Even before the microbiome craze — the hope that the bacteria in your gut holds the key to good health — people were ingesting cultures of living microorganisms to treat a host of conditions. These probiotics have become so popular that they are being marketed in foods, capsules and even beauty products.

Probiotics have the potential to improve health, including by displacing potentially harmful bugs. The trouble is that the proven benefits involve a very small number of conditions and probiotics are regulated less tightly than drugs. They do not need to be proved effective to be marketed and the quality control can be lax.

In a recent article in JAMA Internal Medicine, Pieter Cohen, an associate professor of medicine at Harvard Medical School in the US, urges us to consider the harms as well as the benefits. Among immune-compromised individuals, for instance, probiotics can lead to infections.

Consumers cannot always count on what they are getting. From 2016 to 2017, the American Food and Drug Administration inspected more than 650 facilities that produce dietary supplements and determined that more than 50 per cent of them had violations. These included issues with purity, strength and even the identity of the promised product.

Probiotic supplements have also been found to be contaminated with organisms that are not supposed to be there. In 2014, such a supplement’s contamination arguably caused the death of an infant. Given all of this, what are the benefits? The most obvious use of probiotics would be in the treatment of gastrointestinal disorders, given that they are focussed on gut health. There have been many studies in this domain, so many that early this year the journal Nutrition published a systematic review of systematic reviews on the subject.

The takeaway? Certain strains were found useful in preventing diarrhoea among children being prescribed antibiotics. A 2013 review showed that after antibiotic use, probiotics help prevent Clostridium difficile-associated diarrhoea. A review focused on acute infectious diarrhoea found a benefit, again for certain strains of bacteria at controlled doses. There is also evidence that they may help prevent necrotising enterocolitis (a serious gastrointestinal condition) and death in preterm infants.

Those somewhat promising results — for very specific uses of very specific strains of bacteria in very specific instances — are just about all the “positive” results you can find.

Many wondered whether probiotics could be therapeutic in other gastrointestinal disorders. Unfortunately, that does not appear to be the case. Probiotics did not show a significant benefit for chronic diarrhoea. Three reviews looked at how probiotics might improve Crohn’s disease and none could find sufficient evidence to recommend their use. Four more reviews looked at ulcerative colitis and similarly declared that we do not have the data to show that they work. The same was true for the treatment of liver disease.

Undaunted, researchers looked into whether probiotics might be beneficial in a host of disorders, even when the connection to gut health and the microbiome was tenuous. Reviews show that there is insufficient evidence to recommend their use to treat or prevent eczema, preterm labour, gestational diabetes, bacterial vaginosis, allergic diseases or urinary tract infections.

Reviews looking at the treatment or prevention of vulvovaginal candidiasis in women, pneumonia in patients hooked up to respirators, and colds in otherwise healthy people show some positive results. But the authors note that the studies are almost all of low quality, small in size, and often funded by companies with significant conflicts of interest.

Individual studies are similarly disappointing for probiotics. One examining obesity found limited effects. Another showed they do not prevent cavities in teeth. They do not help prevent infant colic, either.

None of this has prevented probiotics from becoming more popular. In 2012, almost four million Americans used them. In 2014, the global market for probiotics was more than $32 billion.

It is not clear why. Even in specific diarrhoea-focused areas, the case for them is not as strong as many think. As with nutrition research, much of this has to do with study design and how proof of efficacy does not translate into real-world applications.

“Sometimes small studies suggest strains work, but when a larger, more well-done study is performed, they no longer seem to,” Cohen said.

When research is done on probiotics, it usually involves a specific organism, defined by genus, species and even strain. When used in studies, they are pure and carefully dosed. But when we buy probiotics off the shelf, especially when they are in food products, we often have no idea what we are getting.