Week of May 6, 2024
Rain, a bug-filled B&B, and our car breaking down marked a weekend beach trip to celebrate my wife's birthday and our anniversary. It's easy to stew in short-term frustrations. A AAA-funded 2.5-hour drive from Delaware to Annapolis, discussing life with Robbie, the rollback tow truck driver, offered time for meditative introspection, a broader perspective, and a deeper understanding of the towing industry.
---
Wastewater concentrations are rising, and hospitalization rates are falling. Hospitalization data are a (lagging) indicator—they usually rise about two weeks after wastewater mRNA concentrations rise.
The Pandemic Mitigation Collaborative (PMC) website uses wastewater levels to forecast 4-week predictions of COVID rates.
https://pmc19.com/data/
based upon https://biobot.io/data/
The N.Y. Times COVID Tracker reflects only CDC-gathered hospital data.
https://www.nytimes.com/interactive/2023/us/covid-cases.html
------
COVID articles
U.S. hospitals are no longer required to report COVID-19 hospital admissions, hospital capacity, or hospital occupancy data to the U.S. Department of Health and Human Services. You don't know what you don't measure.
https://www.healthcarefinancenews.com/news/cdc-hospitals-no-longer-required-report-covid-19-data
During my Friday clinic last week, after I casually mentioned a patient should stay up to date with vaccines, he emphatically mentioned that he "knew the data" and that the COVID vaccine "killed men." Epidemiologists, using Oregon public health reporting data sets, found that despite the presence of myocarditis in YOUNG men (age 16-30), there is no indication that young men are disproportionately dying after receiving the COVID vaccine. Oddly, my patient was not eager to read the linked MMWR article (The CDC's Morbidity and Mortality Weekly Report).
https://www.cdc.gov/mmwr/volumes/73/wr/mm7314a5.htm#contribAff
Medical Trends and Technology
I found a fantastic review of recently published data demonstrating how cells in the mammalian (and probably other living creatures) brain stem help control the activity of the immune system. The study is elegant, and the work implies there may be new ways to treat autoimmune conditions.
https://www.nature.com/articles/s41586-024-07469-y
excellent Twitter review:
https://x.com/vipintukur/status/1786247932825534760
Articles with titles like "Exagamglogene Autotemcel for Severe Sickle Cell Disease" make it easy to overlook data on the early efficacy of gene editing in humans (with CRISPR). Big words mask the importance of this technology. To be clear, implanting an edited genome still requires the equivalent of an autologous bone marrow transplant. However, it is not hard to see where this technology is going.
https://x.com/samuelbhume/status/1787062257479749931
https://www.nejm.org/doi/10.1056/NEJMoa2309676
Infographics
Fascinating motion infographic on the complexity of hand-eye coordination in hitting a baseball:
https://twitter.com/i/status/1787024178924728467
from
https://x.com/historyinmemes/status/1787024178924728467
Things I learned this week
I learned how pervasive the movie industry was in the early 20th century. In the 1920s, 75% of Americans went weekly to see 800+ films produced yearly.
https://x.com/culturaltutor/status/1786772256900763997
I did not know you could train honeybees to detect chemical scents, like drug or explosive-sniffing dogs. This effort has been going on for more than 15 years. The YouTube video reminds me of the movies Minority Reports and Clockwork Orange if it starred bees instead of humans.
https://www.technologyreview.com/2006/12/07/227361/using-bees-to-detect-bombs/
https://youtu.be/lG-ZqO734tM?si=itHC6I_a_tJlbM2g
https://www.bbc.com/news/business-56344609
Living with A.I.
It is worth reading Eric Topol's substack on clinical A.I. this week. He highlights the first randomized, prospective A.I. study demonstrating that AI-generated alerts for ECG changes in hospitalized patients save lives. The A.I. algorithm identified patients at high risk for death and notified the patient's hospital physician via text message. The physician would (frequently) transfer the patient to the ICU. Though the algorithm used ECG data, cardiac issues were often not the patient's primary illness. Essentially, the physician had to accept the algorithm's identification of the patient as being "high risk," intensively monitoring the patient, and through these actions, they could help the patient.
"[In the] trial, there was a statistically significant 17% reduction of all-cause mortality. The identified high-risk ECG group was where almost all the benefits occurred: a 31% reduction in deaths. That's remarkable and as good or better than our most effective medical treatments (e.g. statins for secondary prevention)."
https://erictopol.substack.com/p/when-medical-ai-is-lifesaving
The study
https://www.nature.com/articles/s41591-024-02961-4
Generative A.I. in pharma is early and already generating some compelling results.
https://www.cnbc.com/2024/05/05/within-a-few-years-generative-ai-will-design-new-drugs-on-its-own.html
A.I. art of the week
"A photo of a rollback tow truck. On the back, a psychiatrist is wearing a safety yellow vest, sitting in a chair, and a patient is lying on a couch. The psychiatrist is taking notes with a pad and a pen. The truck is moving down the road."
https://drive.google.com/file/d/19fPFiZs6MbU3AwLvzXsGk4fqKDy6HeWT/view?usp=sharing
Clean hands and sharp minds,
Adam
Comments
Post a Comment