Minoxidil works for female pattern hair loss. Finasteride, in three trials, did not

Hair loss in women is a different condition from hair loss in men — different pattern, different diagnostic workup, and a different evidence base, most of which is weaker. Topical minoxidil has moderate-quality randomised evidence behind it in women. Oral finasteride, which works in men, did not beat placebo in the three trials that tested it in women. And spironolactone, routinely listed as a first-line option, has never been through an adequate randomised trial at all.

What female pattern hair loss is

It is the most common type of hair loss in women, characterised by progressive shortening of the growth phase across successive hair cycles and progressive follicular miniaturisation, in which thick terminal hairs are replaced by soft, fine, short vellus hairs [s1]. Prevalence rises steeply with age: about 12% of women have symptoms by 29, 25% by 50, and 41% to 50% at 70 and over [s2].

The pattern is the first thing that distinguishes it clinically. Women typically present with central scalp or diffuse thinning, whereas male pattern hair loss usually begins at the temples or crown [s2]. The frontal hairline may or may not be preserved [s1].

The main differential is telogen effluvium — diffuse shedding triggered by psychological stress, weight loss or iron deficiency [s2]. A hair-pull test helps separate the two: if six or more hairs come away easily when a small bundle is pulled, the test is positive, and in female pattern hair loss it is usually negative or positive only over the central scalp [s2].

Two points in the diagnostic guidance run against common practice. Androgen levels are usually normal and are not helpful [s2]. Ferritin and thyroid-stimulating hormone should be reserved for cases where an alternative diagnosis is suspected, rather than ordered routinely, and scalp biopsy is not usually indicated [s2].

The randomised evidence

A Cochrane review included 47 randomised trials covering 5,290 participants [s1]. Only five were judged at low risk of bias; 26 were unclear and 16 were at high risk [s1]. That distribution governs how much weight any single number below can carry.

Seventeen of the trials evaluated minoxidil. Pooling six studies, 157 of 593 participants on minoxidil reported a moderate to marked increase in regrowth against 77 of 555 on placebo — a risk ratio of 1.93 (95% confidence interval 1.51 to 2.47), rated moderate quality [s1]. Seven studies with 1,181 participants produced a concordant investigator-rated estimate of 2.35 (1.68 to 3.28) [s1]. Across eight studies with 1,242 participants, minoxidil increased total hair count by 13.18 hairs per cm² over placebo (10.92 to 15.44), though the review graded that estimate low quality [s1]. Adverse events did not differ significantly from placebo at any concentration tested [s1].

Concentration made no measurable difference. Four trials with 1,006 participants comparing 2% with 5% minoxidil found no difference in the proportion reporting improvement, no difference in adverse events (risk ratio 1.02, 0.91 to 1.20), and no difference in total hair count at study end across three trials with 631 participants [s1].

Finasteride is the clean negative. Three studies compared finasteride 1 mg with placebo: 30 of 67 participants improved on the drug against 33 of 70 on placebo, a risk ratio of 0.95 (0.66 to 1.37), with investigator assessments agreeing at 0.77 (0.31 to 1.90) [s1]. The review's conclusion is that finasteride was no more effective than placebo, on low-quality evidence [s1]. Those trials are small, so this is failure to demonstrate an effect rather than demonstration of no effect — but it is the opposite of the situation in men.

Low-level laser combs came out inconsistent. Two studies with 141 participants found no participant-rated advantage over a sham device (risk ratios 1.54, 0.96 to 2.49; and 1.18, 0.74 to 1.89, both moderate quality), while the same studies found an increase in hair count from baseline in favour of the device (mean differences 17.40 and 17.60, low quality) [s1]. Two measurements of the same trials disagreeing is a reason for caution, not a reason to pick the flattering one.

The gap under the guidance

Current clinical guidance lists topical minoxidil and oral spironolactone as first-line, with oral minoxidil, finasteride, low-level laser therapy, platelet-rich plasma and oral contraceptives as second-line, and notes that response typically takes six to twelve months and may be preceded by transient shedding [s2].

Spironolactone's presence on that list is not supported by randomised trials. The Cochrane review lists it first among the widely used treatments for which further randomised controlled trials are needed, alongside finasteride at other doses, dutasteride, cyproterone acetate and laser-based therapy [s1].

What exists instead is a systematic review pooling randomised and observational studies of oral spironolactone, which reported an overall improvement rate of 56.60%, higher in combination therapy (65.80%) than monotherapy (43.21%), with significant heterogeneity, and with hair loss not improving or improving only modestly in 37.80% of patients [s3]. Reported adverse events included scalp pruritus or increased scurf (18.92%), menstrual disorders (11.85%), facial hypertrichosis (6.93%) and drug discontinuation (2.79%) [s3]. The authors' own conclusion calls for further research to understand the heterogeneity of response [s3].

An uncontrolled improvement rate of 43% in a condition where placebo arms in the minoxidil trials produced moderate-to-marked improvement in 77 of 555 participants is not a comparison anyone can make confidently [s1] [s3]. It is the reason Cochrane asked for the trial.

What is not in dispute

Female pattern hair loss carries psychological consequences, and current guidance recommends screening for anxiety and depression on that basis [s2]. Most of the pharmacological treatments are contraindicated in pregnancy [s2]. Neither of those points depends on which drug turns out to work best.

This article is informational and is not medical advice.

Sources

• [s1] Interventions for female pattern hair loss — Cochrane Database of Systematic Reviews, 26 May 2016. https://doi.org/10.1002/14651858.CD007628.pub4
• [s2] Female pattern hair loss — CMAJ, 2 March 2026. https://doi.org/10.1503/cmaj.251058
• [s3] The Efficacy and Safety of Oral Spironolactone in the Treatment of Female Pattern Hair Loss: A Systematic Review and Meta-analysis — Cureus, 16 August 2023. https://doi.org/10.7759/cureus.43559