home Journal club Conclusions not totally convincing but hand microbiomes worth considering in hospital acquired infection studies

Conclusions not totally convincing but hand microbiomes worth considering in hospital acquired infection studies

So – just was reading this paper: Healthcare Workers’ Hand Microbiome May Mediate Carriage of Hospital Pathogens.  Basically, they showed that the microbiome on health care worker’s hands was correlated with estimates of pathogen carriage. And, it seems to be of potential interest to the microBEnet community.  I confess, I am skeptical of the validity of some of the claims in the paper such as this from the Conclusions:

In summary, our primary finding was that after correcting for individual risk of exposure to the selected potential pathogens, differences in carriage of these potential pathogens might be attributed to the structure of the hand microbial community.

The issue I have with this and some other claims is that I think the evidence is really about a correlative vs. causative connection.  Thus the term “attributed to” seems inappropriate. They do caveat this in other parts of the paper:

However, a limitation of this study is the directionality of the association between carriage of potential pathogens and HCW hand microbiota. That is, it may be that the hand microbiota is itself a result of the carriage of certain pathogens or that both the microbiota and the carriage of specific potential pathogens are associated with yet another factor.

Anyway – whether or not one believes their conclusions about possible causal impacts of hand microbiomes on pathogen transmission in a hospital, I think the data presented here is enough to convince me that studies of transmission in hospitals might want to include analyses of health care workers skin microbiomes in the future.


3 thoughts on “Conclusions not totally convincing but hand microbiomes worth considering in hospital acquired infection studies

  1. They still state potential pathogens in the manuscript which is good as neither the 16s nor the qPCR data can be used to assess pathogen load. Staph presence, even MRSA, is no indication of the potential to infect. We find MRSA cassettes everywhere we find human skin, just in very low abundances.

  2. We can say with confidence that the hands of healthcare workers have been implicated in the cause of infections; but the evidence for this goes back to Semmelweis and company, with much higher rates of infection and clear evidence of handwashing as an effective intervention.

    I’ve followed this line of research in the prior publications from the same group.

    My main issues here are the continued use of “pathogen” (ie ‘3.5 … Pathogen Carriage Detection’ sometimes called ‘potential pathogen’) without further evidence of pathogenicity (i.e. patient cases of infection at the same time in the hospital from which the organism, identified by WGS, is recovered and then disappears during successful treatment), the selection of three samples to establish temporal stability on the hands, which is by definition too few, and the notion of community structure defined so loosely.

    These studies are phenomenally hard to do in a conclusive way. I don’t think that, as an academic community, we are succeeding overall, though some of the WGS HAI work coming out of England is particularly good. I am strongly sympathetic to this handwashing/disease transmission scenario because hypothesizing about colonization resistance led to my first science award and prize money – in early high school. Perhaps we need to find a way to come together more collaboratively as we pursue this line of research.

Leave a Reply

%d bloggers like this: