Dan Morgan
@dr_dmorgan
Followers
4K
Following
11K
Statuses
3K
Infectious disease MD/Prof Epidemiology UMSOM. Exploring how we practice medicine & could do better. Dir. Center for Innovation in Dx/ https://t.co/iMFYwgM5j9
Baltimore, MD
Joined November 2014
How good are doctors at diagnosis? This is the most relevant paper I have written. Not perfect but addresses a huge issue I think could change medicine if acknowledged …it has changed how I think about diagnosis @JAMAInternalMed
@drjohnm 1/n
42
227
689
Great work @JonBaghdadi
Across ~750k inpatients in 928 U.S. hospitals, starting with empiric narrow-spectrum antibiotics escalated to broad-spectrum antibiotics based on cultures was NOT associated with worse outcomes compared with empirically starting broad.
0
0
0
RT @AdamRodmanMD: There is a lot of buzz about our new paper in Nature Medicine on the effects of LLMs (GPT-4) on physician management reas…
0
84
0
@emollick The doctor + AI arms often aren't optimized but basically access to a chatbot. We should prompt LLM output / use as starting point w/ some doctor training on use. Then optimize format It seems like giving doctors ultrasound machines without training--of course it doesn't work..
0
0
0
Great to see talk about Mindlines, evoking the great BMJ paper about actual clinical practice.
Are clinicians’ beliefs & emotions key drivers of their antibiotic prescribing? 🆕💫 Sub study INHALE Guidelines vs mindlines: a qualitative investigation of how clinicians’ beliefs influence the application of rapid molecular diagnostics in intensive care
1
1
5
RT @jbcarmody: Articles like this wear me out. It comes so close to discussing real problems - then parrots back the same platitudes. Do…
0
61
0
RT @JALM_ADLM: Diagnostic stewardship, which extends to clinical medicine, and laboratory stewardship serve complementary roles & collabora…
0
2
0
RT @drjohnm: The use of oseltamivir in the total absence of evidence of benefit meets the criteria of therapeutic fashion. Evidence-based…
0
13
0
Same group wrote: Same group examined hospitalized patients "in severe influenza, oseltamivir and peramivir might reduce duration of hospitalisation compared with standard care or placebo, although the certainty of evidence is low."
0
1
4
Yup
A mess of a figure Likely driven by lead & length time biases without meaningful gains There surely are some real gains here but most IMO is from screening, including incidental diagnoses from rampant testing & imaging Separate by cancer & stage, and look at mortality rates
0
0
3
RT @AdamRodmanMD: But we know how to test efficacy in medicine. Clinical trials are messy, more expensive than in silico studies, and requi…
0
4
0
Interesting to work with Adam and others on this dive into the performance of newer LLM models on clinical tasks. In summary, they’re getting technically better and we need to figure out how they could interface with actual medical care
Preprint out today that tests o1-preview's medical reasoning experiments against a baseline of 100s of clinicians. In this case the title says it all: Superhuman performance of a large language model on the reasoning tasks of a physician Link: A 🧵⬇️
1
0
4
Here’s an advertisement for sample collection from one company! “a clean catch urine sample is not always an option…it can be unpleasant…a swab of a wet, saturated brief is an alternative” (here to trigger my infectious disease friends, not exactly a clean sample)
0
0
1
@moetkacik @5149jamesli @chopracfpb I thought you mean Gilman school! Next door to my kids public school.
1
0
0