Dietary Supplements for Body Composition: Here is the evidence

Our evidence-based analysis features 7 unique references to scientific papers.

Written by Kamal Patel
Last Updated:

In April of this year, a review was published titled "Dietary supplements for improving body composition and reducing body weight: where is the evidence?"[1]

Since the publication of this article, two main things have happened in my life that are directly related to its publication:

  • Numerous amounts of people have contacted me about the 'lack of evidence' for the supplements mentioned in this article

  • I have acquired a new example of "Why you should never take a review article written by a single author as 'evidence' for anything"

Review articles are wonderful things in the sense that they are educated people rambling. It could be very insightful, or the very opposite of that; there really is no standard format.

This review is the latter. In my opinion, the ratio of crap to insightful-thoughts skews too heavily on the crap end. This was seen right in the abstract when the author states that no supplement has ever been shown to produce 'significant weight loss' and randomly quantifies this as 2kg or more. No time frame was given in the entirety of the paper. This allows one to merely shorten the time frame until it falls under the 'not significant' threshold.

The article has the general format of making four categories that fat burners fall under (nutrient blockers, stimulants, nutrient partitioners, appetite suppressants) and lists numerous supplements for each section.

  • In the nutrient blocker section, Phaseolus vulgaris and Chitosan are reviewed. Phaseolus (white kidney bean extract) was said to possess two studies[2][3](which are the extent of Pubmed's database on the topic) and the author claims this is insufficient evidence; a good claim. Chitosan was subject to a meta-analysis[4] that noted statistically significant weight loss of 1.7kg in the trials, but the trials were claimed to be subpar quality. At least in this section, the author seems to accurately report what the literature says.

  • The stimulant section is where it begins to go downhill, with the introduction to the section not talking about what makes a molecule a stimulant but instead opting to discuss ephedrine being banned in sports and citing some studies that conclude that overdosing is a bad idea. After briefly talking about how caffeine can increase the metabolic rate synergistically with ephedrine or nicotine it is then tossed aside in a single paragraph due to a lack of long term evidence. The notion that caffeine can benefit short term fat loss period is lost, despite touting the increases in metabolic rate (which would yield ... fat loss). Ephedra was next, and right after stating what it was the next topic to be mentioned was how there was a 10-fold increase in reports to toxicology centers from 1993 to 2002 (roughly the time period where ephedra became popular and had its usage increase drastically). After brief scaremongering about how it is still available online and citing 5 review articles, the author then cites a single meta analysis that shows 0.6-1kg additional weight loss per month (which would meet her significant criteria by month two) but at the cost of 'stressing the cardiovascular system'. The Meta-Analysis in question[5] did cite higher autonomic and psychonomic side-effects, but claimed insufficient evidence "to draw conclusions about adverse events occurring at a rate less than 1.0 per thousand". Immediately after this meta-analysis, she cites case studies and news reports of teenage athletes using ephedra to enhance performance, and finishes with the 2003 death of a certain baseball pitcher that was confounded with a variety of variables such as heat exhaustion and overreaching (close to overtraining).

  • Perhaps the most glaring bias is when the author examines the evidence for 'ephedra + caffeine', the popular EC stack. She cites two studies that noted "6–9 months (of EC) reported greater weight loss (3–5 kg) and fat loss (1.5–4.3 kg)" that clearly surpasses her own standard of what is significant. Two studies were reviewed[6][7]. To quote Hackman 2006 (her second source): "A dietary supplement containing a low potency ephedra/caffeine mixture appeared safe and effective in causing loss of weight and body fat, and improving several metabolic parameters, including insulin sensitivity and lipid profiles". Apparently dry mouth was bad enough to disregard the decrease in triglycerides, cholesterol, and improvements in lipid levels and insulin sensitivity from the weight loss? If we Examine the ephedra page we have here, we have 10 studies on metabolic rate and 17 and 8 on weight and fat mass in particular, respectively. I won't attempt to say that ephedra and ephedrine are side effect free, as stimulants are not; but disregarding so many trials to find two that can be found by putting 'safety' into Pubmed and looking for ones that reported something bad (whether or not the study authors thought it mattered) is not good review science. It is bias.

  • Green tea Catechins was done well for the most, mainly as the science for it being an agent capable of inducing fat loss is quite context dependent (it seems to work vicariously through caffeine, and loses efficacy during caffeine tolerance). However, apparently 34 reported case studies of green tea consumption and liver damage (no causation linking these two, just reports) paired with the fact that they are unregulated in the US is enough to disregard green tea supplements overall? It pains me that I even have to point out that correlation is not causation.

  • The other compounds studied that I have no qualms with in regards to their analysis are chromium picolinate (admittedly crap for weight loss), Conjugated Linoleic Acid (demonstrated to be crap for weight loss), and Calcium (which the evidence for its links to fat metabolism and fat loss were deemed preliminary, but no side effects noted). Dietary fiber was reviewed for the sole inclusion under appetite suppression and was noted to increase fat loss by 1kg relative to placebo (which is something, but insignificant according to the time-undefined '2kg') Studies on glucomannan in particular were said to induce '1-2kg weight loss', which is clinically relevant in the context of a lifestyle approach, but were ignored. No side effects were noted with soluble fibers.

Now, given the above breakdown; how do her claims stack up?

While some weight-loss supplements produce modest effects (<2 kg weight loss), many have either no or few randomized clinical trials examining their effectiveness

Yes, but it would help to do a comprehensive review on the ones you deem 'bad' (ephedrine), and it would also help to say what time frame this '2kg' loss is supposed to happen in, because ephedrine sure as hell passes that in 2 months independent of diet.

A number of factors confound research results associated with the efficacy of weight-loss supplements

Yes, this is inherent due to the preliminary nature of things. The author's background is in public health prevention, it is understandable that sample sizes can be small relative to that. Some of the ephedrine studies have over 100 persons (enough to draw conclusions) and it is quite expensive to do metabolic chamber testing on that sample size.

There is no strong research evidence indicating that a specific supplement will produce significant weight loss (>2 kg), especially in the long term

I will just outright state wrong on this statement, using ephedrine as an example.

Even then, a multimodal approach (green tea, soluble fibers, ephedrine + caffine) is just adding their fat loss totals together; it adds up in a practical sense.

I will definitely concede that some supplements are crap; good on you for attacking Conjugated Linoleic Acid, because I drew the same conclusions on that one.

Weight-loss supplements containing metabolic stimulants (e.g., caffeine, ephedra, synephrine) are most likely to produce adverse side effects and should be avoided

Most likely? I hope these conclusions aren't drawn from that study citing insufficient evidence to draw conclusions occuring at greater than a 1 in 1000 rate, or was it from the one that noted decreases in blood pressure and improvements in lipid parameters secondary to weight loss? I'm not sure which of the two studies cited out of 15+ interventions noting weight loss you are referring to.


Being skeptical is the right way to go. This review draws false conclusions based on improperly defined criteria and fear-mongers too much on the supplements that are close to, or clearly do, surpass the criteria put forth. It seems to cite very little in the sections where the science does note significant fat burning effects, and drops a plethora of citations on the supplements (chromium, CLA) that have been shown to be quite ineffective.

Take down the supplements that are worth taking down; don't attack all supplements under the assumption that they're bad and not needed.

I understand not everybody likes stimulants and ephedra/ephedrine; that is fine and everybody has their own stance on the issue that isn't inherently wrong. That being said, don't omit facts and say it's not effective for fat loss.

💊 Get unbiased supplement information