I had fun writing abt the high-flow vs standard flow O2 RCT
Weird how using the most unbiased endpoint--mortality--as a primary can mask lots of good things about a Rx
SOHO = technically non-sig, but there were lots of benefits from high-flow
https://www.sensible-med.com/p/when-surrogate-endpoints-best-the
After 6 years, the MSM and public health establishment continue to lecture us about covid “truth” and to rationalize panic and fear-based public health authoritarianism. Apologizing and embracing fundamental reform would be more effective in helping the country to "move on."
Four truths about covid that have become clouded over time https://wapo.st/4uEiMiV
New ACC/AHA lipid guidelines race ahead with loads more class I recommendations than 2018, mostly based on expert opinion
This is a step backwards!
"@DrJMarine is more aligned with the MAHA movement."
let's go Joe!
Guidelines should be written by Neutral Martian types WITHOUT financial dualities of interest.
THIS IS UNSEEMLY 👇🏻
Conflict of Interest with Industry was handled very differently 2018 vs. 2026.
The 2018 guideline preamble explicitly states that all writing committee members were required to be free of recent and relevant industry-related conflicts.
The 2026 guideline states that writing committee members are required to recuse themselves from voting on sections to which their specific relationships with industry apply.
So the ACC/AHA shifted from a blanket "no relevant conflicts allowed on the committee" policy (2018) to a recusal-based model (2026) where conflicted members can serve but must recuse from relevant votes.
Take a look yourself.. of the 33 total members, 12 disclose relevant industry relationships. The column showing how many committee members have relevant ties to each company makes the concentration of relationships around PCSK9 inhibitors, Lp(a)-lowering agents, genetic risk profiling, and the new pipeline drugs pretty hard to miss.
The guideline states that members are required to recuse from voting on sections to which their relationships apply, but the actual recusal record — who recused from which votes —I can't find publicly.
So we know who has relevant conflicts and with which companies, but we don't know whether those individuals actually recused from the specific recommendation votes where their conflicts were most direct — for example, whether Blankstein recused from CAC scoring recommendations, whether Natarajan recused from the genetic testing/PRS sections, or whether members with Amgen/Novartis/Esperion ties recused from the PCSK9 inhibitor, bempedoic acid, or inclisiran recommendations.
But even if assume everyone recused themselves from voting, did they also recuse themselves from the discussion before the vote happened?
It's an important question because the most troubling, recommendations happen to relate to Coronary calcium testing, genetic testing, routine Lp(a) testing, and expanding pcsk9inhibitors to ever broader populations... which just happens to be where the concentration of COI lives.
Happy to hear from anyone with more knowledge of how the sausage was made. DMs open!
2018 vs 2026 Lipid guidelines compared.
The visually striking thing is the number of recommendations made in 2026 vs. 2018 : more than 4x as many recommendations, and observational-evidence recommendations replaced RCTs as the foundation.
The COI story is interesting🧵
Many guidelines have been based on little evidence
Now we have a lipid guideline, one of the best studied areas in medicine, having half of its class I (strongest recommendations) being based essentially on opinion
Half of the Class 1 recommendations in the lipid guidelines are basically opinion based.
Why give them a Class 1 rec ?
I don’t know Eric Topol personally and have no insight into what he is like in person.
What I do know is his public persona. He built a very large social media following in areas around the science of nutrition and health that I’ve been immersed in over the years.
In those domains, much of the content struck me as highly superficial and almost entirely reliant on treating Cell, Nature, and Science publications as proxies for breakthrough.
The underlying studies were not interrogated in a serious way. Instead, they were amplified, uncritically, to a broader audience. In that sense, he functioned less as scientific discussion and more as an extension of the journal and institutional press cycle.
I did not see meaningful skepticism, nuance, or alternative interpretation, even when the underlying studies were far more uncertain than presented.
He also blocked me on X, so I don’t get the sense there is much interest in engaging in substantive discussion.
My earlier comment in reference to the public advertisement highlighting his involvement with a Boston biotech VC firm is not about him or that firm individually. It reflects a broader systemic issue. The same dynamics—hype, lack of criticism, and reinforcement of groupthink—are deeply embedded in that ecosystem.
Tried telling you
That researcher has bad track record and this approach of smoothie large EHR data is confounded bullshit, not hypothesis generating
Hopeful for the addiction trials but unlikely to be anywhere close to add effective as observational studies
https://x.com/i/status/1996624646754746843
This led to some interesting Grok fact-finding and queries. I asked if Jha promoted COVID Vax requirements. Grok said no just targeted groups. After grok listed those groups, I asked it to estimate the % of the population Jha wanted vax’ed or they would be excluded: “Yes, Ashish K. Jha advocated for vaccine requirements (or vaccine-or-test options) in targeted high-priority settings that collectively affected a very large share of Americans — roughly 45% directly via work and school mandates he supported, potentially rising to 50-65% when including air travel.” I learned 44–49% of ppl in the US fly annually.
We tried in 2020-2022 and were shouted down by The Science folks
Leave it to Topol to misinterpret this outcome. Even the intervention group was below target UOP range for stone prevention.
What this does show is that it’s difficult to hit the UOP target of 2.5L/day. It’s very hard to hit that output with work and social/life considerations.
That's not what the trial shows.
The difference in volume between the control and intervention was around 1/10 of a liter...
The intervention group was 40% below the guidelines...
What this study showed was not that hydration doesn't work, but that compliance is challenging.
Pittsburgh Public Schools will be closed for three days, and compel "asynchronous" remote learning, because of the NFL Draft.
Since the city will have a lot of visitors this will "ensure students can continue learning safely and effectively."
Unreal.
https://www.publicsource.org/pittsburgh-public-schools-pps-remote-learning-nfl-draft/
prior auth sucks for all but its a tragedy of the commons
12 years of training does not mean you know how to appraise and apply evidence
I know, because I was that person
a lot of what doctors order, including at elite academic medical centers, is not needed
I see it everyday
Short this heavily based on knowing only their thesis (which is idealistic but really hard to do) and who their advisor is, with a horrible track record akin to flipping a coin
for humanity's flourishing I hope I'm wrong
I was too in medicine until you realize most of the academy do not have the skills to appraise the merits of new ideas, so rely on consensus or authority
They are selected for important but orthogonal skills like clinical practice, teaching or writing papers & grants
and it's inferior
to the "why do we need higher evidentiary standards for FDA approval", this is the reason
rather than asking the company to prove whether something is better, we approve drugs and devices and then sometimes* learn if it does or does not after many people receive it
enormous wealth transfer at the expense of these patients and society at large
*narrator: many things are never properly evaluated even after FDA approval
difference in restricted mean survival time, −0.36 years; 95% confidence interval, −0.70 to −0.01
That's not just Not non inferior. It's inferior.
> 600,000 left atrial appendage devices have been placed
NOT NONINFERIOR 👇🏻
Trial is large, nonindustry funded and done in experienced centers in Germany
Endpoint had both efficacy and safety components and still did not make non-inferiority
I tried to tell you all
Totally agree. MD without residency training is sort of useless.
No special skills. Just a rather generic body of knowledge without application experience.
The VESALIUS-CV trial is a really interesting and important trial that is very informative with regards to lipid lowering therapy and benefit.
It builds on the many foundational randomized controlled trials that have come before demonstrating the benefit of LDL lowering. 🧵
Excellent discussion and an excellent decision by @MartyMakary at FDA
Our former masters made objectively terrible decisions that were obvious in real time, and they regulated the way a duck floating down a river controls the flow of the river, but other than that they were excellent.
Wow! Not just expansion of fellowships and years off before or during med school to do research, now we are seeing something new called a pre-residency.
Not a positive development IMO!
95 CI of ARR of HFNO vs NRB goes from -4.2 to 4.1
No-one thought effects outside that range were possible b4.
2nd paragraph justifies this: "a huge trial would be reqiured"
Correct - that's a reason to *choose a different outcome*, not to run it anyway and learn very little.
1: Kennedy and @CDCgov have the right to change the recommendations, what law says they do not? History is irrelevant
2: The part of the ruling that says he doesn’t get to appoint new members of the ACIP is even more outrageous
3: Where’s the standing here? How are these plaintiffs injured? They just don’t like it. Too bad, you don’t get to sue because you don’t like a policy change
4: This ruling will not survive appeal, it is pure judicial activism
Interesting judicial ruling on ACIP. TL;DR-
1. The President does not have authority over HHS.
2. HHS does not have authority over its agencies.
3. The agencies do not have authority over their advisory committees.
4. "Experts" in the administrative state have ultimate sovereignty over us all.
5. One robed lawyer in Massachusetts gets to decide who the "experts" are.
Everyone clear?
Very important point made in today's @Sensible__Med post. The post summarizes new guidelines recommend reacting to incidental findings of coronary calcification. The recommendation is Class 1 despite evidence lacking such a classification.
I am sensitive to this being that I had a CT a couple of weeks ago that showed incidental coronary calcifications. Admittedly, when I was told about that, I had my Mandrola Moment-What would John think?
I assumed he would ignore it, which is what I did.
That this recommednation was made as a Class 1 is absurd-the evidence supporting the recommendation is just not there.
https://www.sensible-med.com/p/the-problem-of-incidental-coronary
Not personal but I’m not sure of the wisdom of having guideline authors write editorials on their own guideline 👇🏻👇🏻
Here's another goofy thing about lipid guidelines:
authors give use of PREVENT score a class I rec. No RCTs
It's more accurate BUT it outputs lower 10-yr risk than PCE, making fewer statin-eligible people
That is not allowed, so they simply change Rx thresholds down to 3% risk (using a net-benefit calculation that gives statins a 35% RRR)
If risk is that low perhaps there should be fewer people taking statins
The interventional cardiologist is almost never the person who initiates the CAC cascade.
This is under-appreciated: interventional docs are frequently the most conservative group involved
Problem often starts in primary care -> gen cardiology-> by the time the pt gets to cath lab, the train has cannot be stopped
“Even though relative risk is lower at low scores, most cardiovascular events numerically occur in the low-score population because there are a lot more people who have low calcium scores.”
Then there’s the DANCAVAS trials.
RCTs of CAC vs. standard : NO benefit.
An internal medicine colleague of mine once told me that their chief resident refused to have more than 8 patients on their service because any more would be unsafe.
Another got written up for asking a med student what their thought process was for coming up with a diagnosis.
Another was reprimanded for criticizing an intern in front of the team instead of privately.
Class I is a head scratcher. Especially given that zero calcium doesn’t mean zero events, especially if other clinical variables predict high risk.
From a pop health perspective, numerically most events are in zero/low calcium score cohort.
These guidelines are simply grade inflation
You should probably wait three weeks before you read this to see what gets retracted