They included systematic reviews with meta-analyses of randomized controlled trials evaluating oral probiotics for preventing antibiotic-associated diarrhea. Methodological quality was assessed with the AMSTAR 2 tool, and the degree to which the reviews reused the same primary trials was quantified using the corrected covered area (CCA) and the GROOVE tool. The 21 included reviews contained 478 trial entries, but these collapsed to just 170 unique randomized controlled trials, a sign of substantial reuse of the same underlying studies across reviews.

Most of the reviews were rated as low or critically low quality under AMSTAR 2. Overall overlap was moderate, with a corrected covered area of 9.06 percent, although 71 of the 210 possible pairs of reviews showed very high overlap, defined as 15 percent or greater. This matters because when many reviews reanalyze the same trials, pooling their conclusions can create an impression of more evidence than actually exists.

Despite these limitations, the pooled results pointed consistently in one direction. Probiotics were associated with a lower risk of antibiotic-associated diarrhea, with a relative risk of 0.54 (95 percent confidence interval 0.48 to 0.61) and an odds ratio of 0.37 (95 percent confidence interval 0.31 to 0.46). In practical terms, the relative risk estimate suggests roughly a 46 percent lower risk of developing diarrhea during antibiotic treatment among probiotic users compared with controls, although such observational synthesis cannot prove causation for any specific product.

The findings held up under several robustness checks. Leave-one-review-out analyses, in which each review was removed in turn to test whether any single study drove the result, produced similar estimates. The association also remained when the analyses were restricted by methodological quality.

However, when the researchers examined subsets of reviews with low overlap, the effect estimates were weaker and less precise, a pattern consistent with the possibility that duplicated trials inflated the apparent strength of the evidence. Age appeared to matter. Pooled estimates suggested a benefit in both children and adults, but no clear benefit was observed in older adults.

The review did not establish efficacy for any specific probiotic strain, dose, or formulation, and the heterogeneity across the included reviews in these variables remains a central obstacle to translation. On safety, the analysis detected no increase in adverse events among probiotic users, a reassuring signal given that these products are consumed by vulnerable patients already under treatment for infections. The researchers caution, however, that safety reporting in the underlying trials was limited, so the absence of detected harm should not be read as proof of safety.

The authors conclude that published review-level evidence associates probiotics with a lower risk of antibiotic-associated diarrhea, but that confidence is limited by low methodological quality, substantial primary-study overlap, and heterogeneity in strains, doses, and formulations. Their prescription for the field is specific: high-quality, strain-specific randomized trials and more rigorous systematic reviews are needed before probiotics can be recommended with confidence for particular products or patient groups. Until then, the message for clinicians and patients is one of cautious optimism, supported by a large body of evidence that is broad but, by the authors' own assessment, not yet deep.

Subject of Research: Umbrella review and second-order meta-analysis of probiotics for preventing antibiotic-associated diarrhea

Article Title: Probiotics for the prevention of antibiotic-associated diarrhea: an umbrella review and second-order meta-analysis of 21 systematic reviews covering 170 unique randomized controlled trials

Article References: Xi, F., Xu, C., Li, P., Zhou, S., Liu, J., Zhou, S., Wang, Y., & Yu, J. (2026). Probiotics for the Prevention of Antibiotic-Associated Diarrhea: An Umbrella Review and Second-Order Meta-Analysis of 21 Systematic Reviews Covering 170 Unique Randomized Controlled Trials. Frontiers in Nutrition, 13, 1994645. https://doi.org/10.3389/fnut.2026.1994645

Image Credits: AI Generated

DOI: 10.3389/fnut.2026.1994645

Keywords: probiotics, antibiotic-associated diarrhea, umbrella review, meta-analysis, systematic reviews, gut microbiota, randomized controlled trials, AMSTAR 2, study overlap, evidence quality, Frontiers in Nutrition, adverse events