Saw palmettoSerenoa repens

The number that explains everything

In the 72-week trial that forms the backbone of the modern evidence, researchers counted how many men improved by at least three points on the standard symptom score, which is the threshold for a change a man would actually notice.

On saw palmetto: 42.6 percent. On placebo: 44.2 percent.

Nearly half the men taking nothing got meaningfully better. That single pair of numbers explains both why saw palmetto has sold for decades on genuine, sincere testimonials, and why it keeps failing when tested properly. Prostate symptoms fluctuate, men seek help when symptoms are at their worst, and things drift back toward average on their own. Without a placebo group you cannot tell that apart from a working treatment, and for most of saw palmetto's history nobody had one.

What the evidence actually shows

The 2012 Cochrane review pooled 32 randomised controlled trials covering 5,666 men, with trial lengths from 4 to 72 weeks. Twenty-seven were double-blinded, though allocation concealment was adequate in only 14. The 2023 update narrowed the question to placebo comparisons, 27 studies and 4,656 men, and rated the evidence that saw palmetto makes little to no difference to symptoms as high certainty.

Pooling the two high-quality long-term trials, 582 men, saw palmetto was not superior to placebo at reducing symptoms: a mean difference of 0.25 points on the AUA symptom index, with a confidence interval running from minus 0.58 to 1.07. That interval straddles zero, and the effect it is straddling is far too small to feel.

The 72-week trial also tested double and triple the usual dose. Still nothing. It did not significantly reduce night-time urination either.

The reviewers' conclusion is blunt. At double and triple doses saw palmetto did not improve urinary flow, and a one-year trial at the standard dose found no change in prostate size.

Why the early trials were positive

Meta-analyses from around 2000 did find a benefit, and those were not fraudulent. They pooled shorter trials with weaker methods, and when the large, long, well-concealed trials arrived, the effect disappeared.

This is an ordinary pattern in medicine rather than anything peculiar to herbs. Small, short, poorly concealed trials systematically overstate effects. The useful lesson is about reading rather than about saw palmetto: when a supplement's evidence base is many small positive trials and a couple of large negative ones, the large negative ones are usually telling you the truth.

If you are going to take it anyway

Plenty of men will, and that is a reasonable choice to make with the numbers in front of you. Two things make it a better bet.

Use a standardized extract at 320 mg a day, the kind the trials used. That will not make it work where the trials found it did not, but it at least means you are testing the studied material.

And give it a stopping rule. Twelve weeks, with your symptoms written down at the start, and stop if nothing has changed. The trials ran up to 72 weeks without finding anything, so waiting longer is not the answer.

Does it affect a PSA test?

No, on the best evidence. Because saw palmetto is said to act on the same enzyme as finasteride, which halves PSA, it was long assumed to lower PSA too and mask prostate cancer. The CAMUS trial measured it directly. Over 72 weeks, at 320, then 640, then 960 mg a day, PSA rose by 0.23 ng/ml on saw palmetto and 0.16 ng/ml on placebo, a difference that was not significant, and the result held at every starting PSA level.

That fits the rest of the record: a supplement that does not shrink the prostate or improve symptoms in good trials does not behave like finasteride in the blood either. Still list it among your medications, as with any supplement. The caution worth knowing is a different one: a handful of case reports describe bleeding problems around surgery, so mention it to a surgeon. Otherwise it is about as well tolerated as placebo.

What to consider instead

Beta-sitosterol has smaller and older trials but they are consistently positive: better flow and less residual urine across four trials, and a 4.9-point symptom improvement against placebo in the two that measured it, which is more than saw palmetto managed in trials ten times the size.

Rye grass pollen extract has the specific finding on night-time urination: in small, short trials men were about twice as likely to report improvement, which is the symptom most men actually want fixed.

And if symptoms are genuinely bothering you, the drugs work. An alpha blocker produces in weeks what none of these produced in 72.