I thought I would answer some of the questions that I am frequently asked.
What do you think of splitting the MMR shot into 3 shots?
It is not going to happen. It would take years, perhaps a decade, several billion dollars, and unrelenting political pressure on vaccine makers to accomplish.
If a future president made it a major personal goal, and used the entire power of the presidency to lean on every company to get it done (and ensured his/her successor did too), then, like putting a man on the moon, it might happen by the end of the decade, but the reality is no one party can control the presidency for that long, and the next president of the opposing party would halt all efforts.
Do you believe approved drugs and vaccines and devices are safe and effective, and have a net positive benefit-to-harm ratio?
Some of them, sure. All of them? No, certainly not. For many of these products, we know very little about long-term safety risks and the benefit-to-harm ratio. Our current systems are ill-equipped to learn from post-market data.
Do you think Tylenol consumed during pregnancy contributes to autism?
No, I do not. I think it is, as always, low-quality observational data used to draw improper conclusions.
Would you advise a pregnant woman to limit the use of medications unless absolutely necessary?
Yes, that is generally prudent advice.
Has the political left ever used low-quality observational data to draw incorrect, or improperly stated conclusions?
Examples include: the COVID vaccines in kids, Paxlovid, annual boosters, annual flu shots, covid vaccine in pregnancy, long COVID treatments, lockdowns, school closures, how many lives would be saved by Medicare for all, that COVID shots reduce cardiovascular events, whether the shingles vaccine stops dementia, etc. etc.
Do you think SSRIs are overused?
Absolutely. The use of these products far outstrips the evidence. It is disappointing to see so many professionals defend the use of these products in ways that lack a robust evidence base.
Do you think a healthy pregnant woman in 2026 should get a COVID-19 shot?
Absolutely not. I consider this a bizarre recommendation that reflects poorly on the professional societies that make it. It not only lacks evidence of benefit to either mother or child, but also lacks a compelling rationale for even considering it. What is the risk of hospitalization to a pregnant woman in 2026? We can’t even answer that question — it is so low.
What are your thoughts on gender medicine?
Overwhelmingly low-quality evidence from which strong conclusions cannot be drawn. In the meantime, the use of hormone blockers and surgery for minors strikes me as deeply problematic. The fact that (some) American doctors embraced it reflects the massive liberal bias in the academy.
What do you think of this sentiment?
Completely accurate. Such consults will be markedly reduced. And similarly, the ID doctor won’t ask me to interpret iron and ferritin levels.
Has federal funding of science become politicized?
No — it has always been political. It has strongly favored liberal causes and goals for the last two decades, and the composition of those who make funding decisions is overwhelmingly liberal. How else did we get an explosion of disparities research that doesn’t actually help minorities? Now, the bias is shifting, slightly, to the right. I would be happy if it moves to be in line with what most American’s believe
Should doctors be able to own hospitals?
Yes
Should the USPSTF give Grail a B recommendation?
No, it should be I (inconclusive). Anything else would be incorrect.
Academics vs private practice?
For most graduating physicians, private practice is the better option. The pay, autonomy, and net societal impact are better.
What are trials/ drugs you like?
Many. GLPs for overweight diabetics. Daraxonrasib. Treatments for lymphoma. Tec-dara.
Should the FDA approve more favored vapes?
We know very little if more flavors result in more never users starting to use, or shift patterns among those who already use tobacco products. We know little if the availability of flavors contribute to fewer or more people dying of tobacco related illnesses. Yet opinions are strong. See also Gender medicine.
Would you advise a young person to go into medicine?
No, I would discourage it. I think AI will drastically disrupt the space. I trained from age 21 to 31 (Med school, residency, fellowship) — that opportunity cost is too great given AI's potential to replace the core aspects of the job. If you upload your scans, labs, and biopsies into Claude Fable and ask for recommendations, it is already at the level of the average oncologist.
What’s your version of the Turing test — when we know AI is truly thinking?
When you ask AI to compare outcomes in an uncontrolled phase 2 to historical benchmarks or a synthetic control arm, and it refuses, saying that the data will have too much residual confounding and you are better off randomizing. (Prasad’s test for AGI).
Will total pharmaceutical spending go up or down in the next decade?
In the US, it will go up 5 times.
What do you think of ivermectin and febendazole in cancer?
These are medicines that don’t work — have no credible data — and have an odd fan base.
What do you think of statins and metformin in cancer?
These are medicines that don’t work — have no credible data — and have an odd fan base.
What do you think of Moderna-Merck mRNA cancer therapeutic vaccine?
I am waiting to learn the HR for relapse-free survival. If it is < 0.5, I think people will be seriously impressed. If > 0.8, I think people may be disappointed. No matter what, I am not sure it implies this will work in other tumors because melanoma has always been one of a handful of cancers with unique immunologic properties.
What do you think of China’s dominance in biotech?
I agree it is happening. I am not sure how we can truly compete with a country where you can even enroll patients without having informed consent. Most of the proffered solutions are half measures at best.
Are you bearish about lp(A)?
Without seeing the full Horizon results, I think it is hard to know.
Would you take a GLP-1?
No, my BMI is 23, and with a little discipline it will (again) be 22. I am also unclear about the implications of the HR elevation.
Should the CDC have counted the Pennsylvania measles deaths or not?
I have no idea; we have not been provided with enough information. But I doubt anyone's vaccine decisions depend on whether the number is two, four, or zero.
How can public health regain trust?
I don't think it can or deserves trust.
What about those deaths from the COVID vaccine? And the RTF on the Moderna mRNA flu shot?
I will direct you to this video, which I think covers those topics
Has the political right ever used low-quality observational data to draw incorrect, or improperly stated conclusions?
Given that you would recommend against someone going into medicine in the age of AI, what would be your recommendation to someone who already started medical school? Are there specific specialties that are less prone to AI disruption? Are there ones that you would completely discourage going into?