Labels of US Probiotic Products Lacking, Researchers Find

When it comes to buying probiotics, many product labels do not give consumers enough information to make an informed decision, according to a research team led by Georgetown University Medical Center.

In their study, published in the Journal of General Internal Medicine, investigators found that of 93 different bottles of probiotics purchased from four large national probiotic retailers in the Washington, DC area, only 33 (35%) could be linked to medical efficacy as supported by clinical evidence. The researchers point out that the other 65% of products might be supported by clinical studies, but they were not able track down the evidence given insufficient labeling.

The researchers considered a product supported by evidence if the product label disclosed the strain of bacteria or yeast designation(s), the strain was said to be present at a beneficial dose, and at least one controlled human study supported use of the probiotic was listed on PubMed, a web-based search engine for biomedicine and health studies.

“The higher number of strains, higher dose or greater cost were not associated with evidence,”

says the study’s senior investigator, Dan Merenstein, MD, professor of family medicine and director of research programs for the Department of Family Medicine at Georgetown University School of Medicine.

The good news?

“It may be counterintuitive to consumers, but we found that products with fewer strains and lower cost were more likely to be supported by evidence we could trace,” he says.¬†The caveat is that “consumers will have to do legwork themselves to figure out the best product to buy,”

Merenstein says. None of the 33 bottles linked to evidence that the probiotics inside were clinically beneficial had any mention of those studies on the label.

“I think it’s a lot to expect consumers to check the evidence, but trustworthy guides can help them find products backed by evidence,” he says. “Companies could greatly help this process by labeling products with strain designations, dose at expiration, and with proper indication of usage.”

In 2014, a group of experts, two of which are authors on this paper, affirmed the definition of probiotics to be

“live microorganisms that, when administered in adequate amounts, confer a health benefit on the host.”

The robust use of probiotics has led to a number of studies that have found several strains work as intended, but additional studies have shown other strains don’t work better than a placebo. Manufacturers are required to demonstrate that commercial probiotics are safe for the intended consumer and are labeled in a truthful and not misleading fashion.

Of the 93 products examined, 67 were judged to be “unique” based on distinct strain composition. Further, any products that did not disclose strains on labels were considered unique since the precise composition could not be determined. But evidence could not be traced to many of the high-dose, multiple-species, and more expensive products, Merenstein says. Some such products listed some strains that are known to be beneficial, but since the dose of the strain was not disclosed, researchers could not determine if such strains were present at an efficacious dose.


In addition to Merenstin, the study co-authors include John Guzzi, MD, a student at Georgetown University School of Medicine at the time of the study, and Mary Ellen Sanders, PhD, is from Dairy & Food Culture Technologies.

Merenstein reports being an expert for Bayer, Pharmavite, Reckitt Benckiser and Debevoise & Plimpton. Sanders consults with numerous companies engaged in probiotic business, but does not have any financial stake in any company. She serves on scientific advisory boards for Danone, Yakult, DannonWave, Clorox and Winclove. Guzzi reports no disclosures.

Merenstein and Sanders are currently conducting two clinical trials examining the role for probiotics and potential mechanism in preventing diarrhea in children on antibiotics.

Be the first to comment

Leave a Reply

Your email address will not be published.


*