Biotin Human Clinical Evidence: What the Studies Show

Biotin is sold as a beauty vitamin, yet the strongest human data point somewhere else. Biotin human clinical evidence shows clear benefit when true deficiency is corrected, thin and mostly uncontrolled support for brittle nails, no placebo-controlled proof that supplements regrow hair in people with normal biotin status, and a documented hazard: high doses can distort blood tests. The U.S. FDA has reported one death after a falsely low troponin result linked to biotin interference.

If you take biotin, advise people who do, or write about it, the gap between marketing and measurement matters. This article walks through the human studies outcome by outcome, showing who was studied, what dose they received, who funded the work and how much confidence each finding deserves. Animal and laboratory findings are kept out of the efficacy claims.

Biotin Human Clinical Evidence at a Glance

• Treating deficiency: In people with confirmed deficiency (inherited biotinidase or holocarboxylase defects, long-term parenteral nutrition, some medication effects), biotin resolves symptoms in published case reports. No randomized trials exist, and withholding treatment in such patients would be hard to justify.
• Hair: A 2024 review found only one placebo-controlled trial of biotin alone for hair, and it reported no difference from placebo. Evidence of benefit in people with adequate biotin status is absent.
• Nails: Three small studies using 2.5 mg/day report improvement in brittle nails. None had a placebo group.
• Skin: Support is limited to case reports, mainly in infants with deficiency.
• Metabolic outcomes: A 2022 meta-analysis of randomized trials in type 2 diabetes is inconclusive, because confidence intervals for fasting glucose and triglycerides include no effect.
• Multiple sclerosis (300 mg/day): A 154-patient phase 3 trial was positive; a 642-patient phase 3 trial was negative.
• Laboratory tests: Controlled studies in healthy adults show assay interference at 5–10 mg/day. The FDA has issued warnings, including about troponin tests.

The Research Question

This review asks a practical question, structured the way clinical researchers frame it: in humans, does oral biotin (the intervention) change measurable health outcomes (hair, nails, skin, metabolic markers, neurological disability, biomarkers of status) compared with placebo, no treatment or baseline, and does it affect the accuracy of laboratory tests?

Two populations are kept apart throughout. The first is people with deficiency or a condition that depletes biotin. The second is people with adequate biotin status, who make up most supplement users. Evidence from the first group cannot be assumed to apply to the second.

Biotin Clinical Studies: Search and Inclusion Methodology

What this is. A structured evidence review, not a formal PRISMA systematic review. One reviewer searched and appraised sources; there was no duplicate screening and no pooled analysis of my own.

Sources searched (October 2026). PubMed and PubMed Central records, publisher pages, ClinicalTrials.gov, the Iranian Registry of Clinical Trials, the FDA, and the NIH Office of Dietary Supplements (ODS).

Search terms. Combinations of biotin with hair, alopecia, nails, brittle nails, skin, supplementation, randomized controlled trial, type 2 diabetes, glycemic control, multiple sclerosis, MD1003, pregnancy, marginal deficiency, interference, thyroid and troponin.

Eligibility.

• Included: human studies where biotin was the exposure (randomized trials, non-randomized and observational studies, case series, systematic reviews and meta-analyses), plus controlled human studies of biotin ingestion and laboratory-test accuracy.
• Excluded from efficacy conclusions: animal, cell and laboratory work; multi-ingredient products where biotin's contribution cannot be isolated (shown for context only); and registered trials without retrievable results.
• Case reports are cited only to describe the deficiency evidence and are graded lowest.

Verification. Authors, years, journals, sample sizes and identifiers were cross-checked against PubMed, PubMed Central, publisher or registry pages. Where I could see only an abstract or another review's summary of a study, the table says so. No number in this article was estimated or filled in from memory; unavailable details are marked "not reported" or "not retrieved."

Known limits of the method. Full texts of several older studies were not accessible, one database search cannot guarantee completeness, and publication bias cannot be ruled out. Treat the evidence grades as transparent judgments, not formal GRADE ratings.

Biotin Human Trials by Health Outcome

Biotin deficiency versus adequate biotin status

Biotin deficiency from diet alone is rare, and severe deficiency has never been reported in healthy people eating a mixed diet (NIH ODS). Typical intake in Western populations is about 35–70 µg/day, against an adequate intake of 30 µg/day for adults. Groups more likely to run low include people with biotinidase deficiency, chronic alcohol exposure, long-term anticonvulsant therapy, and pregnant or breastfeeding women.

Deficiency signs include thinning hair, a scaly rash around the eyes, nose and mouth, brittle nails, and neurological symptoms. That is why biotin is marketed for hair, skin and nails, but the reasoning has a flaw: correcting a deficiency does not show that extra biotin helps someone who is not deficient.

Patel and colleagues (2017) reviewed published cases and found 18 reports of biotin used for hair or nail changes. Every patient had an underlying cause of poor growth, and every one improved. The authors concluded there was insufficient evidence for supplementation in healthy people.

Trüeb (2016) measured serum biotin in 541 women (aged 9–92) who consulted a Zurich hair clinic. About 38% had levels under 100 ng/L. This is a retrospective clinic sample with no treatment outcomes, and the author later disclosed support from Merz Pharma. Another study (Abdel Rahman 2020) found no difference in biotin levels between 60 women with telogen effluvium and 20 controls. Serum biotin may also be a poor marker of marginal status, according to ODS. These data raise a question about testing; they do not show that supplements help.

Hair

Yelich and colleagues (2024) searched PubMed for biotin used alone for hair, found 330 records, and kept three studies after excluding case reports and series.

• Pawlowski 1966: a double-blind, placebo-controlled trial in 46 women with diffuse hair loss (28 on 10 mg/day, 18 on placebo) for four weeks. Both groups improved, with no significant difference in hair growth or sebum output.
• Aksac 2021: 60 patients taking isotretinoin, 30 of them also given 10 mg/day for four months. Terminal hair density fell in both groups; the biotin group showed a shift toward growing-phase hairs compared with its own baseline.
• Sen 2021: 112 women with hair loss after sleeve gastrectomy; 22 were biotin-deficient. About 23% of deficient and 38% of non-deficient women taking 1 mg/day reported marked improvement, a non-significant difference based on self-report.

The reviewers judged that none of this supports biotin for hair in people with adequate levels. Hair benefits are reported mainly as case reports in children with uncombable hair syndrome (3–5 mg/day, per ODS) and in specific deficiency settings.

Multi-ingredient products are a trap for readers. Ablon (2015) randomized 60 women with self-perceived thinning hair to a marine protein supplement or placebo for 90 days and reported less shedding (P=0.002). The product contained several ingredients, so the result cannot be credited to biotin, and the product belongs to the company that makes it. Registered trials of other multi-ingredient products and of biotin–dexpanthenol mesotherapy exist (for example NCT06605768, NCT07520019 and IRCT 50577), but I could not retrieve results and give them no weight.

Nails

Three small studies, all giving 2.5 mg/day, form the whole base (ODS; Lipner commentary).

• Floersheim 1989: firmer, harder nails in 41 of 45 patients (91%) after a mean of about 5.5 months; brittleness and ridging reportedly returned roughly 10 weeks after stopping. A conference report cites 71 patients for the same work, so enrollment details are inconsistent across sources.
• Colombo 1990: electron microscopy in 32 people. In the eight patients sampled before and after treatment, nail thickness rose 25% (P<0.05); in 14 others with less precise timing it rose 7%, not significant.
• Hochman 1993: a retrospective survey; 22 of 35 respondents (63%) reported improvement, typically within about two months.

ODS notes that none of the three had a placebo group and none reported baseline biotin status. Lipner has described these as the entire evidence base and noted that no rigorously designed trial has shown any supplement to work for brittle nails.

Skin

For skin, ODS identifies only a handful of case reports, all in infants, where 100 µg to 10 mg/day cleared rash or dermatitis in the setting of deficiency. I found no controlled human trials of biotin for skin health in people with adequate status.

Metabolic outcomes in type 2 diabetes

Zhang and colleagues (2022) meta-analyzed randomized trials of biotin in type 2 diabetes (searches through August 2022; trial and participant counts were not retrievable from the sources I could access). For supplementation lasting 28–90 days, the pooled mean differences were:

• fasting glucose −1.21 mmol/L (95% CI −2.73 to 0.31),
• total cholesterol −0.22 mmol/L (−0.25 to −0.19),
• triglycerides −0.59 mmol/L (−1.21 to 0.03).

Insulin did not change significantly, and evidence for HbA1c, LDL, HDL and VLDL was too limited to conclude. Notice that the intervals for glucose and triglycerides cross zero, so despite wording in the abstract, only the cholesterol result excludes no effect. A 2025 umbrella review of water-soluble vitamins in type 2 diabetes described prior meta-analyses as inconsistent and called for better trials. These findings are short-term and do not show that biotin improves long-term outcomes.

Pregnancy and marginal deficiency

In a randomized, placebo-controlled trial, Mock and colleagues (2002) studied 26 pregnant women with raised urinary 3-hydroxyisovaleric acid (3-HIA), a marker of reduced biotin-dependent enzyme activity. After 14 days of 300 µg/day, 3-HIA fell significantly in early and late pregnancy, while it rose slightly on placebo. ODS reports that at least a third of pregnant women develop marginal deficiency despite normal intake, but the clinical meaning is still unclear. A later commentary noted that the 3-HIA marker may not be reliable in pregnancy.

Concern about birth defects comes from mouse studies. That is preclinical evidence and does not establish risk in humans. The adequate intake is 30 µg in pregnancy and 35 µg in lactation.

Multiple sclerosis and high-dose biotin

Pharmaceutical-grade biotin (MD1003) at 300 mg/day, about 10,000 times the adequate intake, was tested in progressive multiple sclerosis. In MS-SPI (Tourbah 2016), 154 patients were randomized for 12 months: 13 treated patients (12.6%) met the improvement endpoint versus none on placebo (P=0.005), with a safety profile similar to placebo.

The larger SPI2 trial (Cree 2020; 642 participants, 90 sites in 13 countries, 15 months) did not replicate this. Disability and walking speed did not improve significantly, and the investigators concluded that MD1003 could not be recommended. They also reported inaccurate laboratory results despite mitigation measures. These doses are far beyond over-the-counter products and say nothing about hair, nails or skin.

Laboratory test interference with thyroid and troponin assays

Many immunoassays capture their target using a biotin–streptavidin bond. Excess biotin in a blood sample can block that step, giving falsely high results in competitive assays and falsely low results in sandwich assays. This is a measurement effect, studied in people, and it is the best-supported finding in the whole biotin literature.

• Li (2017, JAMA): six healthy adults took 10 mg/day for seven days. Interference appeared in 9 of 23 biotinylated assays and none of 14 non-biotinylated assays (P=0.007). One platform's TSH read falsely low enough to suggest thyrotoxicosis, and NT-proBNP read falsely low.
• Biscolla (2017): in 19 volunteers, a single 10 mg dose interfered with TSH, free T4 and T3 assays, with effects gone within 24 hours.
• Trambas (2016): clinicians described euthyroid patients on high-dose biotin whose results mimicked hyperthyroidism (TSH below 0.01 mIU/L, free T4 above 100 pmol/L), sometimes triggering unnecessary antithyroid treatment. They advised stopping biotin 72 hours before blood tests and noted interference above about 5 mg.
• Washout study (2020): in 10 healthy adults, 5 or 10 mg/day interfered with thyroid tests on two of three platforms; interference lasted about 8 hours at 5 mg and 1–2 days at 10 mg.

The AACC guidance reports that multivitamin doses up to 1 mg have not been linked to interference. The FDA's 2017 safety communication and its 2019 update stress falsely low troponin results as the key concern, and in 2022 the agency published a list of troponin assays that had not addressed the risk. Washout times differ between sources and assays, so patients should tell the ordering clinician and laboratory about any biotin and follow their instructions.

Human Studies Evidence Table

NR = not reported in the sources I could access. "Per Yelich 2024" means I relied on that review's summary of the original study.

How the Studies Compare: Design Quality, Participants, Dose and Duration

Study design and risk of bias. Only a few entries are randomized and blinded: Pawlowski (hair), Mock (pregnancy biomarker), the two MS trials, and the trials inside the diabetes meta-analysis. The nail studies are uncontrolled, the hair studies after 1966 are small and unblinded, and the interference studies are small experiments in healthy volunteers. Risk of bias is high for most hair and nail evidence and lower for the MS trials.

Participants. Populations are narrow: women with hair loss, patients on isotretinoin, post-surgical patients, people with progressive MS, pregnant women, and healthy adult volunteers. Few studies report baseline biotin status, which is the single most important variable for interpreting a result.

Dose and form. Doses span four orders of magnitude: 300 µg in pregnancy, 1–10 mg in hair studies, 2.5 mg in nail studies, and 300 mg in MS. Most used oral biotin; the MS trials used a pharmaceutical-grade capsule. Formulation details for the older studies were not retrievable. ODS reports that absorption of free biotin is essentially complete even at 20 mg/day.

Duration. Between 14 days (pregnancy biomarker) and 15–27 months (SPI2). Hair and nail studies run from 4 weeks to about 15 months, long enough to see nail change but not always long enough to judge hair cycles.

Endpoints. Many are surrogate or subjective: urinary 3-HIA, self-reported improvement, electron-microscope thickness. Patient-relevant measures (disability scores, validated hair counts) appear mainly in the MS and a few hair trials.

Clinical Significance

Statistical significance and clinical meaning are different things here. A 25% rise in nail plate thickness in eight patients is measurable but unproven to reduce splitting in daily life. A 12.6% responder rate in MS-SPI looked meaningful until SPI2 failed to confirm it. A drop in 3-HIA shows an enzyme marker moved, not that a pregnancy outcome improved.

The meta-analytic changes in cholesterol are small in absolute terms and based on short trials, so they do not justify using biotin as a cardiovascular strategy. By contrast, the laboratory-interference findings are clinically significant even in small studies, because a falsely low troponin or a falsely abnormal thyroid panel changes care.

Conflicts of Interest and Funding

Disclosure is uneven, which is itself a finding. The MS trials involve the developer of MD1003 (MedDay), whose people appear in author lists and registry records; the first trial was positive and the larger, later one was not, a reminder to weigh independent replication. The serum-biotin study in women with hair loss was supported by Merz Pharma, according to the author's later comment. The multi-ingredient hair trial involves a branded product. The 2024 hair review reports no funding and no conflicts, and the JAMA interference study reported no relevant financial disclosures. For many older nail and hair studies, funding was not reported in the sources I could access.

Consistency Across Trials

• Nails: consistent direction (improvement) across three small uncontrolled studies, but consistency among weak designs is not strong evidence, because the same biases can affect all three.
• Hair: inconsistent and mostly null or uninterpretable; benefit appears chiefly where deficiency or a specific condition exists.
• Diabetes and lipids: inconsistent, with wide intervals.
• MS: inconsistent, and the larger trial contradicts the smaller one.
• Lab interference: consistent across studies, platforms and case series. The size of the effect varies by assay and dose, with some assays unaffected.

Meta-Analyses and Systematic Reviews

No meta-analysis of biotin for hair or nails was found. For these outcomes the best syntheses are narrative or systematic-style reviews (Patel 2017; Yelich 2024), and both conclude that the evidence does not support supplementation in people with adequate biotin status. The only quantitative synthesis located is Zhang (2022) for type 2 diabetes, whose conclusions are limited by imprecision. The 2025 umbrella review of water-soluble vitamins called for higher-quality trials. I found no meta-analysis of the MS trials or of interference studies; the FDA and AACC guidance fill that role for laboratory safety.

Evidence Grading by Health Outcome

Method. Grades follow GRADE logic applied informally. Randomized trials start high and observational or case-based evidence starts low. Certainty is lowered for risk of bias, inconsistency, indirectness (wrong population or dose), imprecision and suspected publication bias, and raised for very large or consistent effects. Four levels are used: high, moderate, low, very low. This is my narrative appraisal of public sources, not a formal GRADE assessment.

Limitations and Research Gaps

• No adequately powered, placebo-controlled trial of biotin alone exists for hair loss or brittle nails in people with measured biotin status, which would be the decisive test.
• Baseline status is rarely measured, and serum biotin may not reflect marginal deficiency, so trials need better markers (such as enzyme-based measures).
• Dose–response is unknown for healthy adults at supplement-label doses, including whether anything between 30 µg and 5 mg does anything.
• Long-term safety data at supplement doses are thin, and no tolerable upper limit exists because toxicity has not been shown, which is not the same as proof of safety.
• Pregnancy: the link between marginal deficiency and outcomes in people rests on biomarker and animal data.
• Interference: washout times and which assays are affected vary by platform; some assays have since been redesigned, so older findings may not apply to current tests.
• My own limits: full texts of several older studies and of the diabetes meta-analysis were unavailable, so details such as trial counts are missing and no claim was inflated to fill them.

Bottom Line: What Biotin Human Clinical Evidence Supports

What human studies establish. Biotin corrects deficiency. High intakes can make some blood tests wrong, sometimes dangerously. At 300 mg/day, biotin did not improve disability in the larger MS trial.

What remains uncertain. Whether supplements help brittle nails beyond small uncontrolled reports; whether people with low-normal biotin and hair loss benefit; whether short-term cholesterol and glucose changes mean anything clinically; and whether correcting marginal deficiency in pregnancy improves outcomes.

What cannot be justified today. Claims that biotin grows hair or improves skin in people with normal biotin status; claims that high-dose products are risk-free; and claims that animal or mouse data prove human benefit.

What to do with this. If you are considering biotin, start with diet and ask a clinician whether you have a reason to be deficient (alcohol use, bowel surgery, certain anticonvulsants, restricted diets, pregnancy). Tell your doctor and the laboratory about any supplement before blood tests, especially for thyroid, heart or hormone panels. Be wary of products promising results that the studies above do not support. This article is educational information, not medical advice.

Frequently Asked Questions

Does biotin make hair grow?

For people with normal biotin levels, no controlled human evidence shows that it does. The only placebo-controlled trial of biotin alone found in a 2024 review (46 women, 4 weeks, 10 mg/day) showed no difference from placebo. Benefit is reported mainly where deficiency or a specific condition exists.

Is biotin proven to help brittle nails?

It is promising but unproven. Three small studies using 2.5 mg/day reported improvement, yet none had a placebo group and none reported baseline biotin status. A well-designed randomized trial is still missing.

Can biotin help type 2 diabetes?

Evidence is inconclusive. A 2022 meta-analysis reported small pooled changes, but the confidence intervals for fasting glucose and triglycerides include zero, and trials were short.

How does biotin affect thyroid and troponin tests?

In assays that use a biotin–streptavidin bond, excess biotin can produce falsely high or low results. Controlled studies show this at 5–10 mg/day for some platforms, and the FDA warns about falsely low troponin results. Tell the clinician and the laboratory about any biotin you take, and ask how long to stop it before testing.

How much biotin do adults need?

The adequate intake is 30 µg/day for adults, 30 µg in pregnancy and 35 µg during lactation. Typical diets in Western countries supply roughly 35–70 µg/day. No upper limit has been set.

Is there a meta-analysis for biotin and hair or nails?

I did not find one. The best summaries are narrative reviews that conclude evidence is limited outside deficiency.

References

Links were checked at the time of writing (October 14, 2026). DOIs and PubMed IDs are given where I could confirm them.

1. Patel DP, Swink SM, Castelo-Soccio L. A review of the use of biotin for hair loss. Skin Appendage Disord. 2017;3(3):166–169. doi:10.1159/000462981. PubMed · Free full text
2. Yelich A, Jenkins H, Holt S, Miller R. Biotin for hair loss: teasing out the evidence. J Clin Aesthet Dermatol. 2024;17(8):56–61. PubMed · Free full text
3. Trüeb RM. Serum biotin levels in women complaining of hair loss. Int J Trichology. 2016;8(2):73–77. doi:10.4103/0974-7753.188040. PubMed · Free full text
4. Trüeb RM. Comment on the use of biotin for hair loss. Skin Appendage Disord. 2017;4(4):345–346. Free full text
5. Abdel Rahman SH, Mohammed Salem R, Hassan Sabry J. Biotin deficiency in telogen effluvium: fact or fiction? J Clin Aesthet Dermatol. 2020;13(3):37–40. PubMed
6. Pawlowski A, Kostanecki W. Effect of biotin on hair roots and sebum excretion in women with diffuse alopecia. Pol Med J. 1966;5(2):447–452. PubMed
7. Aksac SE, Bilgili SG, Yavuz GO, et al. Evaluation of biophysical skin parameters and hair changes in patients with acne vulgaris treated with isotretinoin, and the effect of biotin use on these parameters. Int J Dermatol. 2021;60(8):980–985. doi:10.1111/ijd.15485. PubMed
8. Şen O, Türkçapar AG. Hair loss after sleeve gastrectomy and effect of biotin supplements. J Laparoendosc Adv Surg Tech A. 2021;31(3):296–300. doi:10.1089/lap.2020.0468. PubMed
9. Ablon G. A 3-month, randomized, double-blind, placebo-controlled study evaluating the ability of an extra-strength marine protein supplement to promote hair growth and decrease shedding in women with self-perceived thinning hair. Dermatol Res Pract. 2015;2015:841570. doi:10.1155/2015/841570. Free full text · Registry: NCT02297360
10. Floersheim GL. Treatment of brittle fingernails with biotin. Z Hautkr. 1989;64:41–48. PubMed
11. Colombo VE, Gerber F, Bronhofer M, Floersheim GL. Treatment of brittle fingernails and onychoschizia with biotin: scanning electron microscopy. J Am Acad Dermatol. 1990;23(6 Pt 1):1127–1132. PubMed
12. Hochman LG, Scher RK, Meyerson MS. Brittle nails: response to daily biotin supplementation. Cutis. 1993;51:303–305. PubMed
13. Lipner SR, as reported in: Limited treatment options exist for brittle nail syndrome. MDedge Dermatology. Link
14. Zhang Y, Ding Y, Fan Y, et al. Influence of biotin intervention on glycemic control and lipid profile in patients with type 2 diabetes mellitus: a systematic review and meta-analysis. Front Nutr. 2022;9:1046800. doi:10.3389/fnut.2022.1046800. Full text
15. Umbrella review of water-soluble vitamins on glycemic control and insulin resistance in adults with type 2 diabetes. Asia Pac J Clin Nutr. 2025;34(1):118–130. doi:10.6133/apjcn.202502_34(1).0012. PubMed
16. Mock NI, Quirk JG, Mock DM. Marginal biotin deficiency during normal pregnancy. Am J Clin Nutr. 2002;75(2):295–299. PubMed
17. Mock DM. Adequate intake of biotin in pregnancy: why bother? J Nutr. 2014;144:1885–1886. Free full text
18. Tourbah A, Lebrun-Frenay C, Edan G, et al. MD1003 (high-dose biotin) for the treatment of progressive multiple sclerosis: a randomised, double-blind, placebo-controlled study. Mult Scler. 2016;22(13):1719–1731. PubMed
19. Cree BAC, Cutter G, Wolinsky JS, et al. Safety and efficacy of MD1003 (high-dose biotin) in patients with progressive multiple sclerosis (SPI2): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Neurol. 2020;19(12):988–997. doi:10.1016/S1474-4422(20)30347-1. PubMed · Registry: NCT02936037
20. Li D, Radulescu A, Shrestha RT, et al. Association of biotin ingestion with performance of hormone and nonhormone assays in healthy adults. JAMA. 2017;318(12):1150–1160. doi:10.1001/jama.2017.13705. PubMed · Free full text
21. Biscolla RPM, Chiamolera MI, Kanashiro I, Maciel RMB, Vieira JGH. A single 10 mg oral dose of biotin interferes with thyroid function tests. Thyroid. 2017;27(8):1099–1100. doi:10.1089/thy.2016.0623. PubMed
22. Trambas CM, Sikaris KA, Lu ZX. A caution regarding high-dose biotin therapy: misdiagnosis of hyperthyroidism in euthyroid patients. Med J Aust. 2016;205(4). Link
23. Assessment of biotin interference in thyroid function tests (healthy-adult washout study). Free full text
24. Li D, Ferguson A, Cervinski MA, et al. AACC guidance document on biotin interference in laboratory tests. J Appl Lab Med. 2020;5(3):575–587. doi:10.1093/jalm/jfz010. PubMed · Summary page
25. U.S. Food and Drug Administration. UPDATE: The FDA warns that biotin may interfere with lab tests: FDA safety communication (November 5, 2019; original communication November 28, 2017). Archived copy
26. U.S. Food and Drug Administration. Biotin interference with troponin lab tests: assays subject to biotin interference (content current as of June 21, 2022). FDA page
27. National Institutes of Health, Office of Dietary Supplements. Biotin: fact sheet for health professionals (updated January 10, 2022). ODS page
28. Registered trials without retrievable results, listed for transparency: NCT06605768, NCT07520019, IRCT 50577.