Do prebiotics, the fibres that feed your good bacteria, work just as well as probiotics?
Prebiotics alone are less effective than probiotics in the best-studied situations. For diarrhoea or immune support, probiotics are more useful, but even those effects are smaller than often assumed.
Prebiotics and probiotics are related, but they do not do the same thing. Probiotics are live bacteria; prebiotics are the fibres that feed those bacteria. For the best-studied applications, the answer is clear: prebiotics alone perform worse than probiotics.
The difference is greatest in acute diarrhoea in children. Probiotics, especially the yeast Saccharomyces boulardii, shorten the duration of diarrhoea by 17 to 40 hours. Combining them with zinc produces even better results, particularly in lower-income countries. Prebiotics alone showed no benefit over placebo or standard treatment in the same research. Synbiotics, a combination of probiotics and prebiotics, do work.
The picture is similar for type 2 diabetes. A large analysis of 68 studies showed that prebiotics do not demonstrably improve blood sugar levels1. Probiotics did, but the effect was small and not clinically relevant enough to recommend supplementation2,3.
In athletes, there is reasonable evidence that certain probiotic strains reduce upper respiratory tract infections in frequency, severity and duration. They also appear to support the gut wall, which becomes more vulnerable during intensive exercise4,5. Hardly any research has been done on prebiotics in athletes, so a comparison here simply cannot be made.
In organ transplantation and prevention of colorectal cancer, the evidence is too limited or of too low a quality for reliable conclusions. A review study in kidney and liver transplant patients found no clear difference between synbiotics and prebiotics alone, but the studies were small and of very low quality.
Based on two network meta-analyses and several randomised studies (PMID 30517196, 38134725, 31864419, 36126902, 38551082). The total number of participants is difficult to calculate precisely given the use of network meta-analyses; conservatively estimated, it involves well over 5,000 people across all studies.