Biotin: why high doses matter more for blood tests than for hair

Many people take biotin for hair, skin and nails, yet the EU did not authorize the claim that it is necessary for healthy hair, skin and nails, although narrower claims about maintaining normal hair and skin are authorized separately. The firmest number about high-dose biotin concerns laboratory work: in a meta-analysis of 3 trials, 31 of 662 patients on biotin (4.7 percent) had blood test interference against none on placebo, on high-certainty evidence.

How much you need

• Measured biotin is highest in peanut butter (87.81 ug per 3.5 oz (100 g)) and egg yolk (45.9 ug), while bananas and butter tested at zero across all samples. — USDA Foundation Foods analyzed samples Evidence E source
• Biotin in the same food can vary widely: mature spinach samples ranged from 0 to 8.98 ug per 3.5 oz (100 g) and baby carrots from 0 to 11.0 ug, while flaxseed held 28.2 to 43.9 ug. — USDA Foundation Foods analyzed samples Evidence E source
• The US label Daily Value for biotin is 30 ug for adults and children 4 and over, 35 ug in pregnancy and lactation, 8 ug at ages 1 to 3 and 6 ug for infants. — US food labeling reference intakes Evidence E source
• FDA set the biotin label value from an Adequate Intake because data were too limited to set a Recommended Dietary Allowance. — US food labeling rule Evidence E source

How well your body absorbs it

• In a randomized crossover trial of 16 adults, biotin from a single serving of a multi-ingredient supplement raised plasma biotin over 8 hours against placebo. — 16 healthy adults (8 men, 8 women), single 13 g serving, 1-week washout Evidence B source

What too little does

• The Nordic Nutrition Recommendations 2023 review found symptomatic biotin deficiency rare and data to set biotin reference values limited. — Scoping review for Nordic Nutrition Recommendations 2023 Evidence C source

Who is most likely to fall short

• In an RCT of 26 pregnant women with raised 3-HIA, 300 ug biotin daily for 14 days lowered this marker of low biotin status. — 26 pregnant women with increased 3-HIA excretion, 6-37 weeks gestation, 14 days Evidence B source
• In a 24-week RCT of 80 pregnant women, a prenatal vitamin with 150 ug biotin raised maternal and cord blood biotin more than one with 30 ug. — 80 healthy pregnant women from week 12-16, 62 completed, 24 weeks Evidence B source
• In a controlled feeding study giving 57 ug biotin daily, lactating women excreted 76% more of the biotin breakdown product bisnorbiotin than controls, a sign of faster biotin loss. — 26 pregnant, 28 lactating and 21 control women, 10-12 week feeding study Evidence B source

What the evidence does and does not show

• A meta-analysis pooling 2 RCTs in 796 people with multiple sclerosis found insufficient evidence that high-dose biotin improved composite disability at 12 to 15 months. — 796 people with multiple sclerosis, 2 RCTs pooled for composite disability, high-dose biotin vs placebo Evidence A source
• In a phase 3 RCT of 642 people with progressive MS, 100 mg biotin three times daily improved disability in 12% against 9% on placebo, a non-significant difference. — 642 adults aged 18-65 with progressive MS, 90 clinics in 13 countries, about 20 months Evidence B source
• An earlier RCT of 154 people with progressive MS reported disability reversal in 12.6% on 300 mg biotin daily versus none on placebo, a result the larger trial did not reproduce. — 154 patients with progressive MS, EDSS 4.5-7, 12 months Evidence B source
• A meta-analysis of 5 RCTs (445 people with type 2 diabetes) found 28 to 90 days of biotin lowered total cholesterol by 8.5 mg/dL (0.22 mmol/L), while the drop in fasting glucose was uncertain. — 445 patients with type 2 diabetes, 5 RCTs, 28 to 90 days Evidence A source
• The EU did not authorize the claim that biotin is necessary for healthy teeth, bones, hair, skin and nails. — EU health claims register Evidence E source
• The EU also rejected a claim that a B-vitamin and biotin combination with pumpkin seed oil helps increase hair number. — EU health claims register Evidence E source

What too much does

• In the same meta-analysis, 4.7% of patients on high-dose biotin had laboratory test interference against none on placebo, with adverse events otherwise similar. — 662 patients on high-dose biotin vs placebo, 3 RCTs Evidence A source
• Laboratory medicine guidance warns that biotin supplements can cause clinically significant errors in some immunoassay blood tests, with falsely high or low results depending on the test format. — Clinical laboratories and clinicians Evidence E source
• In 50 children with biotinidase deficiency taking 5 to 20 mg biotin daily, one assay read vitamin D at a median 90.2 ng/mL that fell to 29.1 ng/mL once biotin was neutralized. — 50 children with biotinidase deficiency on biotin 5-20 mg/day Evidence C source
• A case with literature review describes biotin producing thyroid tests that mimicked Graves disease and normalized soon after stopping the supplement. — One patient, tertiary endocrine service, New Zealand, plus literature review Evidence D source

Medicines it interacts with

• In 75 children on valproic acid, those on higher doses had serum biotinidase activity about a fifth of healthy controls. — 75 children with seizures on valproate monotherapy vs 50 healthy controls, 45 days Evidence C source

What to do with this

Ordinary food supplies biotin unevenly. In USDA samples, peanut butter held 87.81 mcg per 3.5 oz (100 g), almond butter 57.31 mcg, egg yolk 45.9 mcg and rolled oats 21.9 mcg, while bananas and butter tested at zero. Egg yolk ranged from 26.4 to 75.2 mcg across 24 samples. Some zeros in other foods, such as spinach and baby carrots, may only mark the limit of the test method.

The US label Daily Value is 30 mcg for adults and children 4 and over, 35 mcg in pregnancy and lactation, 8 mcg at ages 1 to 3 and 6 mcg for infants. FDA based it on an Adequate Intake, because the data were too limited for a Recommended Dietary Allowance.

If you take biotin and have a blood test coming up, tell the laboratory and the person who ordered it. Laboratory medicine guidance says biotin can cause clinically significant errors in some immunoassays, falsely high or falsely low depending on the test format. Thyroid tests and vitamin D are among the documented cases. The guidance does not list every affected test, and we hold no card on troponin or other heart tests. So do not assume any blood test is safe from biotin, and mention it in an emergency too.

If you are pregnant or breastfeeding, or a child takes valproic acid, whether a biotin supplement makes sense is a question for the person treating you. We hold no card setting an upper limit for biotin. That is a gap in our sources, not a promise of safety at any dose.

Where to get it from food

USDA does not report biotin for individual foods, so this site cannot rank foods by it. The sections above name the foods the studies themselves looked at.

The bottom line

For most readers, biotin matters through blood tests. A single case report describes thyroid results that looked like Graves disease and returned to normal soon after the supplement was stopped. In 50 children with biotinidase deficiency taking 5 to 20 mg a day, one Roche assay read vitamin D at a median 90.2 ng/mL, which fell to 29.1 ng/mL once the biotin was neutralized, and other methods were not affected. The hope for multiple sclerosis did not hold up. An early trial in 154 people with progressive MS reported disability reversal in 12.6 percent on 300 mg a day against none on placebo. A larger phase 3 trial in 642 people found 12 percent improved against 9 percent, a difference that did not reach significance (OR 1.35, 95% CI 0.81 to 2.26). In 5 trials of 445 people with type 2 diabetes, 28 to 90 days of biotin lowered total cholesterol by 8.5 mg/dL (0.22 mmol/L). The fall in fasting glucose had a confidence interval that crosses zero. In a feeding study, breastfeeding women excreted 76 percent more of a biotin breakdown product than controls, a sign of faster loss. In pregnant women with a raised urine marker of low biotin status, 300 mcg a day for 14 days lowered it. These are markers, not health outcomes for mother or baby. In children on valproic acid, those on higher doses had biotinidase activity about a fifth of healthy controls, in a study that was not randomized. Symptomatic deficiency is rare, according to the review behind the Nordic recommendations of 2023.

Sources

1. btn-daily-r4-01 · Position of an expert body · food composition dataset
   Peanut butter, creamy | Biotin 87.81 UG | range 78.35-94.65 | samples 8 ; Eggs, Grade A, Large, egg yolk | Biotin 45.9 UG | range 26.4-75.2 | samples 24 ; Almond butter, creamy | Biotin 57.31 UG | range 53.2-65.6 | samples 8 ; Oats, whole grain, rolled, old fashioned | Biotin 21.9 UG | range 19.2-23.9 | samples 8 ; Bananas, ripe and slightly ripe, raw | Biotin 0.0 UG | samples 12 ; Butter, stick, salted | Biotin 0.0 UG | samples 6 (value from the source's table, not a sentence)
   USDA FoodData Central, Foundation Foods, release 2026-04-30 · checked 2026-10-02 · independently re-checked
2. btn-daily-r4-02 · Position of an expert body · food composition dataset
   Spinach, mature | Biotin 4.252 UG | range 0.0-8.98 | samples 8 ; Carrots, baby, raw | Biotin 2.396 UG | range 0.0-11.0 | samples 8 ; Flaxseed, ground | Biotin 33.8 UG | range 28.2-43.9 | samples 8 ; Mushroom, maitake | Biotin 32.57 UG | range 18.85-47.3 | samples 8 (value from the source's table, not a sentence)
   USDA FoodData Central, Foundation Foods, release 2026-04-30 · checked 2026-10-02 · independently re-checked
3. btn-daily-r4-03 · Position of an expert body · Regulatory reference value
   Biotin [RDI, Micrograms (mcg)] 21 CFR 101.9(c)(8)(iv) | 30 Adults and children >= 4 years | 6 Infants 1 through 12 months | 8 Children 1 through 3 years | 35 Pregnant women and lactating women (value from the source's table, not a sentence)
   21 CFR 101.9(c)(8)(iv), Reference Daily Intakes · checked 2026-10-02 · independently re-checked
4. btn-daily-r4-04 · Position of an expert body · Regulatory decision
   Accordingly, we are proposing to use AIs to set the RDIs for biotin, chloride, choline, chromium, manganese, pantothenic acid, potassium, and vitamin K.
   FDA, Food Labeling: Revision of the Nutrition and Supplement Facts Labels, proposed rule, 2014 · checked 2026-10-02 · independently re-checked
5. btn-daily-r4-17 · Randomized controlled trial(s) · Randomized crossover trial
   Plasma concentrations of folate, calcium, zinc, vitamin C, biotin, nicotinamide, pyridoxine, riboflavin, thiamine, and hesperidin were measured. ... Results: AG1 significantly increased AUC0-480 min values (p < 0.05) for all measured nutrients except pyridoxine which revealed a weak trend (p = 0.1205).
   Front Nutr, 2026 · checked 2026-10-02 · independently re-checked
6. btn-daily-r4-20 · Observational data · Scoping review
   Symptomatic biotin deficiency is rare, and few studies have investigated biotin requirements in relation to health outcomes. Data to support the setting of dietary reference values for biotin are limited.
   Food Nutr Res, 2024 · checked 2026-10-02 · independently re-checked
7. btn-daily-r4-13 · Randomized controlled trial(s) · Randomized controlled trial
   In the early-pregnancy group, 3-HIA excretion decreased (P < 0.006) by 11.7 +/- 3.6 mmol/mol creatinine (mean +/- SEM) in the 5 women who received biotin supplements, whereas 3-HIA excretion increased by 1.6 +/- 0.6 mmol/mol creatinine in the 5 women who received placebo. In the late-pregnancy group, 3-HIA excretion decreased (P < 0.002) by 7.1 +/- 1.2 mmol/mol creatinine in the 8 women who received biotin supplements, whereas 3-HIA excretion increased by 0.9 +/- 1.8 mmol/mol creatinine in the 8 women who received placebo.
   Am J Clin Nutr, 2002 · checked 2026-10-02 · independently re-checked
8. btn-daily-r4-14 · Randomized controlled trial(s) · Randomized controlled trial
   At both week 12 and 24, the intervention group had significantly higher serum biotin concentrations than the control group (p < 0.01). Cord blood biotin concentrations were also higher in the intervention than in the control group (p < 0.01).
   J Diet Suppl, 2026 · checked 2026-10-02 · independently re-checked
9. btn-daily-r4-15 · Randomized controlled trial(s) · Controlled feeding study
   Lactating women excreted 76% more (vs. control; P = 0.001) of the biotin catabolite bisnorbiotin, indicating that lactation accelerates biotin turnover and loss.
   J Nutr, 2014 · checked 2026-10-02 · independently re-checked
10. btn-daily-r4-05 · Meta-analysis or systematic review · Meta-analysis
    At 12 to 15 months, there is insufficient evidence that the HDB and placebo arms differed in terms of composite improvement of MS-related disability (relative risk (RR) 2.87; 95% CI 0.29-28.40; 2 trials; 796 participants; I2 = 66%) [low COE]
    Mult Scler Relat Disord, 2021 · checked 2026-10-02 · independently re-checked
11. btn-daily-r4-07 · Randomized controlled trial(s) · Randomized controlled trial
    For the primary outcome, 39 (12%) of 326 patients in the MD1003 group compared with 29 (9%) of 316 in the placebo group improved at month 12, with confirmation at month 15 (odds ratio 1·35 [95% CI 0·81-2·26]).
    Lancet Neurol, 2020 · checked 2026-10-02 · independently re-checked
12. btn-daily-r4-08 · Randomized controlled trial(s) · Randomized controlled trial
    A total of 13 (12.6%) MD1003-treated patients achieved the primary endpoint versus none of the placebo-treated patients (p = 0.005).
    Mult Scler, 2016 · checked 2026-10-02 · independently re-checked
13. btn-daily-r4-09 · Meta-analysis or systematic review · Meta-analysis
    It was suggested that biotin supplementation for 28 to 90 days significantly decreased the level of fasting blood glucose (FBG) (MD: -1.21 mmol/L, 95% CI: -2.73 to 0.31), total cholesterol (TC) (MD: -0.22 mmol/L, 95% CI: -0.25 to -0.19) and triglycerides (TG) (MD: -0.59 mmol/L, 95% CI: -1.21 to 0.03).
    Front Nutr, 2022 · checked 2026-10-02 · independently re-checked
14. btn-daily-r4-18 · Position of an expert body · Regulatory decision
    POL-HC-6774 Biotin Necessary for healthy teeth, bones, hair, skin and nails Non-authorised (value from the source's table, not a sentence)
    EU Register on nutrition and health claims, claim POL-HC-6774, snapshot 2026-09-21 · checked 2026-10-02 · independently re-checked
15. btn-daily-r4-19 · Position of an expert body · Regulatory decision
    POL-HC-8405 A combination of thiamin, riboflavin, niacin, pantothenic acid, pyridoxine, D-biotin and pumpkin seed oil (Cucurbita pepo L.) Helps to increase hair number -/- Non-authorised (value from the source's table, not a sentence)
    EU Register on nutrition and health claims, claim POL-HC-8405, snapshot 2026-09-21 · checked 2026-10-02 · independently re-checked
16. btn-daily-r4-06 · Meta-analysis or systematic review · Meta-analysis
    Synthesized data on incidence of any AEs (RR 0.98; 95% CI 0.92-1.04; 3 trials; I2 = 0%) [high COE] and any serious AEs (RR 0.98; 95% CI 0.77-1.24; 3 trials; 889 participants; I2 = 0%) [moderate COE] were not significantly different between HDB and placebo groups. Out of 662 pooled patients in the HDB group, 31 patients (4.7%) were found to have laboratory test interference compared to zero event in the pooled placebo group [high COE].
    Mult Scler Relat Disord, 2021 · checked 2026-10-02 · independently re-checked
17. btn-daily-r4-10 · Position of an expert body · Guideline
    Laboratory tests that use streptavidin-biotin binding mechanisms have the potential to be affected by high circulating biotin concentrations, which would produce positive and negative interference in biotinylated competitive and noncompetitive (sandwich) immunoassays, respectively. Consumption of high-dose biotin supplements for cosmetic or health-related reasons has drawn attention to biotin interference in clinical laboratory tests. Case reports and in vivo studies show that ingestion of supplemental biotin can cause clinically significant errors in select biotinylated immunoassays.
    J Appl Lab Med, 2020 · checked 2026-10-02 · independently re-checked
18. btn-daily-r4-11 · Observational data · Clinical study
    In contrast, the median 25-OHD level was seen to decrease from 90.2 [35.9-105.3] ng/mL to 29.1 [22.6-37.6] ng/mL after neutralization with the Roche assay (p < 0.0001). While there was no statistically significant difference in the values recorded before and after neutralization in PTH analysis using Beckman assay, the median PTH levels increased from 7.8 [1.6-21.6] pg/mL to 28.2 [22.5-41.9] pg/mL after neutralization with the Roche assay (p < 0.0001).
    Clin Biochem, 2022 · checked 2026-10-02 · independently re-checked
19. btn-daily-r4-12 · Mechanistic or laboratory data · Case report and review
    The patient had markedly abnormal TFTs that did not match the clinical context. After biotin cessation, TFTs normalized far more rapidly than possible given the half-life of T4, consistent with assay interference by biotin.
    J Clin Endocrinol Metab, 2016 · checked 2026-10-02 · independently re-checked
20. btn-daily-r4-16 · Observational data · Clinical study
    On the contrary, biotinidase activity was significantly statistically lowered (p < 0.001) in groups B and C (1.22 +/- 1.11, 0.97 +/- 0.07 mmol/min/L respectively), as compared with controls (5.20 +/- 0.90 mmol/min/L).
    Epilepsia, 2001 · checked 2026-10-02 · independently re-checked