A study that was withdrawn

The Green Coffee Study That Was Retracted, And What Replaced It

One trial made green coffee bean extract famous, and it no longer stands. The ingredient is still on this front panel, and the evidence behind it today is smaller, slower and considerably more trustworthy.

By SodaSlim Team · Last updated November 4, 2026 · 9 min read · A study that was withdrawn · Reviewed by Dr. Lindsey Wells, ND

What people remember about green coffee

Somewhere around 2012, green coffee bean extract became briefly unavoidable. It was on daytime television, in supplement aisles and on the front of a great many bottles that had contained something else the previous year. The numbers attached to it were large and the timescale was short.

What persists from that period is a vague and confident impression: that green coffee does something dramatic. Impressions of that kind are durable precisely because they are vague. Nobody remembers the study design, the dose or the duration, so there is nothing specific to revise.

The left column is what survived in public memory. The right column is what the literature currently supports, which is smaller, slower and still real.

The right-hand column of that table is the honest position, and none of it is as exciting as what the left column claimed. It is also, unlike the left column, supported by pooled randomised evidence.

The trial, and its withdrawal

The study at the centre of the boom was a randomised, double-blind, placebo-controlled linear-dose crossover in overweight subjects, published in 2012 and reporting substantial reductions on green coffee extract in a journal record that today carries the label retracted publication.

It was retracted. The PubMed record for it is categorised as a retracted publication, and that categorisation is returned by the database itself rather than being an editorial characterisation added here.

This website cites it anyway, with the label attached, and the reason is worth stating. A reader who half-remembers the green coffee story is carrying a claim with no ending. Omitting the paper leaves that claim standing and unanswered; citing it and saying what happened is the only version of this section that actually informs anybody.

What replaced it

A decade of more careful work, most of it smaller in its claims and larger in its samples.

The three trials above the line are the current evidence base. The fourth is the one most people remember.

The 2023 meta-analysis is the most useful of these for a buyer. It pooled randomised trials using at least 500 mg a day of chlorogenic acid and reported a mean reduction of about 1.3 kg in body weight against placebo, with no heterogeneity between studies and no publication bias detectable by the usual tests in a systematic review.

A figure of 1.3 kg is not a headline. It is, however, a figure that survived pooling, which is more than most supplement ingredients manage, and the absence of heterogeneity means the individual trials were pointing the same way rather than cancelling each other out.

A broader review of chlorogenic acid across the coffee literature covers the compound's wider effects in a comprehensive review, and a 2017 randomised trial gave 64 obese women 400 mg a day for eight weeks alongside energy restriction measuring serum adipocytokines and body composition.

The four-week floor

The single most practical finding in this literature has nothing to do with the retraction. A dose-response meta-analysis across fifteen randomised trials reported that supplementation periods shorter than four weeks produced no effect at all on any anthropometric measure.

That is a hard floor and it explains a great deal about how people experience this category. A buyer who takes a green coffee formula for ten days and concludes it does nothing has conducted a valid experiment and reached the answer the literature predicted. They have not, however, tested the ingredient.

It also explains why a return window matters more here than the terms themselves suggest. A thirty-day window forces a decision at roughly the point the evidence says to start looking. A sixty-day window lets somebody reach four weeks, form a view, and still send it back.

What that means for this bottle

Green coffee bean extract is one of five plants named on this front panel, and the panel gives no amount for it. The pooled trials are dosed against chlorogenic acid content, not against extract weight, so even a printed extract figure would leave a further question about standardisation.

What a reader can take from this piece is therefore not a conclusion about the capsule. It is a conclusion about the ingredient: it is real, the effect is about 1.3 kg in pooled trials at 500 mg a day of chlorogenic acid, it needs four weeks minimum, and the famous study everybody remembers is not part of the evidence any more.

Whether this particular capsule reaches anything like that dose is a question the artwork cannot answer, and no page on this website pretends otherwise.

How a headline becomes an ingredient

The sequence is worth naming, because it repeats. A single trial reports a striking result. A broadcast or a magazine picks it up, because a striking result is the only kind that makes a story. Manufacturers respond to the demand that follows, and within a year the ingredient is on hundreds of labels.

Nothing in that chain requires anybody to lie. The trial was published, the coverage reported it accurately enough, and a manufacturer adding a popular ingredient is answering a market rather than making a claim. The weakness is structural: the whole edifice rests on one study, and one study is the least reliable unit of evidence there is.

What follows the boom is slower and quieter. Other groups attempt replication, most of them find smaller effects, and eventually somebody pools the lot and publishes a figure nobody reports. By then the ingredient is established and the original number is what people remember.

Green coffee ran that course with an unusually clean ending, because the founding trial was withdrawn outright rather than merely failing to replicate. Most ingredients in this category do not get such a clear correction, which makes this one worth studying.

The transferable lesson

Retractions are rare and the ingredients attached to them are not thereby worthless. What the green coffee episode teaches is narrower and more useful: a single striking trial is the weakest form of evidence that gets treated as the strongest, because it is the one that makes the news.

Pooled analyses of many trials are duller, later and far more reliable, and they almost always report smaller effects than the trial that started the excitement. That pattern repeats across this entire category, and a reader who expects it will be disappointed less often and misled less often.

Applied to any supplement, the rule is: ask what the meta-analysis says, not what the study said.

References