andrewparsonsMD Profile Banner
Andrew Parsons Profile
Andrew Parsons

@andrewparsonsMD

Followers
937
Following
2K
Media
62
Statuses
1K

Assoc Dean for Clinical Competency @medicineUVA | Director of Research, Hospital Medicine | @MaastrichtU PhD Candidate | Author #ClinicalReasoning

Charlottesville, VA
Joined June 2010
Don't wanna be here? Send us removal request.
@andrewparsonsMD
Andrew Parsons
3 months
Excited to announce the publication of our new book! We hope this is a wonderful resource for clinicians and medical educators. #clinicalreasoning #meded AVAILABLE: https://t.co/WLsSTaUqaZ
1
8
43
@CoryRohlfsen
Cory Rohlfsen
3 days
I had this bookmarked w/ a similar critique. Hard to activate management scripts independent from illness scripts. Deliberate practice is not linear, stepwise, or strictly pathophys. It’s nuanced pattern recognition amidst highly varied scripts in which management can’t wait.
@doc_BLocke
Brian Locke, MD MSCI
3 days
It’s kind of demoralizing that the rare diagnostic reasoning pub in a top tier journal is framed against 2-4 decades old paradigms (thinking fast-slow dichotomy, errors mostly cognitive biases, etc) that are mostly junk. LLMs or not, just get with the current lit.
0
2
8
@AnilMakam
Anil Makam
3 days
Great thread Clinical reasoning & diagnostic errors have moved too far from the root cause: the doctor's knowledge & ability to reason using problem representation & illness scripts Cognitive debias, Swiss cheese, & mechanistic thinking only help at the edges Not the core
@doc_BLocke
Brian Locke, MD MSCI
3 days
It’s kind of demoralizing that the rare diagnostic reasoning pub in a top tier journal is framed against 2-4 decades old paradigms (thinking fast-slow dichotomy, errors mostly cognitive biases, etc) that are mostly junk. LLMs or not, just get with the current lit.
3
5
30
@doc_BLocke
Brian Locke, MD MSCI
3 days
It’s kind of demoralizing that the rare diagnostic reasoning pub in a top tier journal is framed against 2-4 decades old paradigms (thinking fast-slow dichotomy, errors mostly cognitive biases, etc) that are mostly junk. LLMs or not, just get with the current lit.
@JAMA_current
JAMA
5 days
💬 Viewpoint: Medical education should shift from pattern recognition to critical thinking and pathophysiological reasoning to reduce diagnostic errors and adapt to #AI advancements. #MedEd https://t.co/yyhCG5IclQ
6
36
153
@CoryRohlfsen
Cory Rohlfsen
15 days
This diagnostic reasoning question has been asked many times over the past 30 years but #AI is a game changer in #MedEd Come find out how and why at 11:30am today to learn more about novel quantitative signatures of #diagnosticexcellence ; Rm 205B #DxEx
1
4
11
@AdamRodmanMD
Adam Rodman
12 days
Highly recommend attending this session! @CoryRohlfsen is on the frontier of measuring diagnostic excellence, with important implications for doctors, AI systems, and the interface between the two!
@CoryRohlfsen
Cory Rohlfsen
15 days
This diagnostic reasoning question has been asked many times over the past 30 years but #AI is a game changer in #MedEd Come find out how and why at 11:30am today to learn more about novel quantitative signatures of #diagnosticexcellence ; Rm 205B #DxEx
1
3
12
@AdamRodmanMD
Adam Rodman
12 days
@EvidenceOpen This is generally when I'm much more confident that management reasoning will remain within the purview of physicians for quite some time (as opposed to diagnosis @jbcarmody)
1
2
8
@AdamRodmanMD
Adam Rodman
12 days
For more context, these results aren't surprising to me at all. Commercial LLMs are remarkably good at diagnosis, which can be (more) easily rewarded. Management reasoning is notoriously difficult to reward, and not generally in any model pre-training.
@rohanpaul_ai
Rohan Paul
12 days
Another bad news for Medical AI. This paper shows that medical LLMs often give different answers to the same hospital question. The core finding is that these tools are unstable for judgment-heavy bedside calls. The team tested 6 models on 4 common inpatient cases where either
3
8
33
@ledje
Lou Edje, MD, MHPE, FAAFP (she/her)
15 days
That’s a wrap! So invigorated by the medical educators who came to our Medical Education Symposium to hear Vice Dean of Edcuation, @KarenHauer4 (UCSF), Vice Chair of Research Department of Emergency Medicine, @DowinHugh (NYU), Dr. @andrewparsonsMD & our in-situ presenters. We
0
1
8
@andrewparsonsMD
Andrew Parsons
15 days
Thanks for the invite to speak with @UMichMedSchool at #MedEd Symposium. What an engaging, thoughtful, and kind group of clinician educators led by @ledje
1
1
7
@CoryRohlfsen
Cory Rohlfsen
29 days
@andrewparsonsMD @CMichaelGibson Humans evolved to navigate info deserts (foragers). Our super power is recognition when it ‘doesn’t fit’ (we attend disproportionately to surprises). AI thrives in info rich contexts & has diff biases. My take on how we’re complimentary hasn’t shifted:
first10em.com
Try to define a physician’s role in diagnostic reasoning as artificial intelligence emerges: a distributed cognition framework
0
2
3
@EricTopol
Eric Topol
29 days
Will A.I. lead to de-skilling of physicians? And if so what can be done to preempt it? @tberzin and I address this in a new @TheLancet piece https://t.co/rK5Lf77YtK
42
283
827
@andrewparsonsMD
Andrew Parsons
1 month
Managing Clinical Uncertainty: Formalizing Management Reasoning in Emergency Care Delivery ⁦— fun to collaborate with such a wonderful group led by ⁦@ADHaimo⁩ ⁦@AnnalsofEM
Tweet card summary image
annemergmed.com
Emergency care faces a persistent tension between the imperative for diagnostic accuracy and the complex social, environmental, resource, and time constraints inherent to clinical practice. Classical...
0
2
3
@UNMC_HospMed
UNMC Hospital Medicine
1 month
Can diagnostic excellence be measured? What is #entropy and how does it relate to clinical uncertainty? The answers might surprise you. Check out the newly published work from @CoryRohlfsen and @andrewparsonsMD #DxEx
Tweet card summary image
link.springer.com
Journal of General Internal Medicine - Navigating uncertainty is fundamental to sound clinical decision-making. With the advent of artificial intelligence, mathematical approximations of disease...
0
2
2
@saibalamdjd
Sai Balasubramanian, M.D., J.D.
1 month
0
5
8
@andrewparsonsMD
Andrew Parsons
2 months
Outstanding editorial by ⁦@RachelKon⁩ ⁦@MemariMD⁩ ⁦@UvaDOM⁩ in ⁦@JournalGIM#PIF
Tweet card summary image
link.springer.com
Journal of General Internal Medicine -
1
4
4
@CoryRohlfsen
Cory Rohlfsen
2 months
But entropy is not ready to serve as a standard for diagnostic excellence until it becomes linked to values-based, decision thresholds & relevant outcomes. Any ideas on how it can become fit for purpose? To better define, measure, or enrich a clinical context. 2/3
1
1
2
@CoryRohlfsen
Cory Rohlfsen
2 months
Pub🚨 What does #entropy have to do w/ #clinicalreasoning? We answered this Q in the form of a narrative review of 20 studies👇 As a measure & construct of complexity, we can start measuring, modeling, & managing states of pt ambiguity. #microstates https://t.co/OHHKEcvZtf
Tweet card summary image
link.springer.com
Journal of General Internal Medicine - Navigating uncertainty is fundamental to sound clinical decision-making. With the advent of artificial intelligence, mathematical approximations of disease...
2
10
14
@AlissaHigMD
Alissa Higinbotham, MD
2 months
Day 1 of @MedicineUVA’s Clinical Reasoning Faculty Scholars Program! Coaching struggling learners is something I am very passionate about, and I am thrilled to have the opportunity to develop these skills in this year’s cohort. #MedEd #NeuroX 📚
0
1
12
@andrewparsonsMD
Andrew Parsons
2 months
No Immunity From Having to Choose: An Exercise in Management Reasoning from Grace Yi, MSPH ⁦@dgsomucla⁩ ⁦@JournalGIM#CR #MedEd
Tweet card summary image
link.springer.com
Journal of General Internal Medicine -
0
1
0