No Priors Ep. 130 | With OpenEvidence Founder Daniel Nadler

Danielle, thanks for doing this.

Happy to be here.

So, uh, give us a sense of this incredibly viral sensation that has been OpenEvidence, uh, in terms of what type of, um, coverage it has of American doctors today.

As much as we would like to think that it's going especially well for us, I would sort of say as a qualifying point that, um, in all of the sub-industries of AI, you're, you're seeing an acceleration and compression, right? So the, the adoption cycles, even outside of OpenEvidence, before we get to OpenEvidence, in other fields of knowledge where encoding and so on are hyper compressed, right? It used to take, you know, half a decade or a decade for something to become standard, and now it seems to happen in two years or, or a year. So the same thing's happened with OpenEvidence. In about 18 months, it's become the operating system for clinical knowledge in the United States. Uh, it is used something like 20 times more than the next most used platform of any kind in our specific segment, which is high stakes clinical decision support for doctors. So high stakes clinical decision support for doctors is a specific category of medicine. It's distinct from, say, paperwork, or it's distinct from scribing. Um, those things are, you know, part of the workflow of being a doctor, uh, but the stakes and the consequences, uh, are different. Um, if you get it wrong, you can go back and do it again. Uh, that's not the case with a patient. Uh, you have to get it right, and you have one shot to get it right. And so clinical decision making, uh, of which clinical decision support, uh, is in service of, is unquestionably the highest stakes area of medicine. We're probably the only company working at the tip of that s- spear. Most people have self-selected themselves out of the problem of high stakes clinical decision making, uh, certainly through an AI lens, um, because they view it as ambitious.

And could you explain an order of evidence? Because I think fundamentally, it's about taking information and then translating that into specific either recommendations or diagnoses for a patient. Can you tell us more about how that works?

Yes. One way to sort of simplify it down is at its foundation, it's a search problem, but it's a very semantic search problem. Uh, so most search traditionally works with keywords, right? So like, you know, flights to Barcelona or hotels in Barcelona. Most of the, you know, most of the keywords there can be captured in, like, a couple of words, and certainly in a sentence. And that's sort of traditional Google search. Even if you were to think about clinical decision support as a search problem, simply describing your search query, if you want to think about it that way, usually takes many sentences. So an example I like to give is you have a 44-year-old female patient, she has moderate to severe psoriasis. That's the red stuff on your skin. Um, y- you know, you're a dermatologist. That's so far, so simple. You would just prescribe one of the many creams you see commercials for on television. Except, uh, she has, um, MS. Uh, so now it gets interesting because you want to treat her psoriasis, um, but you don't want to make the MS worse. And you are not a neurologist, you're a dermatologist, so neurology is not your specialty. Um, but you don't want to go refer her to a neurologist because you want to treat her psoriasis and, and if you just keep referring people in circles, medicine never happens. From the ether, you might have heard as a dermatologist that the new classes of psoriasis treatments, um, which are biologics, they're IL-17 inhibitors and IL-23 inhibitors, might have some interactivity, uh, with the neurological dimension of a patient's condition. That's about all you know. Um, you didn't learn this in medical school because IL-23s were FDA approved in 2019, right? Uh, so one of the great themes of OpenEvidence is that this sort of golden age of biotechnology is sort of the dark ages of physician burnout because it's just impossible to keep up with all the new drugs and all the new mechanisms of action and so on. So y- you know, it was approved in 2019, you might have graduated medical school in 2005, right? (laughs) So y- you didn't cover it in medical school and that's it. That's kind of, that's what you know. So your question then is, you know, for a 44-year-old female patient with moderate to severe psoriasis, is an IL-17 inhibitor or an IL-23 inhibitor more appropriate and more safely tolerated with respect to not aggravating the MS? Now that's, that's not a academic question. Um, that's a very consequential question. IL-17 inhibitors will actually make the MS worse. Uh, IL-23 inhibitors are safe and well-tolerated in case of MS. Th- that's an example of where medicine can go wrong because even five or 10 years ago, um, either you're referring that person to a neurologist, in which case you're just getting r- referrals in circles and medicine is not happening, or unfortunately, what would more likely happen is they would just 50/50, and that MS might be aggravated. And, y- you know, it's well-known and it's been often repeated that medical error is the third leading cause of death in the United States after heart disease and cancer. But even that kind of, that statistic kind of understates it because that's just looking at death, right? In the case of my, in my example, um, this patient is not gonna die as a result of taking an IL-17 inhibitor. She's going to have a, a relapse of MS. And so it's not just that medical error historically was a leading cause of death, it's that, uh, as many people died from medical error, probably a factor of 10 to 100 as many people had a, a comorbidity or condition that became aggravated and got worse and so on. So coming back to your question, that whole string is the search query. And so you can't just do search in a traditional way where you sort of say, you know, IL-17, 'cause that's not really what the question's about. Um, nor does the physician have the time to go read book chapters on this stuff. What you need is a semantic understanding of the, of the query in the way that another human physician would semantically understand that query. And then it's actually quite-... deterministic and simple after that. Um, once you semantically understand the query, uh, you can from the world of published biomedical literature, you can find the exact snippets in a phase 3 RCT, Randomized Controlled Trial in The New England Journal of Medicine that tested each of these things and found that one aggravated MS and the other didn't, right? So once- once you have a semantic understanding of the query, uh, the rest is fairly deterministic and it's almost a search problem. Um, but all of the- all of the juice is in, you know, connecting the very complex semantic meaning of a medical scenario to the answer where the answer might be in a phase 3 RCT in The New England Journal of Medicine and a snippet in- in, not even in the- in the abstract, but in the methodology section or in the population section.