The side-effect record, sorted by what kind of evidence it rests on

Almost everything written about finasteride's side effects fails in the same way: it takes one kind of evidence and presents it as the whole picture. A page that quotes only the trial table makes the drug sound trivially safe. A page that quotes only the reports makes it sound catastrophic. Both are quoting real documents. This page keeps the three kinds of evidence in separate rooms and says what each can and cannot establish.

Evidence type one: the controlled trials

In the twelve-month hair-loss trials, investigators recorded which adverse experiences they judged drug-related. This is the only part of the record with a placebo group, which makes it the only part that can establish a rate caused by the drug rather than a rate observed alongside it.

Reduced sex drive 1.8% against 1.3% on placebo; erectile dysfunction 1.3% against 0.7%; ejaculation disorder 1.2% against 0.7%; at least one of the three, 3.8% against 2.1%. The label adds four qualifications that get dropped when these numbers travel: 1.4% of men on the drug stopped because of a drug-related adverse experience, against 1.6% on placebo; the effects resolved in men who discontinued and in most who continued; yearly incidence fell to 0.3% or below by the fifth year; and a more sensitive questionnaire found differences in favour of placebo in three of four domains at twelve months, but no significant difference in overall satisfaction with sex life.

One thing the trials did not find: breast tenderness and enlargement, hypersensitivity reactions and testicular pain occurred no more often than on placebo in the hair-loss studies.

Evidence type two: post-approval reports

The second section of the label lists what has been reported since approval. The label states its own limitation in the first sentence: the reports come from a population of uncertain size, so it is not always possible to estimate frequency or establish that the drug caused anything. Read that as it is meant — not as proof, and not as nothing.

• Sexual dysfunction that continued after treatment was stopped, including erectile dysfunction, libido disorders, ejaculation disorders and orgasm disorders.
• Male infertility and poor semen quality, with normalisation or improvement reported after stopping; also testicular pain and blood in the semen.
• Depression, and suicidal ideation and behaviour.
• Male breast cancer, and breast tenderness and enlargement.
• Hypersensitivity reactions including rash, itching, hives and swelling of the lips, tongue, throat and face.

Notice what the first line does. The phrase "that continued after discontinuation of treatment" is in the FDA-approved labelling of the drug. People arguing that persistent effects are an internet invention are arguing with the label.

Evidence type three: what regulators concluded

Four regulators have now written about this, and their documents are the most useful thing in the whole subject, because a regulator has to say what it thinks the evidence supports and then act on it.

The record runs from the two American approvals — the prostate tablet on 19 June 1992, the hair-loss tablet on 19 December 1997 — through a long quiet period, and then four documents in under three years, including the FDA's alert on compounded topical products on 22 April 2025.

The UK went first and hardest. On 29 April 2024 the MHRA published a safety update reminding prescribers of psychiatric and sexual side effects, including the potential for sexual dysfunction to persist after treatment has stopped, and introduced a patient card for every pack. Its own Yellow Card totals: 426 reports of finasteride and sexual dysfunction since 1992, with the outcome recorded as not recovered or not resolved in almost half of them, and 281 reports of depressed mood disorders and suicidal or self-injurious behaviour since 1993.

The EU review followed. The Pharmacovigilance Risk Assessment Committee examined trial data, the EudraVigilance database and the literature, and the European Commission issued a binding decision on 22 August 2025. Suicidal ideation was confirmed as a side effect of finasteride tablets, at a frequency the EMA describes as unknown — it cannot be estimated from the available data. The review identified 325 relevant cases of suicidal ideation, 313 for finasteride and 13 for dutasteride with one reported for both, set against an estimated exposure of around 270 million patient-years for finasteride and around 82 million for dutasteride. Most concerned patients treated for hair loss.

Then the UK again, on 11 May 2026, after reviewing the European referral, its own reports and other regulators' actions. Its expert group's phrase for the published studies is the most honest sentence written about this drug by anyone: the literature data showed mixed outcomes. The product information is being updated to say that sexual dysfunction may contribute to mood disorders, and that sexual dysfunction has also been reported without mood alterations. Yellow Card figures given in that update run to 31 May 2025: 170 reports of suicidal ideation and related terms for finasteride, of which 19 were fatal reports of suicide, and 5 reports for dutasteride with no fatal reports.

What the regulators told people to do

This is advice from named bodies, quoted as theirs. It is not this site's advice, and it does not replace your prescriber's.

1. Patients prescribed the hair-loss tablet should stop taking it and contact a doctor as soon as possible if they develop depression or suicidal thoughts (MHRA, 11 May 2026).
2. Patients on the prostate tablet or on dutasteride should contact a doctor as soon as possible in the same circumstances, without the instruction to stop.
3. Anyone experiencing sexual dysfunction on either medicine should contact their doctor about it.
4. Prescribers should ask about a history of depression or suicidal ideation before prescribing, and review patients for both kinds of effect afterwards.
5. Suspected reactions should be reported — through the Yellow Card scheme in the UK, and through MedWatch in the US.

The effects that are not about sex or mood

They get less attention and two of them matter for other appointments entirely.

• PSA is lowered. The label reports mean PSA falling from 0.7 to 0.5 ng/mL over twelve months in men aged 18 to 41 on the hair-loss tablet, and roughly halved in older men on the prostate tablet. Any confirmed rise from your lowest value while taking it should be evaluated even if it is inside the normal range. Tell whoever orders a PSA test that you take this.
• High-grade prostate cancer in a prevention trial. Over seven years at five times the hair-loss strength, Gleason 8-10 cancers were found in 1.8% of men on finasteride against 1.1% on placebo; a four-year trial of dutasteride showed 1.0% against 0.5%. The label says the significance for men taking the hair-loss tablet is unknown.
• Breast changes. Tenderness, enlargement and, in post-approval reports, male breast cancer — with the label stating that the relationship between long-term use and male breast tumours is currently unknown.

What nobody can tell you

Three questions are asked constantly and have no published answer. How often the lasting form occurs: the EMA calls the frequency of suicidal ideation unknown, and no regulator has published an incidence for persistent sexual dysfunction. Who is at risk: no predictive test or risk factor has been established. Whether stopping early changes the odds: not established either, although the regulators' instruction to stop and seek advice at the first sign of mood change is the nearest thing to a position on it.

If anyone quotes you a precise percentage for post-finasteride syndrome, ask which document it came from. That page sets out what is and is not known, and who says which.