A Guide To Metabolism Supplements: What Is In Them And What They Do

Three of them, and a formula in this aisle is almost always pulling at least one.

The first is energy expenditure. Raise the rate at which the body spends energy and, all else equal, the daily balance shifts. Stimulants do this most reliably, which is why caffeine appears in more of these bottles than anything else.

The second is appetite. Eat less and the same balance shifts from the other direction. This is also where the most durable results in the wider weight literature come from, which is worth remembering when reading a label that emphasises burning over eating.

The third is absorption and handling, meaning what happens to sugar and fat after a meal rather than before it. Chlorogenic acid works here, as do fibres such as glucomannan, and the mechanisms are slower and less dramatic than the first two.

A formula that claims all three at once, as many do, is claiming each of them weakly rather than all of them strongly. That is not dishonest, and it is the single most useful thing to understand about the category.

Caffeine is the best-evidenced compound in this field by a considerable margin. A dose-response meta-analysis of randomised trials found intake associated with reductions in weight, body mass index and fat mass, scaling with dose across the pooled data. It acts inside an hour, it reduces appetite for a few hours, and its pharmacology is understood well enough that a reader can plan around it.

It also has a ceiling that anybody buying a stimulant formula should know. Two safety reviews converge on 400 mg a day from all sources for healthy adults in a systematic review of adverse effects and in the earlier review behind it, and a controlled crossover found 400 mg disturbing sleep as far out as twelve hours before bed while 100 mg was tolerable at four in 23 men.

The practical consequence is that a caffeinated formula is a different proposition for a heavy coffee drinker than for somebody who drinks none, and any bottle that does not print its caffeine content is asking a reader to guess at an arithmetic that matters.

The founding measurement gave subjects 90 mg of EGCG with 50 mg of caffeine three times a day and recorded a rise of about 4 per cent in 24-hour energy expenditure inside a respiratory chamber, with a shift toward fat oxidation against placebo and against caffeine alone. That result is the reason catechins appear in this category at all.

Pooled work since has been consistent and unexciting. A meta-analysis found the thermogenic effect real across studies of catechin-rich teas with caffeine, while a Cochrane review of green tea preparations for weight loss judged the effect small and not clinically important across fourteen trials. Another analysis found the effect weaker in people whose habitual caffeine intake exceeded 300 mg a day in its subgroup analysis.

There is a safety literature attached to concentrated extract that does not attach to the drink. A safety assessment proposed 300 mg of EGCG a day as a supplement ceiling on hepatic grounds, and a study of confirmed cases found one HLA allele strongly over-represented among those who reacted suggesting individual susceptibility. Two products both carrying green tea extract should not be taken together.

Green coffee bean extract is made from the unroasted seed because roasting destroys most of the chlorogenic acid. Pooled trials at 500 mg a day and above found a mean difference of about 1.3 kg against placebo, with no heterogeneity and no detectable publication bias in a 2023 meta-analysis.

The most useful finding in this literature is about time rather than dose. A dose-response analysis across fifteen trials reported that supplementation periods shorter than four weeks produced no measurable effect at all on any anthropometric measure. A second dose-response analysis examined the same measures across adults in fifteen trials, and an eight-week trial gave 64 obese women 400 mg a day alongside energy restriction measuring body composition.

This ingredient also carries the category's most famous retraction. A 2012 crossover reported large losses on green coffee extract; the PubMed record for it is labelled a retracted publication and carries that label today. Anybody who remembers the green coffee boom of the early 2010s is remembering that trial.

Garcinia cambogia has persisted in this category for three decades on a clean-sounding mechanism: hydroxycitric acid inhibits an enzyme used to convert surplus carbohydrate into fat.

The largest randomised test of that idea gave 1,500 mg a day for twelve weeks to overweight adults, against placebo, with both arms on a high-fibre, low-energy diet. Both groups lost weight and neither separated from the other on weight or on fat mass in JAMA. A later pool of nine trials found 0.88 kg and called its clinical relevance uncertain in a systematic review.

There is also a safety literature worth knowing about. The Drug-Induced Liver Injury Network described moderate to severe injury associated with garcinia alone or with green tea across its case series, and individual reports describe acute liver failure in one case and a transplant in another. These are rare, and they are the reason anyone with existing liver disease should treat this category differently from a multivitamin.

Every category has some, and this one's best-known example is raspberry ketone. It is a single aroma compound, and there is no randomised controlled trial of it in people.

The published work is in mice. A ten-day study gave 165, 330 and 500 mg per kilogram of body weight and reported substantial mortality at the two higher doses in obese animals. A meal-pattern study found the compound changed how the animals ate rather than acting on fat directly in mice, and a bioavailability study measured how much of an oral dose survives digestion in mice.

Animal findings are not human findings, and a formula that presents mouse data in human language has crossed a line the research itself is careful about. The reasonable position for a reader is to treat such an ingredient as unproven rather than as disproven, and to weigh the rest of the bottle accordingly.

The honest summary is uncomfortable for the marketing and useful for a buyer. The best-evidenced ingredients here produce group average differences against placebo of roughly one to one and a half kilograms, measured over eight to twelve weeks, in people who were also eating to a plan.

That figure is small. It is also real, replicated and measured, which is more than most of the category can say, and it is on top of what the eating pattern was doing anyway. Nobody in these trials was told to eat as they pleased.

Two conclusions follow. The first is that a supplement in this category is an addition to a plan rather than a substitute for one, and a label that implies otherwise is claiming something its ingredients cannot support. The second is that individual variation around a group average this small is wide enough that any one person's month tells them very little.

The NIH Office of Dietary Supplements keeps a plain summary of this literature for consumers, and the NCCIH maintains one for green tea specifically including its safety. Both are worth reading beside any product page, including this one.