Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

Physician in the Loop

Physician in the Loop artwork

Daily Briefing for November 3, 2026

Nov 3, 20265 minAudio
ShareXLinkedInFacebookWhatsAppEmail

Child's measles death is New York's first in more than 30 years The voices on this episode are generated, and so is the script, from the day's post at Physician in the Loop. Every fact in the post links to its source. Read the post and its sources Get the free email briefing A child in Lewis County, N.Y., has died in association with the state's measles outbreak, New York's first known measles-associated death in more than 30 years, the state health department said Nov. 2. The department did not release the child's age or vaccination status. New York had counted 178 measles cases this year as of Oct. 24, and the governor declared a state disaster emergency about four weeks before the death was announced to widen who may give the vaccine. In a Japanese validation study in NEJM AI, an AI model listening through a digital stethoscope identified reduced ejection fraction more accurately than cardiologists in a 100-patient comparison, and a German trial in the same journal found AI lay summaries shortened specialist visits by about 8 minutes. In cardiology, the PROGRESS trial found that early valve replacement for moderate aortic stenosis reduced death or heart failure events at two years. In this episode A child in Lewis County dies in New York's measles outbreak, the state's first measles-associated death in more than 30 years AI listening through a digital stethoscope flags ejection fractions of 40% or less, beating cardiologists 82% to 62% in a 100-patient test AI plain-language summaries sent before visits cut neurosurgery consultations from 33 to 25 minutes in a 300-patient randomized trial Early valve replacement for moderate aortic stenosis cuts death or heart failure events at 2 years, 744-patient PROGRESS trial finds Bicarbonate for metabolic acidosis in the ICU shows no clear survival gain but likely fewer dialysis starts, 1,544-patient meta-analysis finds

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show.

Voice 1: This is Physician in the Loop, the daily briefing on AI in medicine. It's Tuesday, November 3rd, 2026.

Voice 1: A child in Lewis County, New York, has died in connection with the state's measles outbreak. State health officials say it's New York's first known measles-associated death in more than 30 years.

Voice 2: Do we know anything about the child, age, vaccination status?

Voice 1: No. The health department didn't release either one. What we do know is the state had counted 178 measles cases this year as of October 24th.

Voice 2: And the governor had already acted on this before the death was even announced?

Voice 1: Right, about four weeks earlier Governor Hochul declared a state disaster emergency specifically to widen who's allowed to give the measles vaccine.

Voice 2: What does the state's guidance actually tell clinicians to do now?

Voice 1: An early MMR dose for infants 6 months and older who travel to or live in a community with transmission, then the usual two doses later. Twenty-one days of quarantine for exposed people who aren't immune. Isolation until four days after the rash starts. And a call ahead before a sick patient shows up at a clinic or ER.

Voice 2: That last one matters. A measles patient walking into a packed waiting room is its own kind of outbreak risk.

Voice 1: Which is exactly why the state is spelling it out now instead of waiting for the next case.

Voice 2: A very different kind of listening test came out of Japan this month.

Voice 1: An AI model listening to a digital stethoscope picked up reduced ejection fraction better than cardiologists did, in a 100-patient comparison. It was published in NEJM AI.

Voice 2: Better by how much?

Voice 1: 82% accuracy for the AI against 62% for a panel of seven cardiologists reading the same cases. The model also picked up reduced ejection fraction with an area under the curve above 0.9 in a larger validation group of 964 Japanese patients referred for echocardiography.

Voice 2: Is anyone suggesting this replaces the echo?

Voice 1: No, the authors call it a step before echocardiography, not a replacement, and they say it needs more validation before wider screening use. It's also worth saying a Japanese company whose employees co-authored the study helped fund it.

Voice 2: A stethoscope that can flag heart failure before the echo is ordered is still a useful front door, funding caveat and all.

Voice 1: A second NEJM AI study looked at something much simpler: what patients are told before they even sit down with the specialist.

Voice 2: A German trial of 300 patients at a neurosurgery clinic sent people an AI-generated plain-language summary of their own findings before their visit. Consultations ran shorter, about 25 minutes instead of 33, roughly a 21% cut.

Voice 1: Shorter, but was the visit actually better, or just quicker?

Voice 2: Patients rated the interaction higher, and the gain was driven by more shared decision-making, with no difference in how empathetic the physician seemed. One odd note: physicians themselves rated their own support of patient autonomy slightly lower in that group.

Voice 1: So the patients felt more involved, even as the doctors felt a bit less sure they'd delivered that. The authors want confirmation on outcomes and workload before anyone adopts this routinely.

Voice 2: Cardiology got real trial results too, not just AI validation studies.

Voice 1: The PROGRESS trial randomized 744 patients with moderate aortic stenosis to early valve replacement or watchful surveillance. At two years, death or a heart failure event happened in about 17% of the replacement group versus about 24% on surveillance.

Voice 2: That's a meaningful gap. Was it driven by fewer deaths, or fewer heart failure episodes?

Voice 1: Mostly heart failure events, about 11% versus 19%. All-cause death was actually similar between the groups, around 8% either way.

Voice 2: So the benefit is in keeping people out of heart failure, not necessarily keeping them alive longer, at least at two years.

Voice 1: That's what the numbers show. Current guidelines just call for echo surveillance every one or two years for moderate stenosis, so this is a real challenge to waiting. Edwards Lifesciences funded the trial.

Voice 2: A quieter ICU study closes out the research today.

Voice 1: A meta-analysis of four trials, about 1,500 critically ill patients with metabolic acidosis, found sodium bicarbonate didn't clearly improve short-term survival.

Voice 2: So bicarbonate just doesn't do much here?

Voice 1: Not quite. It was linked to a lower chance of needing to start dialysis, with a 97% posterior probability of benefit on that specific outcome. Everything else measured was small or imprecise, and no clear safety signal turned up either way.

Voice 2: So it may keep some patients off dialysis without changing whether they survive. That's the kind of nuance that gets lost if someone just reads the headline.

Voice 1: The authors say the next step is figuring out which patients are actually likely to benefit. Before we go, something to watch. The AMA's Interim Meeting opens November 6th in Orlando.

Voice 2: Anything specific on the agenda worth flagging?

Voice 1: It's the first House of Delegates meeting since the AMA's framework on the physician's role alongside AI, so expect resolutions on AI liability and payment.

Voice 2: Liability and payment are exactly the two questions nobody's settled yet. Worth watching who actually brings resolutions on either one.

Voice 1: Sources and corrections are at physicianintheloop.org, where you can sign up for the free email briefing. This briefing was read by AI voices.

Open the transcript file

Hosts and Guests

Nobody is credited on this episode yet.

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. Special Topic: What changes on Jan. 1

    Nov 20, 20267 minAudio

    Laws in at least eight states that touch AI in health care take effect on Jan. 1. Most bind insurers and chatbot companies. A few, in California and Colorado, reach the practice itself. Read the special topic and its...

    Credit the people on this episode

  2. Friday Letter: On a short leash

    Nov 20, 20267 minAudio

    Software has begun to act for doctors, not just advise them. It should be kept on a short leash, because the doctor watching it rarely notices when it slips. Read the letter and its sources:...

    Credit the people on this episode

  3. Daily Briefing for November 20, 2026

    Nov 20, 20266 minAudio

    Hospitals race to patch a Java flaw found with AI in an hour The voices on this episode are generated, and so is the script, from the day's post at Physician in the Loop. Every fact in the post links to its source. Read...

    Credit the people on this episode

  4. Special Topic: Not the doctor's to sign

    Nov 19, 20267 minAudio

    Free AI tools for doctors come with a HIPAA contract one doctor can accept alone. A hospital's records need the hospital's signature, and the companies should make the safe path the default. Read the special topic and...

    Credit the people on this episode

  5. Daily Briefing for November 19, 2026

    Nov 19, 20267 minAudio

    AI hacking agents seize some networks in under 3 hours, FBI warns The voices on this episode are generated, and so is the script, from the day's post at Physician in the Loop. Every fact in the post links to its source....

    Credit the people on this episode