Week of February 6, 2023
I love learning from my patients; some weeks are a gold mine of random ephemera.
This week, I learned that the acronym A.I. also means "artificial insemination" in cattle farming. Unfortunately, this realization spawned 10 minutes of me asking further questions:
How do you "A.I." a cow? (A: see link below)
What are success rates? (A: On average, 30-50%, but my patient claimed a solid 55% pregnancy rate for his efforts.)
Do the cows mind? (A: a little)
https://www.fwi.co.uk/livestock/livestock-breeding/8-step-guide-artificially-inseminating-dairy-cow
The veterinary sidebar made me late for my next patient, who mentioned their DIY ghost-hunting with electromagnetic field (EMF) meters (purchased from Amazon). Thanks to the "other A.I.," a deeper dive into this topic has to wait.
Learn about the scientific foibles and fallacies of "ghost hunting."
https://gizmodo.com/meet-the-emf-meter-the-little-tool-that-ghost-hunters-5875212
and
https://www.amazon.com/Meter-Deluxe-BLACK-New-Improved-Design/dp/B009MADZ10/ref=sr_1_1
---
Fourteen-day U.S. national-level COVID data continues to trend favorably. There are still 450 people dying every day from COVID, however.
N.Y. Times Tracker
https://www.nytimes.com/interactive/2021/us/covid-cases.html
------
A loyal reader shared this thread on the use and interpretation of wastewater RNA monitoring. The Missouri Sewershed Surveillance Project has 100 weeks of longitudinal data offering an understanding of variant distribution and intensity.
thread
https://twitter.com/solidevidence/status/1621921056989609986?s=46&t=5iYvw1057syRRizu968RaQ
Project website
https://storymaps.arcgis.com/stories/f7f5492486114da6b5d6fdc07f81aacf
The Cochrane Library is an international non-profit known for meta-analysis publications. Last week they released a meta-analysis looking at the impact of hand washing and masking (both surgical and N95) in 78 controlled trials from 2009 to 2023, including several from the COVID pandemic. The aggregate data indicated that, at a population level, masking makes little difference in the spread of influenza-like illnesses (ILI) or COVID-19. Handwashing had a small but measurable improvement in the transmission of ILI.
https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD006207.pub6/full
For those that want the plain language Cliff-notes summary:
https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD006207.pub6/full#CD006207-abs-0002
These data are provocative and polarizing. Those who are against vaccination and masking feel vindicated. Medical professionals can find numerous problems with the underlying studies; for instance, the meta-analysis used multiple pre-COVID era trials looking at communities with low adherence and minimal enforcement of masking. And it is crucial to remember that population-level data is not the same as individual risk mitigation - masks are just one part of a broader set of needed behaviors and tactics.
These data do not drive me to alter my behavior nor advise my patients to change how they protect themselves, especially the most vulnerable (elderly and immunosuppressed). I suspect a selection of more homogenous studies (same mask types, routine and rigorous behaviors) would demonstrate more compelling outcomes of mask-wearing. Still, masks are just one piece of a more comprehensive approach to population health:
https://twitter.com/SkivvyHoncho/status/1621826661372002304/photo/1
Here are some related data to consider:
https://onlinelibrary.wiley.com/doi/abs/10.1002/rmv.2336
and
https://www.sciencedirect.com/science/article/pii/S2666535222000568
and masks are just one piece of a larger approach to population health
Twitter threads on this topic:
https://twitter.com/JenniferNuzzo/status/1621881097452666883
and
https://twitter.com/IanCopeland5/status/1621781348460466178
Medical Trends and Technology
This week, I found several interesting articles about managing blood pressure and blood pressure control. It is interesting to see the blurry lines between the recommended goals, observed physiologic changes, and the voluminous associated evidence.
The American Academy of Family Physicians (AAFP) recently released new blood pressure guidelines. In essence, low-risk patients should aim for <140/<90. For patients with diabetes or cardiovascular disease (and kidney disease!), aim for <130/<80. The Medscape article explains the evidence behind these recommendations and why pushing for lower blood pressure goals can be challenging in otherwise healthy individuals.
https://www.medscape.com/viewarticle/987184
Nevertheless, some data suggest that more intensive blood pressure control, even for non-diabetics and those without cardiovascular disease, is worth attempting in appropriate patients. For example, I found this press release about an American Stroke Association conference abstract demonstrating that more intensive blood pressure control was associated with fewer structural brain changes associated with dementia.
As always, guidelines are guidelines, and any one person's care has to be customized for their needs. But there is a ton of associative data, and lower goal blood pressures may be better when tolerated.
And there are other associations with blood pressure control. PLOS (a minimal peer review, open-access journal) published a Swedish registry study looking at patients before and while taking Beta Blockers, a typical class of blood pressure medications. In the population of 1.4 million Swedes, the authors noted a 13% lower risk of being charged with a violent crime when patients routinely filled prescriptions for beta blockers.
https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1004164
Association is not causation, but Yahoo! News employs the "just asking a question" headline to imply a stronger relationship between these variables. "Could a Heart Medication Stop Violent Crimes From Happening?"
https://news.yahoo.com/could-heart-medication-stop-violent-190000754.html
Bonus: Learn about "Betteridge's Law of Headlines" or any headline ending with a question mark can be answered, "no." Otherwise, it would have been an assertion.
https://en.wikipedia.org/wiki/Betteridge%27s_law_of_headlines
Infographics
A data-heavy timeline of the Rise of A.I.
https://ourworldindata.org/uploads/2022/12/ai-training-computation-3-1663x2048.png
from an excellent article on OurWorldinData.org
https://ourworldindata.org/brief-history-of-ai
Things I learned this week
The company attempting to resurrect a woolly mammoth has opted to take a smaller bite of the "let's mess with genetics despite Michael Crichton's warnings" fortune cookie. I hope Hannukah 2027 will feature my own de-extinct pet dodo bird.
https://arstechnica.com/science/2023/01/the-next-de-extinction-target-the-dodo/
Keep your eye on Avian Influenza - H5N1. There are concerns about the bird-to-mammal spread and more frequent mammal-to-mammal spread.
https://www.nytimes.com/2023/02/03/opinion/bird-flu-h5n1-pandemic.html
We should be learning to Live in a world with A.I.
A recent review of CNET articles found that not only were they written with A.I., but the A.I. plagiarized actual human writers' work. Another excellent example of the brave new world we are entering - regurgitated knowledge from A.I. trained on known sources (either human or A.I.) with obscured sources. Thinking critically and doing a bit of research appears more crucial than ever.
https://futurism.com/cnet-ai-plagiarism
Rushabh Doshi, a medical student at Yale, published this article on the uses of ChatGPT in healthcare. Trade-offs abound!
https://www.statnews.com/2023/02/01/promises-pitfalls-chatgpt-assisted-medicine/
A.I. art of the week
"a ghost, a cow, a dodo bird, and a wooly mammoth looking at an EMF reader, digital art"
https://labs.openai.com/s/0p0oaWJpUPdcfuQdstQK3r2a
Clean hands and sharp minds,
Adam
Comments
Post a Comment