Whoa...it’s rare that we get Breaking News in health services research but here’s some. New @NEJM article shows Camden Coalition intervention did not reduce hospital readmissions when studied in RCT. (1/) https://www.nejm.org/doi/full/10.1056/NEJMsa1906848?query=TOC

Yes, this is Camden Coalition of @Atul_Gawande @NewYorker hot spotters fame. Remember this 2011 article that helped set off the hot spotting craze? (2/) https://www.newyorker.com/magazine/2011/01/24/the-hot-spotters/amp

I’ve been waiting and wondering since then if a RCT would be done and whether it would show the same results as pre-post reports. It seems that the answer for now is “no.” (3/)

Now, this comes as little surprise probably to the other research nerds for whom “regression to the mean” is common parlance. But highlights the critical importance—especially when talking about frequent users—of having adequate comparison groups in studies. (4/)

Now I think there will be lots of good questions about intervention design and also about what outcomes we do and should care about for these sorts of interventions. So this thread is To Be Continued. In the meantime, discuss. (5/5 for now).