Curt Morrison, MD, FACC

Board certified cardiologist. Fellow of the American College of Cardiology. CFA charterholder. Booth School of Business MBA. Erstwhile equity analyst.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@Jesse_Leg Cilantro?

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@Notwokenow Making an exit isn't going to help Clark. Covering the game and advocating for her would.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@M_L_Dighton @Benaskren It's a streak. Can take place in one season. 5 for 5 has nothing to do with this. Is Joe DiMaggio's 56 game hitting streak a career record? No. Pete Rose's 4,256 hits is a career record.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

As a child in the US in the 1960s this was normal, no yellow cap required. We went everywhere without adults, riding bikes all over town, climbing trees in the woods, searching through construction sites for discarded to soda bottles to claim the two-cent deposit, sneaking into the municipal pool at ages 7-9 by ourselves, and yes, walking to school or riding buses by ourselves. This was the norm forever prior to recent decades.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@mdburnell Has Sophie commented on her meme? I wonder what she thinks of it.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@_WhoDeyRay You should have been there in the 70s. Those teams were magnificent.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@GrulkeLamonte Tesla Model Y with FSD.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@MichaelAlbertMD One caveat regarding heart disease: I don’t believe that we are getting coverage for heart failure with preserved ejection fraction, even though the SUMMIT and STEP-HFpEF showed benefit.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@RandPaul Fauci is powerful? What is the source of that power?

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@jimiuorio It is already means tested in that higher earners pay more in SS taxes, but receive proportionally less because of the bend points.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

You don’t need to ask lawyers. I am an MD and of course patients are allowed to record their encounters and an occasional patient has done this during their visit with me. Some patients also use their phones to include relatives or others so that those people can also hear the encounter. Beyond that any patient can request a copy of the visit note and all of their tests as well.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@TotalProSports 😅

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@RuxandraTeslo edit: but many intuitively appealing ideas with plausible rationales have proven to be useless or harmful when tested.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

I haven't seen physicians write that the Midjourney Medical device is necessarily a bad thing. Rather they (I) have simply written that it is not necessarily (or even probably) a very useful thing, based on the limitations of ultrasound and experience with other attempts at improving outcomes via early detection of disease. For example:

1. Thyroid ultrasound screening (South Korea). Opportunistic thyroid ultrasound screening produced a 15-fold increase in thyroid cancer incidence with no change in mortality. An estimated 90% of detected cancers in Korean women were overdiagnosed, exposing patients to unnecessary thyroidectomy with 11% risk of hypoparathyroidism and 2% risk of vocal cord paralysis.

2. General health checks in adults. Comprehensive periodic health examinations for asymptomatic adults have not been shown to reduce morbidity or mortality in a Cochrane review, while increasing overdiagnosis and unnecessary treatment.

The idea that more images must be better than fewer is intuitively appealing, but many intuitively appealing ideas with plausible rationales have proven to useless or harmless when tested.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@stationerycar Every medical trial in every medical journal contains a discussion of P-values. What is the source of this opinion?

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@DrCasteelEM I've been reading the comments, and I haven't seen a single non-physician acknowledge anything at all.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

@jeffreytucker Not so sure about that, partner.

[Curt Morrison, MD, FACC @CMorrisonMDFACC]

I've been there only once, in 1994, and the people were fantastic.

Just one example: I went to Surfer's Paradise in September, not yet in season, and there were no surfboards for rent on the beach. I asked a lifeguard where I might rent one and he gave me directions to a house about one-half mile away.

I found the place with 8-10 boards leaning against the side of the house, so I knocked on the screen door. A barefoot guy in cutoff shorts answered, and I asked if I could rent a board. He said I could just grab one and return it when I was done, and he wouldn't take any money.

So I carried a board to the beach and surfed for a couple of hours and then carried it back and leaned it against the wall again.

All the cabbies and doormen and everyone else was friendly and full of shark and crocodile stories. People were laid back and fun. I think the culture may have changed quite a bit in the last 30 years, as it has in the UK.

[🐊ZK For Tre🐊 @ZKForTre]

I want to try to explain why tech bros and people in the medical field have been having a... robust disagreement about how to interpret the new Midjourney ultrasound technology (a phenomenon that may only be taking place in my cursed feed, idk). This will be very long.

For those who didn't see, Midjourney published an ultrasound-based whole-body imaging technique that allegedly takes only 60 seconds immersed in water. They have a whole page around the spa they are going to build around it, but at its core is based on this technology https://t.co/QCjXWZRgzU.

Before I get into the debate about this, let me highlight what is obvious good about this and what the obvious limitations are:

Good: no radiation, fast imaging is a win. Period. Stacking AI for tissue-level resolution is an exciting proposition.

Not Good: this isn't going to happen for critically ill patients who need to get scanned repeated for so many reasons. Also, this isn't scanning anything meaningful in the cranial vault.

Okay, so what is the debate-- the debate centers around the concept that if this technology is cheap and fast, people can just get scanned in a preventative capacity all the time. Monitor for everything-- early cancer detection, monitor fat accumulation, ASCVD???-- what's not to love.

What's not to love is a phenomena that medicine has known for a long time--- more data doesn't always improve patient outcomes, and it sometimes harms them. See attached for a graph of when South Korea decided to screen thyroid cancer as an example. More detection, more biopsies, more doctors visits, more spending... no changes in thyroid cancer mortality.

Okay but more data is always better and then its on doctors and the medical system to learn how to use it correctly, right?

I'm going to illustrate all the ways in which this fall apart in the real world.

First, let's be very generous in our assumptions:
1. This technology is implemented like MRIs or CTs, but it only costs $20 per scan. Dirt cheap.
2. The scan quality gets really good. We end up trusting them to like 90% of our current trust in MRIs.

If we then used this as an opportunity to bootstrap a new universal healthcare standard where patients get scanned using this method all the time, here is how that would play out:

1. people would get one of these at the equivalent of a yearly physical or when they are in the hospital

2. The number of incidental findings skyrockets

3. Doctors now hold legal and moral responsibility to follow up on incidental findings that cannot be ruled out as pathological

4. in the outpatient setting-- more biopsies, more doctors visits, more concerns about pathologies that, statistically speaking, are far far more likely that not to be irrelevant to health trajectory

5. in the inpatient settings-- neuro and ortho and derm and nephrology and IR and path and GI (you get the point) all have their consult burden explode. If you thought your resident friend was overburdened now, just wait.

6. You think the hospital is hiring more residents to make up the difference? seriously?

7. Sure, the scan costs $20, but the apparatus doesn't. Plus, what kind of premium are hospitals of doctors offices or spas or imaging centers charging?

8. If it's covered by insurance, at what rate? What about the follow ups?

9. How do people that already struggle to afford visits pay for all the extra follow-up appointments? Is it not covered by Medicaid or lower-tier insurance plans? So this is just rich person concierge medicine or inpatient medicine now?

10. If we just skyrocketed the amount of medium-cost imaging we do, that will unequivocally get absorbed into what are already sky high insurance premiums to cover it-- I have not seen one of the high horse tech boosters try to do a cost-basis analysis of the cost of extra imaging versus prevented cancers (and associated costs).

11. If we are just doing these as cash only and insurance ignores it as a modality, how does it interface with services that would do the follow ups? It's already not simple to address what happens when people are concerned over non-pathological CGM readings, so what happens when every disc abnormality, osteophyte, or microadenoma is now being passed on to service? What happens if the service does not view the image finding as an issue? Who makes that call? And, most importantly, who holds the liability?

12. If you think you already don't get enough time with your doctor, imagine if they add 10-15 "screening" cases per week. Who is absorbing this patient volume?

13. remember, most of these findings will be incidental, so when their is inevitably some small percentage of patients who are harmed in the process of a biopsy or who do go on develop a cancer of unclear relation to the screening scan, who is taking liability for that? Certainly not the imagers.

14. There is published evidence of the negative consequences of patients being told about incidental findings that have a microscopic chance of being pathological. Are we going to get everyone full knowledge of Bayesian interpretation in the next half decade? The tech bros don't seem to have a good grasp of it as it currently stands.

15. What are the cutoffs for when to follow up on microadenomas and when not to? The Korea/thyroid example should lay out in plain detail why just following up on every imaging finding is a poor exercise. All of the tech bros saying doctors are so dumb-- you want them to make this call, then?

16. Lastly, remember that most people in this country aren't nominally wealthy people that can visit the doctor pretty frequently with minimal cost concern. What happens with losses to follow up? What happens when parents can make 3 doctors appointments in a month? What happens when someone doesn't want to pay 4 or 5 co-pays more than they normally would?

All of this discussion strikes me as the utopian dream of tech bros that have plenty of money, plenty of time, and are deep into the health-maxxing space but won't even spare a moment for the concepts that the medical fields has been studying and considering for decades.

Such is their right, but dont expect me not to criticize them for their ignorant arrogance.

[Joseph Younis, MD @YounisJoseph]

I think the tech community is conflating two different things when they talk about “false positives.”
What they seem to mean is artifact, which is a known technical noise or measurement anomaly that better sensors, more data, or improved reconstruction methods can reduce

What doctors mean by false positives are real findings that appear concerning on imaging but turn out to be clinically insignificant post workup. These “incidentalomas” trigger a series of additional scans, specialist visits, biopsies, anxiety, and cost with no net benefit to the patient…

The irony of the critiques is that they flaunt profound understanding of probabilities while directly contradicting Bayesian principles. This low pre-test probability settings (scanning asymptomatic people without clinical context or symptoms) causes even a highly specific test to generate a large number of findings that require investigation but ultimately prove irrelevant

More covariates and clinical context improve the positive predictive value, the findings that actually end up being true problems that need intervention. Raw population screening does the opposite

With that said, I’m actually not concerned with false positives because I believe in informed consent. I’d defer this risk to the patient after I do my best to educate

My real concern with something like the Midjourney Scanner is: it’s not fundamentally superior technology being deployed where it’s needed. It’s consumer tech framing whole-body imaging as wellness and prevention likely to collect data for other purposes and lead to false reassurance