Incredible paper and team tracing the history of the 6' rule and droplet/aerosol issues. Amazing how thin the basis for some of these things are, yet get codified into our guidance.

(I have one non-covid example that is my least-cited, but maybe most important, paper)

THREAD https://twitter.com/linseymarr/status/1387781398291959813

I led the Legionnaires' disease outbreak investigations for my prior firm. And everywhere I went hospitals would cite this '30% cutoff' for determining risk.

If <30% of water samples tested positive, ok. If >30%, higher risk.

Never made sense to me, for obvious reasons (um, what if you took 100 samples and 99 were negative, but there was one positive sample with high levels of L. pneumophila in a bone marrow transplant unit?)

So I started tracing the history of this 30% metric. Turns out it came from a paper in 1983 (Best et al.), where they did an oversimplified analysis and merely *suggested* a 30% cutoff made sense.

From there it got strange. A few agencies, like the Allegheny County Health Dept, started using this metric, citing Best et al.

Then others started to use it, too, citing Allegheny...

When we dug deeper it just became a web of self-citations, with all of the studies actually only relying on Best et al.

(here's us trying to untangle the web!)

(we cleaned up the handwritten notes for the paper...but notice how much ties into Best et al. and Allegheny)

We did a literature review of all outbreaks we could find to formally test this 30% metric. Guess what? it doesn't work.

And we published a paper showing this, which I'm pretty sure has 0 citations, but I still think it's one of my most important papers. Because it did have one key, non-academic citation...

Our paper got attention after there was an outbreak in a VA hospital in ....ALLEGHENY COUNTY (!)

The VA Inspector General investigated. And, for the first time I'm aware, flagged that this 30% rule may not be all it was cracked up to be, citing our work (!)

There were hearings on this...where they discussed the false sense of security from this "30%" metric

And then...the VA revised its guidance and DROPPED THE 30% RULE!

Same with Allegheny County!

Side note: this is why I also think things like H-Index are nonsense. There are ways to have impact that go behind citations of your work in peer-reviewed journals.

-end-