Thread by @ml_barnett

Thanks for the heads up on this @mikerose42

Buckle up guys

I want to describe a paper that must be in the top 5 most unknown-but-wild RCTs in the literature

It's an RCT of continuity of care no joke

https://twitter.com/nickdawson/status/1087779694877896704

@mikerose42 "Continuity of Outpatient Medical Care in Elderly Men: A Randomized Trial"

JAMA 1984

jamanetwork.com/journals/jama/…

@mikerose42 This insane RCT randomized 776 men over 55yo treated at the White River Junction VA in Vermont

Patients randomized to "discontinuity" had a DICE ROLL to decide which doctor they would see in follow-up

Any appointments with the same provider twice in a row were also reassigned

@mikerose42 Here's the balance table, looks pretty good. This is not a small RCT.

@mikerose42 The patients in the discontinuity group predictably had fewer visits with the same provider

The effects after 30 months were dramatic and all worse in the discontinuity group:
- 9.1 vs. 5.6 hospital days
- 39% vs. 20% were emergent admissions
- Satisfaction was worse

@mikerose42 So there you go. Wild RCT that I can never imagine an IRB approving today in any way.

Seems like continuity of care is important, more than even I would have expected.

See this slide deck from Hopkins for a bit more critique of the study

ocw.jhsph.edu/courses/HSRE/P…