AUA 2026 establishes time-dependent sexual side effect risks for finasteride
Here's the paradox we keep navigating: A popular drug to treat hair loss, finasteride, can reduce the size of the prostate and regrow hair, yet a subset of men report sexual side effects that outlast the prescription. For physicians, the tension is sharpest when a 28-year-old asks about hair loss and you're weighing a 1-mg pill against an uncertain but real risk of persistent erectile dysfunction.
Why the evidence has been so hard to interpret
The challenge has always been the quality of the evidence. Prior data on finasteride's sexual side effects in young men lacked well-controlled denominators and time-to-event analyses in the specific population most commonly requesting the drug: reproductive-aged men with androgenetic alopecia (AGA). What clinicians needed was a matched cohort study with enough statistical power to separate signal from noise.
That's where evidence coming out of AUA 2026 contributes something concrete. A propensity-matched cohort study using the TriNetX network examined 5,091 men aged 18–45 with AGA prescribed finasteride 1 mg, matched 1:1 to AGA controls on age, BMI, race, and key comorbidities including depression and anxiety.
ED signal appears later
At 6 months, ED risk was statistically indistinguishable (RR 1.32, 95% CI 0.85–2.06). But by 1 year, finasteride users had significantly higher ED incidence—1.61% vs 0.96% (RR 1.67, p=0.004)—and by year 3 that gap persisted at 3.73% vs 2.36% (RR 1.58, p=0.0001). A sensitivity analysis against oral minoxidil showed a similarly elevated risk of PDE5i prescriptions at 3 years (RR 2.19, p=0.007). The absolute numbers are modest, but the signal is real, time-dependent, and now grounded in a decent-sized matched cohort rather than case reports.
An individualized approach
Practically, this sharpens informed consent. For the young AGA patient, you're quoting roughly a 1-in-27 chance of new-onset ED (defined by diagnosis code or PDE5i prescription) by 3 years—low, but not negligible for someone whose baseline risk is near zero.
For the older patient with benign prostatic hyperplasia (BPH) already carrying vascular comorbidities, finasteride's benefit in reducing progression and avoiding surgery may easily justify a similar or slightly higher relative risk of sexual side effects (and the AUA 2026 BPH data showed an RR of just 1.07 for ED with 5-ARIs in that population).
Before prescribing, make the patient’s baseline risk, treatment goal, and tolerance for a 3-year sexual side-effect signal explicit—and document that discussion.