Vitamin E

The VerdictLOW CONVICTION

It works for two specific liver and brain conditions under a doctor.

Next time you reach for a vitamin E capsule for "antioxidant support" or "heart health" — put it back, and eat a handful of sunflower seeds or almonds with your next meal instead. The dietary version is linked to lower disease risk; the supplement version is not, and chronic 400 IU/day has produced a documented harm signal.

The evidence: In 35,000 healthy men taking 400 IU/day for 5.5 years, prostate cancer risk went up 17%. Across 19 trials and 135,000 people at the same chronic dose, all-cause mortality went up slightly. For healthy-adult prevention, the direction is wrong.
What most people get wrong: They take a "400 IU" multivitamin assuming the form doesn't matter. Synthetic dl-α-tocopherol delivers about half the retained vitamin E per IU compared to natural RRR-α. The IU on the label hides the form.

Vitamin E is a family of fat-soluble compounds found in nuts, seeds, vegetable oils, and leafy greens. Your liver has a special carrier protein (α-TTP) that grabs only one specific natural form (RRR-α-tocopherol) and loads it onto fat-carrying particles to send around your body. Pour in more than that carrier can use and the rest gets thrown out. Take 400 IU per day for 5 years and the leftover starts interfering with blood clotting and other signalling — that's where the prostate cancer and mortality signals come from.

That's the general answer. Your stack is different.

Check your whole stack

Tonight, ask yourself one question: am I taking vitamin E for "prevention" or "antioxidant support"? If yes, the bottle goes back and a handful of sunflower seeds with your next meal does the actual job.

A 35,000-man trial at 400 IU/day for 5.5 years raised prostate cancer risk by 17%. The food version of vitamin E is linked to lower disease risk. The pill version is not — and at chronic high dose, it points the wrong way.

Takes less than 30 seconds. The seeds take 60.

The Verdict

It works for two specific liver and brain conditions under a doctor. For "antioxidant support" in healthy adults, it has caused harm.

Vitamin E is a family of fat-soluble compounds found in nuts, seeds, vegetable oils, and leafy greens. Your liver has a special carrier protein (α-TTP) that grabs only one specific natural form and loads it onto fat-carrying particles to send around your body. Pour in more than that carrier can use, and the rest gets thrown out. Take 400 IU per day for 5 years and the leftover starts interfering with blood clotting and other signalling. That is where the prostate cancer and mortality signals come from.

The evidence In 35,000 healthy men taking 400 IU per day for 5.5 years, prostate cancer risk went up 17%. Across 19 trials and 135,000 people at the same chronic dose, all-cause mortality went up slightly. For healthy-adult prevention, the direction is wrong.
What most people get wrong They take a "400 IU" multivitamin assuming the form on the label is the form that matters. Synthetic dl-α-tocopherol delivers about half the retained vitamin E per IU compared to natural RRR-α. The IU on the label hides the form.
The protocol in plain English Get 15 mg per day from food (about one handful of sunflower seeds or almonds, plus salad with olive oil dressing) and skip the pill. If a hepatologist diagnoses NASH or a neurologist diagnoses moderate Alzheimer's, ask about the specific clinician-supervised protocols.

Best for

Biopsy-proven non-diabetic NASH under a hepatologist (800 IU/day natural RRR × 96 weeks). Moderate Alzheimer's for slowing day-to-day function loss under a neurologist (2000 IU/day × 2 years; not memory). Primary period pain (200-500 IU/day for the days around menstruation).

Skip if

You are a healthy adult wanting "heart health," "longevity," "antioxidant support," or "cognition" benefit. Healthy men ≥50 chronically on ≥400 IU/day are the SELECT prostate cancer cohort. Anyone on warfarin, blood thinners, antiplatelets, scheduled for surgery in 2 weeks, in active cancer treatment, or with a history of bleeding stroke.

Want the full evidence? Keep scrolling

The Protocol

Vitamin E is a clinician-directed product for narrow indications. The healthy-adult retail dose is in the harm zone. The dietary RDA is reachable from food alone.

Population	Dose	Timing	Form	Duration
General adult — food only (RDA)	15 mg/day α-TE	With meals	Food matrix	Daily, lifelong
Biopsy-proven non-diabetic NASH (hepatologist-supervised)	800 IU/day natural RRR (≈1 standard high-dose capsule)	With largest fat meal	RRR-α-tocopherol (natural)	96 weeks (PIVENS protocol)
Moderate AD — functional decline only (neurologist-supervised)	2000 IU/day (≈2-4 capsules)	With fat-containing meal	dl-α-tocopheryl acetate (Sano trial form)	2 years (Sano 1997 protocol)
Primary dysmenorrhea	200-500 IU/day perimenstrually (1 capsule)	With food, 2 days before menses through day 3	Mixed acceptable	2-3 cycles trial
PCOS metabolic adjunct (gynaecology-supervised)	200-800 IU/day (varies)	With food	Variable	Specialist-directed
Male infertility adjunct (urology-supervised)	200-400 IU/day (often + vit C)	With food	Mixed acceptable	3-6 months (one sperm cycle)
Healthy-adult prevention (CVD, cancer, mortality, cognition, longevity)	NOT INDICATED	—	—	SELECT + Miller 2005 harm signal applies at chronic ≥400 IU/d

Forms — and the trick the IU on your multivitamin hides

The α-tocopherol transfer protein in your liver only binds one specific form of vitamin E. The "400 IU" on a multivitamin label can mean two very different things depending on whether the form is natural or synthetic.

Natural RRR-α-tocopherol (d-α)

Retention 1.0 (reference)

The form used in the PIVENS NASH trial. The only form fully retained by your liver carrier protein.

£10-25/mo retail; £20-45/mo at NASH dose

RRR-α-tocopheryl acetate

Retention ≈1.0 after digestion

Shelf-stable natural form. Esterase in your gut cleaves the acetate back to free RRR.

Moderate-high

Synthetic dl-α-tocopherol (all-rac)

Retention ≈0.5 of natural

The form in most multivitamins and in the SELECT trial. Half the IU retention of natural.

Low

Synthetic dl-α-tocopheryl acetate

Retention ≈0.5 of natural

Filler grade for cheap multivitamins. Same retention penalty as free synthetic.

Lowest

Mixed tocopherols (α + β + γ + δ)

α retained; β, γ, δ not retained

Marketing position. Non-α forms are rapidly excreted. No human RCT shows outcome superiority over α alone.

High

Tocotrienols (annatto, palm, rice bran)

Data unavailable for outcome

Theoretical biochemistry only. Human RCTs are small, short, mostly biomarker-only.

High

Absorption Tips

Take with the largest fat-containing meal of the day. Vitamin E is fat-soluble and needs bile, pancreatic enzymes, and dietary fat to be absorbed.
Conditions that reduce fat absorption (cystic fibrosis, cholestatic liver disease, short bowel, chronic pancreatitis) reduce vitamin E absorption proportionally.
Separate from iron supplements by at least 2 hours — iron can oxidise vitamin E in the gut.
Adequate vitamin C from food regenerates oxidised vitamin E back to its active form. High-dose vitamin C supplements are not needed and have their own exercise-blunting harm signal.

Safety & Interactions

The same antiplatelet and vitamin-K-antagonism mechanism that makes vitamin E useful as a chain-breaking antioxidant at dietary doses is what drives every meaningful interaction at supplemental doses. Disclose vitamin E to any prescriber.

Warfarin Severe

Vitamin E at ≥300-400 IU/d competes with vitamin K-dependent γ-carboxylation and inhibits platelet aggregation. Documented INR elevation and clinically meaningful bleeding events. Avoid >100 IU/d without prescriber-supervised INR monitoring.

Antiplatelet drugs (aspirin, clopidogrel, ticagrelor, DOACs) Moderate–Severe

Additive antiplatelet effect at ≥400 IU/d. Increased bleeding risk in post-cardiac-intervention, post-stroke, and high-CV-risk populations. Discuss with prescriber before continuing.

Active chemotherapy or radiotherapy Moderate–Severe

Theoretical antioxidant interference with oxidative-stress-mediated tumour kill. ASCO survivorship guidance discourages high-dose antioxidant supplementation during active treatment. Oncology team approval required.

Statin + niacin combination Moderate

The HATS trial showed a high-dose antioxidant cocktail (vit E + vit C + β-carotene + selenium) blunted the HDL-raising effect of simvastatin + niacin. Avoid concurrent high-dose antioxidant cocktail with this specific lipid regimen.

Vitamin K (supplemental) Moderate

Vitamin E at high doses antagonises vitamin K-dependent γ-carboxylation. Relevant for clients with marginal vit K intake or on warfarin therapy.

Iron supplementation Mild

Iron may oxidise vitamin E in the GI tract before absorption. Separate doses by at least 2 hours.

Contraindicated populations

Anyone on warfarin at any chronic dose >100 IU/d without prescriber-supervised INR monitoring.
Anyone on DOACs or antiplatelet therapy at high CV risk at chronic ≥400 IU/d.
Elective surgery scheduled within 2 weeks — discontinue at least 2 weeks before procedure per surgical guidance.
Active chemotherapy or radiotherapy at high dose without oncology team approval.
History of haemorrhagic stroke at ≥400 IU/d (Loh 2021 trend toward increased risk).
Healthy adult men ≥50 considering chronic ≥400 IU/d for prevention — this is the SELECT prostate cancer cohort.
Smokers considering high-dose multi-antioxidant cocktails — historical ATBC trial context.
Pregnancy and lactation at supplemental doses above the RDA — dietary 15 mg/d is adequate; high-dose supplementation is not characterised for outcome safety.

Tolerable Upper Intake Level 1000 mg α-tocopherol/d (≈1500 IU natural d-α / ≈1100 IU synthetic dl-α) for adults aged 19+ (IOM DRI 2000). The practical harm signal in healthy-adult prevention starts to appear at chronic ≥400 IU/d, which is well below this UL. The UL was set on bleeding-risk reasoning; the prostate-cancer and mortality signals from SELECT and Miller 2005 sit inside that margin.

Conviction

LOW overall — endpoint-stratified

Strong negative recommendation for healthy-adult prevention. Moderate positive for three narrow clinician-directed indications.

What would change this?

An independent double-blind placebo-controlled RCT in healthy adults aged 50-70 of 400 IU/day natural RRR-α-tocopherol × 5 years, with prostate cancer incidence + all-cause mortality + haemorrhagic stroke as pre-specified primary endpoints and formal stopping rules, showing no excess in any of the three endpoints, would upgrade healthy-adult prevention conviction from DEBUNKED-NET-HARM to LOW-NEUTRAL. No such trial is currently in flight. Separately, a multicentre biopsy-anchored RCT of 800 IU/day natural RRR × 96 weeks in non-diabetic NASH (N ≥ 300) replicating PIVENS resolution and adding ≥1-stage fibrosis improvement as co-primary, with histology adjudicated by independent blinded pathology panel, would upgrade NASH conviction from MODERATE to HIGH in that specific population.

Worth Your Money?

Weekly cost£1-6 per week at standard retail 200-400 IU/d (no consumer benefit). £5-11/wk for the PIVENS NASH 800 IU/d clinician-directed protocol. £6-15/wk for "premium full-spectrum" blends (no outcome evidence).

Worth it ifYou have biopsy-proven non-diabetic NASH and a hepatologist supports the 800 IU/d natural RRR × 96-week protocol. You have moderate Alzheimer's and a neurologist supports the 2000 IU/d Sano protocol for functional-decline-only. You have primary period pain and want a 2-3 cycle perimenstrual trial.

Lower priority ifYou are a healthy adult. Your first dollars in the supplement aisle go further on creatine (the one supplement that genuinely works), vitamin D (in genuine deficiency), magnesium (sleep / metabolic / vit-D cofactor), or omega-3 (triglyceride lowering + depression). Vitamin E from food — sunflower seeds, almonds, wheat germ oil, leafy greens — is the only intake target backed by an inverse association with disease risk, and it costs nothing extra.

Skip (healthy adults) Conditional Value (clinician-directed)

Go Deeper

Want to stop wasting money on supplements that don't work — and quietly avoid the ones that cause harm? The Verdict reviews one supplement every week, free.

Subscribe — free weekly review

Sources

Loh HC et al. (2021). Effects of vitamin E on stroke: a systematic review with meta-analysis and trial sequential analysis. Stroke Vasc Neurol. 18 RCTs N=148,016. Total stroke null; ischaemic borderline; haemorrhagic trending up. PMID 33109618.
Vrolijk MF et al. (Loomba) (2021). Network meta-analysis: pharmacologic therapies for fibrosis improvement and NASH resolution. Aliment Pharmacol Ther. 26 RCTs biopsy-proven NASH. Vitamin E significantly > placebo for fibrosis improvement. PMID 34435378.
Vadarlis A et al. (2021). Vitamin E supplementation in adults with NAFLD. J Gastroenterol Hepatol. ALT −7.37 IU/L; AST −5.71 IU/L; histology improved. PMID 32810309.
Farina N et al. (Cochrane) (2017). Vitamin E for Alzheimer's dementia and mild cognitive impairment. AD: functional-decline slowing in single trial; MCI: no benefit. PMID 28128435.
Lampousi AM et al. (2024). Vitamins C, E, and β-carotene and risk of type 2 diabetes. Adv Nutr. Dietary vit E 12 mg/d T2D RR 0.72 GRADE HIGH; supplementation RR 1.01 NS. PMID 38493875.
Wang L et al. (2022). Vitamin E intake and risk of prostate cancer. Nutrients. Pooled supplemental RR 0.99 NS; Europe subgroup 0.81. PMID 36615673.
Aune D et al. (2018). Dietary intake and blood concentrations of antioxidants and CVD/cancer/mortality. Am J Clin Nutr. Dietary biomarker inversely associated; supplementation not protective. PMID 30475962.
Liu Z et al. (2022). Vitamin E and C on male infertility. Int Urol Nephrol. N=832. Pregnancy rate RR 1.86; sperm parameters improved. PMID 35604582.
Tefagh G et al. (2022). Vitamin E in PCOS: cardiometabolic, inflammatory, hormonal. Sci Rep. 12 RCTs. TG, LDL, HOMA-IR, WC, sex hormones improved. PMID 35388031.
Kashanian M et al. (2022). Vitamin E on intensity of primary dysmenorrhea. Clin Nutr ESPEN. GRADE 8 RCTs. SDM −1.16 month 1, −1.83 month 2. PMID 36513486.
Sanderson J et al. (Cochrane) (2018). Vitamin/mineral supplementation for cognitive maintenance in healthy mid/late life. 28 studies, >83,000 participants. No effect. PMID 30556597.
Talebi S et al. (2022). Dietary antioxidants and Parkinson's disease risk. Adv Nutr. GRADE. Vit E 5 mg/d increment RR 0.84. PMID 35030236.
Miller ER 3rd et al. (2005). High-dosage vitamin E supplementation may increase all-cause mortality. Ann Intern Med. 19 RCTs N=135,967. ≥400 IU/d: +39/10,000 persons. Preflight-anchored landmark safety MA cited by NIH ODS.
Klein EA et al. SELECT (2011). Vitamin E and prostate cancer risk: SELECT trial. JAMA. N=35,533 men, 400 IU/d dl-α-tocopheryl acetate × 5.5 y. HR 1.17 (99% CI 1.004-1.36, p=0.008). Preflight-anchored landmark safety RCT.
Sanyal AJ et al. PIVENS (2010). Pioglitazone, vitamin E, or placebo for nonalcoholic steatohepatitis. N Engl J Med. N=247 biopsy-proven non-diabetic NASH, 800 IU/d natural RRR × 96 wk. NASH resolution 36% vs 21% (p=0.001). Indication-defining RCT.
Sano M et al. (1997). Selegiline, α-tocopherol, or both for Alzheimer's disease. N Engl J Med. Moderate AD, 2000 IU/d × 2 y. 230-day median delay in functional-decline composite. Indication-defining RCT.
IOM Food & Nutrition Board (2000). Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium, and Carotenoids. RDA 15 mg/d α-tocopherol; UL 1000 mg/d. Foundational DRI reference.
NIH Office of Dietary Supplements. Vitamin E — Health Professional Fact Sheet. Preflight source. Synthesises trial conclusions, drug interactions, UL.