The Saw Palmetto Dose Gap, and Why It Does Not Matter

In Seven of Eleven we drew a single bar on a single row. Saw palmetto was the one botanical on this panel where a published reference amount existed to draw against, roughly 320 mg a day of standardised extract in the trials NCCIH summarises, and the panel declares 84.5 mg. Divide one by the other and the row comes out at about a quarter of the studied amount.

That was a fair thing to point out. It also quietly assumed something. A bar drawn against 320 mg treats 320 mg as the amount that works, and implies the shortfall is the problem. This post checks the assumption, because the published trials happen to answer it directly, and the answer is not the comfortable one for either side of the argument.

If you came here through a search for does saw palmetto work or saw palmetto dosage, the short version is above. The rest is the reading behind it: which trials, what they gave, what they measured, and what the reviews that pool them conclude.

The most-cited answer in the general press comes from a trial published in the New England Journal of Medicine in 2006 (PMID 16467543). Bent and colleagues randomised 225 men over the age of 49 with moderate-to-severe symptoms of benign prostatic hyperplasia to one year of saw palmetto extract, 160 mg twice a day, or placebo. That is 320 mg a day, exactly the reference amount the bar was drawn against.

The primary outcomes were the American Urological Association Symptom Index and the maximal urinary flow rate. Neither moved. The mean difference in symptom score was 0.04 points (95% CI −0.93 to 1.01), and in flow rate 0.43 mL per minute (95% CI −0.52 to 1.38), with the confidence intervals comfortably straddling zero. Prostate size, residual urine volume, quality of life and PSA were also unchanged, and side effects were similar in the two groups. The authors’ conclusion was blunt: saw palmetto did not improve symptoms or objective measures of benign prostatic hyperplasia.

So at the reference amount, in a properly blinded year-long trial of 225 men, the herb did not beat placebo. It is worth being fair about one thing: one trial, however well run, is one trial. Which is why the next study matters.

A single trial can mislead, and so can two, so it is worth checking what the systematic reviews do when they pool everything. There are three worth reading, from three different eras.

The Cochrane review updated in 2012, published in BJU International, pooled 17 randomised trials involving 2,008 men, with a typical dose of 320 mg a day (PMID 22551330). Its conclusion is the one to hold onto: in a meta-analysis of the three high-quality, moderate-to-long-term trials (661 men), saw palmetto was no better than placebo on the symptom index (weighted mean difference −0.16 points, 95% CI −1.45 to 1.14) or on peak flow (0.40 mL/s, 95% CI −0.30 to 1.09). It also spelled out the dose-escalation finding in plain terms: double and triple doses did not improve the score, and the proportion of responders was 43 per cent on saw palmetto and 44 per cent on placebo, a risk ratio of 0.96. One place it found something was in short-term trials of a single branded extract, with nine trials pooled showing fewer trips to the toilet at night (−0.79 a night, 95% CI −1.28 to −0.29), though it flagged substantial heterogeneity and no such effect in the long-term high-quality trial.

The 2023 Cochrane update, by Franco and colleagues, included 27 studies and 4,656 participants (PMID 37345871). It narrowed its question to comparisons against placebo, and its main results rest on sensitivity analyses restricted to studies at low risk of bias. On that basis, saw palmetto results in little to no difference in urinary symptoms at three to six months (IPSS mean difference −0.90, 95% CI −1.74 to −0.07, nine studies, high-certainty evidence) and at twelve to seventeen months (mean difference 0.07, 95% CI −0.75 to 0.88, three studies, high-certainty evidence). Adverse events probably did not differ (risk ratio 1.01, 95% CI 0.77 to 1.31, moderate certainty). Its authors’ conclusion is unambiguous: saw palmetto alone “provides little to no benefits” for men with lower urinary tract symptoms due to benign prostatic enlargement. The count of 27 studies matches the 2023 review that NCCIH’s saw palmetto page mentions, the one quoted in our earlier post.

With the evidence laid out, the row can be read three ways, and it is worth being clear about what each does and does not support.

If the trials are right, and Cochrane has now said so with high certainty, the amount is almost beside the point for the outcome those trials measured. The row is not shy of a working dose, because no dose of the extracts tested has been shown to work for that outcome. That makes our own bar in the earlier post accurate on the arithmetic and misleading on the implication, and we would rather say so than leave it standing.

If you are asking a different question, the trials say nothing. Every study in this post was in men with lower urinary tract symptoms. The label makes no such claim, and this site does not either. The who it is for page lists buying for urinary symptoms among the reasons this bottle does not suit, and that judgement is consistent with everything above. A man with urinary symptoms needs a clinician, because, as MedlinePlus puts it in the passage quoted on this site’s footer, the symptoms “could be from a more serious health problem”.

If you are asking about the rest of the formula, this is the row where the sums are most tidy, and it is not the row that carries the panel. The ginseng post does the same exercise for the one botanical with a positive federal sentence, with a different result: there the trials gave twenty-odd times the label’s dose, and the effect was small but real. Comparing the two rows tells you something about the panel as a whole. Two botanicals with a federal page and a trial literature behind them tell the same broad story: unproven at the printed amount in one case, and unproven at any amount in the other.