individual can never experience a symptom, and they do not answer every question about lifetime dementia risk. [5] [1]
UncertainBleeding-stroke risk may be slightly higher
A 2024 review of 33 statin trials involving 216,258 participants estimated a 17% relative increase in haemorrhagic stroke, meaning bleeding in the brain. This does not mean 17% of users had a stroke. The absolute risk was small, and the estimate was sensitive to analysis choices. People with a previous brain haemorrhage need an individual discussion. [4]
What this does not prove
Trial averages cannot diagnose the cause of a particular symptom. They also cannot exclude every rare reaction, predict the effect of every drug interaction or establish that one statin is universally the safest.
The 2026 analysis reused data from established trials. It did not recruit 123,940 new patients, and a statistically undetected risk is not the same as zero risk.
When symptoms need attention
Do not stop a prescribed statin solely because of something you read online. Speak with the prescriber or pharmacist, who can assess the symptom, the medicine and possible interactions.
Australian atorvastatin information advises urgent assessment for unexpected muscle pain, tenderness or weakness, especially with fever or feeling unwell. Yellow skin or eyes with dark urine requires prompt medical attention. Breathing difficulty or swelling of the face, lips, mouth, tongue, throat or neck requires emergency care. Statins can interact with other medicines, and Australian atorvastatin product information lists pregnancy as a contraindication. [8]
Do the benefits outweigh the risks?
That depends on baseline cardiovascular risk. In a participant-level analysis of 27 randomised trials, each 1 mmol/L reduction in low-density lipoprotein cholesterol was associated with a 21% lower rate of major vascular events. Among people with a five-year event risk below 10%, this was about 11 fewer major vascular events per 1,000 people over five years. [7]
Those benefits are not directly interchangeable with side-effect counts. A muscle ache, diabetes diagnosis, coronary procedure, stroke and death have different consequences, and the studies measure them over different periods. A useful decision combines the person’s expected benefit, possible harms and preferences.
Key studies and their limits
On mobile: swipe sideways to view all columns.
| Source | What it found | Main limitations |
|---|---|---|
| Cholesterol Treatment Trialists’ Collaboration, 2026 Participant-level meta-analysis; 123,940 in placebo trials | Small increases in liver tests; most other label-listed outcomes had no causal trial signal. | Reuses older trials; adverse-event capture varied; rare-event power is limited; authors disclosed industry relationships. |
| Cholesterol Treatment Trialists’ Collaboration, 2024 Participant-level meta-analysis; 154,664 across two comparison sets | Confirmed a dose-dependent relative increase in diabetes diagnoses. | Absolute risk depended heavily on testing frequency; contributing trials included industry funding. |
| Cholesterol Treatment Trialists’ Collaboration, 2022 Participant-level meta-analysis; 123,940 in placebo trials | Small excess of mostly mild muscle symptoms, mainly in year one. | Selected trial populations; symptoms were recorded differently; overlaps with the 2024 and 2026 datasets. |
| StatinWISE, 2021 Double-blind crossover trials; 200 symptomatic patients | No average symptom-score difference between atorvastatin 20 mg and placebo. | Only 151 contributed both treatments; one statin dose; cannot dismiss an individual reaction. |
| Ott and colleagues, 2015 Systematic review; 25 trials, 46,836 participants | No average cognitive harm attributable to statins. | Different cognitive tests and follow-up periods; not a study of every rare symptom or lifetime dementia risk. |
The practical answer
Statins have well-established cardiovascular benefits and several real but generally small risks. Muscle symptoms and diabetes diagnoses increase slightly on average. Serious muscle injury is rare, liver blood-test changes are not the same as liver failure, and randomised evidence does not support frequent memory impairment. The right decision depends on the reason for treatment and the individual’s risk.
Common questions
How likely is muscle pain from a statin?
During the first year, the best large analysis estimated 11 extra reports per 1,000 person-years. Most muscle symptoms reported by people taking statins were caused by something other than the statin, but any new symptom still deserves assessment. [3]
Will a statin give me diabetes?
Usually not. Low or moderate intensity treatment was linked to about 1.2 extra diagnoses per 1,000 people each year in the analysed trials. Risk was concentrated among people already closer to the diagnostic threshold. [2]
Can statins damage the liver?
They can slightly increase abnormal liver blood-test results. A large blinded-trial analysis did not find an increase in clinical liver outcomes, but rare reactions and personal symptoms still require medical assessment. [1]
Do statins cause memory loss?
Randomised trials have not shown average cognitive harm from statins. This does not mean every new memory symptom should be ignored; other causes and the medicine should be reviewed. [5]
Does a higher dose cause more side effects?
Several risks increase with treatment intensity, including diabetes diagnoses and liver-test abnormalities. The suitable dose depends on the expected benefit, other medicines and individual risk. Do not change it without speaking to the prescriber. [2]
Should I stop my statin if I have side effects?
Speak with the prescriber or pharmacist before changing treatment. Seek urgent care for severe or unexpected muscle symptoms with illness, jaundice or dark urine, breathing difficulty, or facial and throat swelling. [8]
References
- Cholesterol Treatment Trialists’ Collaboration, 2026: Assessment of adverse effects attributed to statin therapy in product labels. Participant-level meta-analysis of double-blind randomised trials.
- Cholesterol Treatment Trialists’ Collaboration, 2024: Statin therapy, new diabetes diagnoses and worsening glycaemia. Participant-level meta-analysis.
- Cholesterol Treatment Trialists’ Collaboration, 2022: Effect of statin therapy on muscle symptoms. Participant-level meta-analysis.
- Bétrisey and colleagues, 2024: Cholesterol-lowering therapy and haemorrhagic stroke risk. Systematic review and meta-analysis.
- Ott and colleagues, 2015: Do statins impair cognition? Systematic review and meta-analysis of randomised trials.
- Herrett and colleagues, 2021: StatinWISE series. Randomised placebo-controlled crossover trials in people reporting muscle symptoms.
- Cholesterol Treatment Trialists’ Collaboration, 2012: Statin benefits in people at low vascular risk. Participant-level meta-analysis of 27 randomised trials.
- Australian Commission on Safety and Quality in Health Care: Atorvastatin SZ medicine information. Consumer and product information.
Evidence checked: 9 September 2026.
Disclaimer: This article is for general educational information only and is not medical advice. It does not diagnose, treat, or replace care from a qualified health professional. Always speak with a qualified health professional about personal health questions, medicines, pregnancy, medical conditions, or before changing treatment. Do not stop or change a prescribed statin without discussing it with the prescriber. Seek urgent medical care for severe symptoms or signs of an allergic reaction.
