Thyroid, Hashimoto’s and Hair Loss: What the Evidence Supports

In the largest series of hair loss patients ever published, 3,028 people with diffuse shedding, thyroid dysfunction turned up in 4.6% of them. Vitamin D deficiency turned up in 72.2%. The thyroid gets blamed for hair loss more often than almost anything else, and it is one of the least common explanations.

That isn’t an argument against checking. Thyroid disease does cause hair loss, the mechanism is well characterised, and it is one of the few causes that is fully reversible by treating the underlying condition. It’s an argument for knowing the base rate before you build a theory on it.

What does explain diffuse hair loss

A retrospective review of 3,028 patients with telogen effluvium recorded what their laboratory workups found.^[1]

The mechanism, which is more direct than people expect

The intuitive story is that thyroid hormone controls metabolism, hair is metabolically expensive, so a slow metabolism means less hair. That isn’t wrong, but it undersells what’s happening.

The hair follicle is a direct thyroid hormone target organ. In cultured human follicles, T4 upregulated hair matrix keratinocyte proliferation and prolonged anagen, both T3 and T4 reduced apoptosis, and both stimulated pigment production. The follicles were also shown to transcribe the deiodinase genes D2 and D3, meaning they can convert T4 to active T3 locally.^[2]

A useful way to hold it: androgens are a local council making decisions about specific streets, and thyroid hormone is the national grid. When the grid dips, every street dims at once. That’s why thyroid loss is diffuse and pattern loss is a map.

That mechanism predicts the clinical picture correctly. Thyroid hair loss is diffuse, all over, because a hormone circulating everywhere affects every follicle at once. It isn’t patterned, because there’s nothing regional about it.

Hypothyroid, hyperthyroid, and a surprise

Both directions are associated with hair loss, and they aren’t equal. In 500 women with telogen effluvium, split into 248 euthyroid, 150 hypothyroid and 102 hyperthyroid, the hypothyroid group had significantly more severe hair loss. The hyperthyroid group did not differ significantly from euthyroid.^[3]

Then a primary care study complicates that neatly. Across 2,035 patients tested, the total diagnostic yield of thyroid testing was 2.1%. Hair loss carried an odds ratio of 21 for an abnormal TSH result, which sounds enormous. But hair loss was one of the features associated with a low TSH, meaning overactive rather than underactive.^[4]

Hashimoto’s, and the distinction that matters most

Here’s the thing most articles get wrong, and it changes what you should test for.

Thyroid autoimmunity and thyroid dysfunction are different, and they associate with different hair conditions. A meta-analysis of 17 studies covering 262,581 alopecia areata patients and 1,302,655 controls found alopecia areata significantly associated with thyroid peroxidase antibodies and thyroglobulin antibodies, and not significantly associated with diagnosed or serological hypothyroidism or hyperthyroidism.^[5]

The antibodies separate the groups. The hormone levels don’t. A case-control study of 102 patients reproduced exactly that pattern at individual-study scale: raised anti-TPO in 5 patients versus 0 controls, significant at p equal to 0.024, while thyroid dysfunction occurred in only 2 patients versus 0 controls and was not significant.^[6]

What that reflects is autoimmune clustering. If your immune system is attacking one self-tissue it is somewhat more likely to attack another, and pooled across 18 studies covering 462,945 alopecia areata patients, autoimmune thyroiditis carried an odds ratio of 1.57, well below vitiligo at 6.61.^[10] Note also that alopecia areata and androgenetic alopecia are different diseases, and this autoimmune evidence is specific to the former. I could find no source giving thyroid antibody prevalence in pattern hair loss.

Worth knowing that the literature is not unanimous. A larger review of 50 studies reached a more pro-screening conclusion, finding higher odds of abnormal thyroid function tests and autoantibodies in alopecia areata patients.^[7] The two disagree on what to do clinically, which is itself informative about how strong the underlying signal is.

The ferritin detour, because it is the same mistake

Nearly everyone worried about thyroid and hair is also being told to push their ferritin above 70. The best study on this compared 381 women with hair loss against 76 controls, and the result is worth reading twice.

Among the premenopausal women, at a ferritin threshold of 15 or below, iron deficiency occurred in 12.4% of those with female pattern hair loss, 12.1% of those with chronic telogen effluvium, and 29.8% of the controls. At a threshold of 40, the figures were 58.8%, 63.8% and 72.3%. In that group the controls were the most iron-deficient at both thresholds. Postmenopausally the pattern doesn’t hold, and the numbers run the other way. The finding that survives across both groups is the one in the paper’s conclusion: iron deficiency is common in women and not increased in hair loss patients compared with controls.^[8]

That’s the pattern to internalise. A number can be abnormal in a lot of people with hair loss and still not explain the hair loss, because it is equally abnormal in people without it. Without a control group you can’t tell the difference, and most of the advice in this space has no control group behind it.

What to test, and what to leave alone

Get a TSH. It’s cheap, it’s one tube of blood, and thyroid disease is one of the few reversible causes on the list. The 2.1% yield figure is an argument against expecting it to be the answer, not against looking.

Add TPO antibodies only if there’s a reason: patchy hair loss rather than diffuse, a personal or family history of autoimmune disease, or an abnormal TSH. That is where the antibody evidence applies. Ordering antibodies for ordinary diffuse shedding will mostly generate anxiety.

Don’t chase ferritin on the basis of hair. The controls in the best-designed study were more deficient than the patients.

And be careful about the recovery timeline, because there’s a gap here I should name. I looked for studies reporting how long hair takes to recover after thyroid treatment, and I couldn’t find one. Not a weak study, none. The commonly quoted three to six months follows the normal telogen-to-anagen lag, which is reasonable physiology and is not a published follow-up figure. If someone gives you a confident number, they’re inferring it too.

The last thing, and it’s the most common error I see. If your hair is thinning at the temples and crown in a recognisable pattern, a normal TSH won’t make that go away, and an abnormal one won’t explain it. Thyroid causes diffuse shedding. Pattern loss is a different mechanism entirely, and I have written about telling diffuse thinning apart from pattern hair loss. Both can be present at once, which is exactly why people spend years treating the wrong one.