SAMe Supplements Guide (2026): Depression & Joint Evidence, Dose

By Erin Rose · Updated September 3, 2026 · Methodology · About Us

What the trials actually found

SAM-e is unusual among supplements in having been tested against real antidepressants rather than only against placebo. The results split depending on which question you ask.

As a first-line treatment: no. A Cochrane review pooled eight trials in 934 adults comparing SAM-e with placebo, imipramine, desipramine or escitalopram, and concluded plainly that SAM-e “was not different from placebo and established antidepressants” — with the evidence graded low to very low quality throughout (PMID 27727432). If you are choosing between SAM-e and starting treatment, that is the answer.

As an add-on when an SSRI has not worked: a real signal. A separate six-week trial enrolled 73 people who had already failed to respond to a serotonin reuptake inhibitor and added SAM-e at 800 mg twice daily. Response rates were 36.1% against 17.6% on placebo, and remission 25.8% against 11.7% — a number needed to treat of roughly one in six for response (PMID 20595412). One trial, 73 people, six weeks: not definitive, but it is the most specific finding here, and it describes a specific person rather than everyone.

For osteoarthritis, comparable to a drug in one small trial. A head-to-head against celecoxib found no significant difference between the two by the second month, with both groups improving from baseline (PMID 15102339). SAM-e is slower to act than the drug, which matters if you are in pain now.

Liver support: the claim you will see most, with the least behind it

SAM-e is marketed heavily for liver health, which traces back to its role in methylation rather than to outcome trials. A Cochrane review pooled nine randomised trials in 434 patients with alcoholic liver disease and found no significant effect on all-cause mortality (RR 0.62, 95% CI 0.30 to 1.26) or liver-related mortality (RR 0.68, 0.31 to 1.48). Its conclusion is worth quoting exactly, because it is not the same as “it does not work”: the reviewers “could not find evidence supporting or refuting the use of SAMe” for these patients, and called for long-term trials before it could be recommended (PMID 16625556). Eight of the nine trials were placebo-controlled but methodological quality was generally low, and only one used adequate methodology.

So if you are buying SAM-e for your liver, you are buying an open question rather than a demonstrated benefit — at the prices below, worth knowing.

The dose that worked is four times what the shelf sells

This is the part that decides whether SAM-e is worth it, and it is not on the bottle. We read every SAM-e label the NIH database lists — 377 products, all of them retrieved — and only 27 declared a per-serving amount we could parse. That is 7% of the category, which is poor label transparency even by supplement standards.

So the typical product delivers a quarter of the dose the positive trial used. Nothing on the label tells you that, because the label is not comparing itself to a trial. In practice, reaching 1,600 mg a day means taking four of the 400 mg tablets below — which turns a $0.62–$0.82 daily cost into roughly $2.50–$3.30 a day, or $900–$1,200 a year. That is the honest price of the evidence, and it is the single most useful thing to know before buying.

Safety, and the one signal worth taking seriously

SAM-e is generally well tolerated — in the Cochrane review, fewer people had troublesome adverse effects on SAM-e than on imipramine, and drop-out rates did not differ from placebo. The exception is mood switching: the same review recorded two reports of mania or hypomania among 441 people taking SAM-e, and there is a published case report of apparent SAM-e-induced mania (PMID 29950497). If you have bipolar disorder, or a family history of it, this is a conversation to have with a clinician before starting rather than after.

The three products we track

All three are 400 mg enteric-coated tablets, which is the format the trials used and the only serving size the label census found. None carries third-party certification.

Check price →

Common questions

Does SAM-e work for depression?

Not as a first-line treatment — Cochrane found it no different from placebo or from established antidepressants across eight trials. The one place it has a specific result behind it is as an add-on when an SSRI has already failed, at 1,600 mg a day.

How much should I take?

The trial worth citing used 800 mg twice daily. Most products are 400 mg per tablet, so that is four tablets a day — price it before you commit.

How long before I know?

The add-on trial ran six weeks; the osteoarthritis comparison took about two months to match celecoxib. SAM-e is not a fast-acting option.

Is it safe to take with an antidepressant?

That is exactly the situation the positive trial studied, but it was supervised. Ask your prescriber — combining serotonergic agents is not something to improvise, and the mania signal above is the reason.

Why do so few labels state a dose?

Only 27 of the 377 SAM-e labels in the NIH database declare a per-serving amount our parser could read. Many are blends or state the amount only in prose. It is a category with unusually poor label transparency, which is its own reason to buy a single-ingredient product.

Method: all 377 SAM-e labels the NIH Dietary Supplement Label Database (DSLD) v9 lists were retrieved on 2026-08-29 and read at the Supplement Facts panel; 27 carried a parseable per-serving amount and 25 also directed a daily amount, across 18 brands. What it cannot tell you: what people actually take, or whether the 350 labels without a parseable dose are higher or lower — they are absent from the median, not counted as zero.

About our data

Every comparison uses clinical evidence from PubMed systematic reviews, product label data from the NIH Dietary Supplement Label Database, and third-party certifications (USP, NSF). Products are ranked by cost per clinically-effective dose. See our methodology and editorial standards.