Using data on ~3M households, households in which a member had bday 2 wks prior were ~30% more likely to have a covid diagnosis
ja.ma/3gK7eH12/ Idea: We wanted to study the effect of small, social gatherings on disease spread. This is hard to study b/c you need data on who is gathering, whether a participant gets covid 19, and a way to solve bias in who tends to gather vs not (eg could differ in masking, travel, etc)
Dec 11, 2020 • 10 tweets • 3 min read
Do surgeons have worse outcomes when they operate on their birthday?
It was a fun question but broader idea was to identify situations where surgeons may be distracted and to study the effect of those distractions on their patients' outcomes.
Why distracted? Texts, phone calls, pages, evening plans, etc.
Feb 19, 2020 • 11 tweets • 4 min read
Our new study in @NEJM on a cognitive error that doctors make called left digit bias - reason why goods cost $4.99 instead of $5 is why heart attack patients aged 79y 50wks are more likely to get CABG than patients aged 80y 2wks
Summary: in a range of behaviors doctors & non-doctors act similarly, including use of low value care & med adherence. Implications below
nber.org/papers/w260382/ Many public health policies & policy efforts to improve value in health care om sharing information & knowledge w/ patients.
Many interventions have lackluster results. Information alone may be hard to change behavior.
Doctors are highly informed & a natural group to study
Feb 26, 2019 • 11 tweets • 4 min read
Who's doing your surgery?
Our new @JAMA_current study on the controversial practice of overlapping surgery & what it means for patients (led by @StanfordMed Eric Sun), using data on ~66,000 surgeries performed at 8 medical centers. Thread on findings
ja.ma/2EhwXlC2/x Overlapping surgery is when the start of an operation being conducted by a surgeon occurs before the end of previous surgery the surgeon was involved in. Concurrent surgery, which is a subset of overlapping surgery, is when critical portions of a surgery overlap.
Jan 30, 2019 • 6 tweets • 2 min read
Our piece in @HarvardBiz today on the 'Art of Evidence-Based Medicine,' where @ChrisWorsham & I argue that EBM, while impt (b/c knowing something is clearly better than knowing nothing), is something we should be cautious to apply in rote fashion. THREAD.
hbr.org/2019/01/the-ar…2/x There is a sentiment that docs who depart from EBM - perhaps they prescribe a drug that's inferior in RCTs - either are clueless as to the evidence or worse, they think their anecdotal evidence outweighs systematic investigations. But, this isn't what RCT evidence tells us.
Nov 29, 2018 • 4 tweets • 2 min read
Take 2 kids. One born in Aug & the other in Sept. In states with a Sept 1 cutoff for school entry, Aug-born kids are 35% more likely to be diagnosed AND medically treated for ADHD. Our study w/ @timothyjlayton@ml_barnett@tannerhicks42 in @NEJM
Thread
2/
About 1 in 4 kids who are born in Aug wouldn’t have been diagnosed w/ & treated for ADHD had they been born in Sept
Why?
In states w/ Sept 1 cutoff, Aug-born kids are youngest in class, Sept-born oldest.
Inattentive behavior in Aug kids interpreted as ADHD vs immaturity
Jun 20, 2018 • 6 tweets • 2 min read
It’s Time To Rethink The Anatomy Of Physician Behavior
Our @Health_Affairs blog piece where @LisaRotenstein & I argue that the fee for service payment system, while not ideal, IS NOT the main culprit driving waste in U.S. health care. Thread for why...
healthaffairs.org/do/10.1377/hbl…
Dartmouth studies show large variation in FFS Medicare across patients that are arguably similar (at least more similar than not). The fact that doctors who are all paid same way (FFS) have such different utilization suggests it’s something about doctors not the payment mechanism
Feb 28, 2018 • 5 tweets • 2 min read
Reduction in Firearm Injuries during NRA Annual Conventions 1/N
Our new @NEJM study
- 20% reduction in gun injuries during convention dates vs identical days in surrounding 3 weeks
- Largest reductions in men, who comprise majority of convention attendees nejm.org/doi/full/10.10…2/N - Largest reductions in states w/ high gun ownership
- Largest reductions when convention is held in a given individual's state in a given year (hypothesis: conventions are easier to attend)
- Data: large privately insured population, ~75 million visits over 9 years
