What does the evidence really say about probiotics?

The evidence on probiotics is mixed: a few uses have support, many popular claims do not, and results depend on the strain. Probiotics are live microorganisms intended to support or improve the health of a host, and the World Health Organization's wording adds that they must be given in adequate amounts to confer a health benefit. The global retail market for them was worth US$41 billion in 2015.

The idea has an odd origin story. Around 1907, the Nobel laureate Élie Metchnikoff postulated that yoghurt-eating Bulgarian peasants lived longer, and proposed that fermented milk would "seed" the intestine with harmless lactic-acid bacteria. But in 1920, Rettger and Cheplin reported that his "Bulgarian Bacillus" could not live in the human intestine.

Here is what the evidence looks like today. A 2015 Cochrane review concluded that some probiotics had a protective effect against antibiotic-associated diarrhoea in children. For infectious diarrhoea, they were of no benefit to people ill for more than two days. For depression and anxiety, a 2019 meta-analysis found low-quality evidence of a small improvement. For preventing eczema, no good evidence shows they work.

Two reasons make it hard to generalise. There is no scientific basis for extrapolating an effect from a tested strain to an untested strain. And as of 2019, the European Food Safety Authority had rejected all manufacturers' health-claim petitions for probiotic products, citing insufficient evidence of cause and effect.

One more caution comes from the other side: some people, such as those with weakened immune systems and premature infants, may be at higher risk of adverse events.