The role of the microbiome in human health and disease: an introduction for clinicians

As the author of this article, I appreciate comments from readers regarding the content. I received the following from a reader. This reader felt that I appeared to be much more in favor of fecal microbiota transplant than probiotics. I did not mean to imply this, but in case others have the same opinion I figured that the reader's opinions should be made available to others. I have contacted the reader and he is in agreement to have his comments posted anonymously by me. I would appreciate any other comments regarding this.

Vincent Young

***email message, edited to maintain anonymity of the author***

Thank you for writing the recent review in BMJ. I found it very useful and will have all my students in lab read it. However, I found the part about probiotics very upsetting. I will likely write a letter to the editor but I feel like it is more fair to just write you first. Now like all of us I have my bias and I have made a living doing probiotic RCTs. But I feel like you very much shortchanged the 100's of high quality studies that have been conducted. My opinion aside, many groups such as Cochrane, Rand, other evidence-based reports have done meta-analysis/systematic reviews and found level A evidence for things like colic, AAD, traveler's diarrhea, UC, etc. One negative study, the one you reference of AAD/CDI, does not prove that probiotics do not work. In fact, I find negative and null studies reassuring that quality research is being done and questioning the field. I study the 2nd most commonly used probiotic, BB12, and many of my studies have found no difference in the primary outcome but I would never say that means BB12 has no efficacy. That just means in my studies, of my population, with my inclusion/exclusion, etc were not positive studies. There isn't a drug on the market that hasn't had a null finding, if that was the bar, our diabetics would have no medications to use.

While shortchanging probiotics you are very positive with FMT. I think I find that more upsetting than I do the probiotic section. FMT for CDI has three total RCTs, one of which found over 60% improvement when patients got their own contaminated stool back. Most of the FMT studies are case controls, which besides a few very extenuating circumstances have no place in an evidence-based review or clinical medicine. The data comparison of probiotic efficacy and well done studies compared to the 3 very small FMT studies is a disservice to practicing physicians.

Finally, you stress mechanism. I am a researcher who studies and respects mechanism. But clinical trials are what changes patients lives and what practicing physicians care about. I teach an entire semester undergraduate class that we review the difference between disease oriented and patient oriented data, the former based on mechanism has a long history of leading us down the wrong path. That said many mechanisms of probiotics are well known and have been reproduced. But I would hope my patients don't stop their lipitor, tylenol or many other drugs until I can prove to them the mechanism.

Again, I know I sound like I didn't thoroughly enjoy the article and that isn't the case. I think it was a great well written article. But I am afraid your negative take on probiotics and positive endorsement of a blind non mechanism FMT approach will be a disservice in the long run. I appreciate your taking the time to read this very long email