New study from UK: closing windows increases infection risk


Wow.  Just got pointed to this news story: BBC News – Closing hospital windows ‘increases infection risk’.  It reports on work from Cath Noakes from the University of Leeds.  The work is published in a non open-access journal “Building and Environment” which I do not have access to right now so I cannot really say much more about the work.  Not sure whether they reference prior work on this topic from Jessica Green (e.g., this ISME paper).

Anyway – the project seems very interesting and I hope to track down the paper at some point.  For related topics see:

UPDATE 4/27/13:

UGGH.  Double UGGH.  Turns out, the news story is pretty misleading.  Someone sent me a copy of the paper and, well, there is nothing in the paper that actually measured microbes, pathogens, or infection.  What they did was related to surrogate estimates of infection risk based on certain models of air flow.  Not quite what I imagined when I read “dramatically reduce the risk of infection”.  Once again it just goes to show – it is always important to look at the paper behind a news story on science.  And this shows yet another reason why open access / paper access is critically important.  I am a bit embarrassed about promoting this work as I did when it is MUCH more limited in scope than what I and BBC led people to believe.


18 thoughts on “New study from UK: closing windows increases infection risk

  1. The work was reasonably well-done, but limited in its scope, as your update points out. This is also not the first time that British researchers have overstated or overgeneralized the findings from the work on infection risk and hospital ventilation. Previous examples include Escombe et al, 2007, PLoS Medicine. (If you read it, please also read my comments available on the same page).

    There is a lesson for scientists as communicators here — if you want press coverage, be aware that the press will take what is perceived by the reporter/producer/editor as the most “newsworthy” or interesting aspect without regard to the scientific evidence available to support it.

    In this case, there was one sentence in the published paper’s Abstract that was not carefully and appropriately constrained, that fed the media (in this case, BBC) feeding frenzy. It said: “Closing the windows to represent winter conditions dramatically increases infection risk, with relative exposure to the tracer increased fourfold compared to the scenarios with the windows open.”

    I have been in a dialogue with the research group leader who concedes that this sentence overstated the finding. To their credit, the authors did not include any such statement in the Conclusions section of the paper, but that didn’t stop the BBC team from dramatizing the implications of the research – that open windows reduce infection risk compared with closed windows.

    The paper is Gilkersen et al, 2013, Building and Environment. “Measurement of Ventilation and Airborne Infection Risk in Large Naturally Ventilated Hospital Wards.” (

    In fact, this over-generalization may well be true according to what is known from theory, CFD (computational fluid dynamics) modeling, controlled chamber studies, and application of the famous Wells-Riley equation. But limitations of the study itself really undermine confidence that this study “proved” that open windows reduce infection risk. In most cases studied, open windows resulted in more air exchange and, therefore, more dilution of the surrogate used to study the spread of infection from a single point release. More dilution without outdoor air means lower concentration of a pollutant (including an infectious agent) from an indoor source when the pollutant is absent or present at lower concentrations in outdoor air.

    Further, as the group leader pointed out in a personal message, on days when the weather conditions were not favorable to natural ventilation (little-to-no-wind, and very small temperature difference between indoors and outdoors), the open windows did not have much of an effect — not surprising, but one of the factors (or facts) that undermine the generalizations.

    I have written an overview of Natural Ventilation in healthcare facilities (read, “open windows”) in a paper presented at the ASHRAE IAQ 2010 conference That paper is available at

Leave a Reply

Jonathan Eisen

I am an evolutionary biologist and a Professor at U. C. Davis. My lab is in the UC Davis Genome Center and I hold appointments in the Department of Medical Microbiology and Immunology in the School of Medicine and the Department of Evolution and Ecology in the College of Biological Sciences. My research focuses on the origin of novelty (how new processes and functions originate). To study this I focus on sequencing and analyzing genomes of organisms, especially microbes and using phylogenomic analysis (see my lab site here which has more information on lab activities).  In addition to research, I am heavily involved in the Open Access publishing and Open Science movements.