Probiotics 'ease upset stomachs'

“Probiotics cut stomach upsets by one day,” The Daily Telegraph has reported. The news is based on the findings of a comprehensive review by the Cochrane Collaboration, which looked at 63 studies featuring a total of 8,000 people.

This large, well-conducted review pooled the results of studies from across the globe to investigate the use of specific probiotic bacterial cultures in the treatment of acute diarrhoea (lasting less than 14 days) believed to be due to an infectious cause. Overall, it found that probiotics reduced the duration of diarrhoea by about 25 hours compared with no treatment.

However, the studies included highly variable populations, used different definitions of diarrhoea and recovery, and used a range of probiotic types, strains and doses. As such, the review authors highlight the need for further high-quality studies, particularly ones that determine which population groups probiotics may be useful for. As such, they say that no recommendations can be given for using probiotics during infectious diarrhoea in adults or children.

Rehydration remains the most important consideration in diarrhoea, and medical attention should be sought if diarrhoea persists or there are any concerns over the severity of the person’s illness.

Where did the story come from?

This Cochrane review was carried out by researchers from Swansea University and the University of the Philippines. Financial support was provided by Swansea School of Medicine, the Cochrane Infectious Disease Group and the Liverpool School of Tropical Medicine.

The Daily Telegraph’s coverage of this research includes some errors, including the fact that it related this study to Clostridium difficile infection.

What kind of research was this?

This was a systematic review and meta-analysis in which researchers conducted a comprehensive review of medical literature to identify all relevant studies examining the effectiveness of probiotics in the treatment of infectious diarrhoea. This is the best study method for examining the effectiveness of a particular treatment, though the individual studies included in systematic reviews are likely to have assessed different study populations, feature variable methods, and examine different outcomes. This needs to be taken into account when performing reviews of this kind.

What did the research involve?

The researchers carried out a search across medical databases. They also contacted individual researchers and pharmaceutical companies to identify all relevant randomised controlled trials that had compared a specific (i.e. named) probiotic with either an inactive placebo or no probiotic in people with acute diarrhoea (less than two weeks duration), which was either presumed or proven to be due to an infectious cause (a ‘bug’).

The researchers included studies in both adults and children, and were only interested in those that had reported the effect upon the diarrhoea itself, such as change in frequency or duration. Studies examining yoghurts or other fermented products that did not name specific probiotic organisms were excluded. The researchers examined the studies for quality, extracting data and pooling results from any relevant studies that had used comparable outcome measures.

What were the basic results?

The researchers identified 63 relevant studies involving a total of 8,014 participants. Of these trials, 56 had included infants and young children. Overall, 81% of the study population was less than 18 years old.

Characteristics of the included study populations varied widely. Trials were of international origin, with 19 of them conducted in countries with high child and adult mortality rates. Some 44 of the 63 studies dealt with hospitalised participants. Criteria for defining diarrhoea (e.g. description of stool, stool frequency) varied widely across studies, as did the study outcomes when defining the end of diarrhoea illness.

As would be expected, the type and dose of probiotic used varied widely. Most studies tested live preparations of lactic acid bacteria and bifidobacteria, though studies varied in whether mixed or single organisms were used.

The overall analysis pooled 35 relevant studies, including 4,555 participants. It found that probiotics reduced the duration of diarrhoea by an average of 24.76 hours compared with the control groups (95% confidence interval 15.9 to 33.6 hours). However, there was wide variability between duration of diarrhoea among intervention groups and among control groups. In pooled analysis of 29 trials (2,853 participants), there was a reduced risk of having diarrhoea lasting four or more days when taking a probiotic (risk ratio 0.41, 95% confidence interval 0.32 to 0.53).

In a pooled analysis of 20 trials (2,751 participants) a person taking the probiotic was likely to have slightly fewer stools on the second day after starting the intervention compared with someone in the control group (average difference 0.8 stools, 95% confidence interval 0.45 to 1.14 stools).

The observed relationship was not affected when the researchers adjusted for possible confounding factors, such as the type of organism causing infection, the severity of the diarrhoea, probiotic dose, strain or number of strains, study location or study quality. The studies did not report any adverse events related to probiotic use.

How did the researchers interpret the results?

The authors of the review conclude that, when used alongside rehydration, probiotics appear to be safe and reduced stool frequency and shortened the duration of infectious diarrhoea. However, they say that more research is needed to guide the use of particular probiotic regimens in certain patient groups.


This is well-conducted research that demonstrates that probiotics reduce the duration of infectious diarrhoea by about one day.

Points to consider:

  • The 63 studies included in the review varied widely in many areas, including their study populations, probiotics used, definitions of infectious diarrhoea, methods and outcomes assessed. However, despite the differing nature of these studies (known as ‘heterogeneity’), the studies have demonstrated broadly similar and consistent effects of probiotics, reducing duration and frequency of diarrhoea.
  • None of the study authors reported adverse effects related to probiotic use. Although vomiting was common across studies, this was attributed to the illness itself. However, compliance with the intervention and reasons for non-compliance or withdrawal were not reported in all studies.
  • Another aspect of safety, the researchers say, is that the majority of included participants were previously healthy. Therefore, there is a need to study any potential adverse effects in people with longer-term health problems, particularly those with malnourishment or gastrointestinal problems.
  • Due to the mixed populations studied and the fact that the total study population was predominantly under 18, the researchers report that it was not possible to analyse the effects in adults specifically.
  • As these studies were performed in variable populations worldwide, the differences between the studies may be related to other unmeasured and unexplored environmental and host factors (e.g. different immunity and different natural ‘friendly’ gut organisms). 

As the researchers of this review appropriately discuss, "the marked clinical variability among studies complicates meta-analysis” and therefore “weakens the evidence base to inform clinical practice”. They call for more large, well-designed studies of "specific probiotic regimens in specific settings". In particular, the groups of people that may most benefit from the use of probiotics needs further consideration.

As such, no recommendations can yet be given for the use of probiotics during infectious diarrhoea in adults or children. Rehydration remains the most important consideration in diarrhoea, and medical attention should be sought if diarrhoea is persisting or there are any concerns over the severity of the person’s illness.

NHS Attribution