EP109: Dr. Ricardo Nuila on algorithmania and the humanities in medicine

Emily Dickinson was one of the first poets that I was introduced to when I was younger, and Dr. Ricardo Nuila refers to one of her lines, Tell the truth but tell it slant. We talk about this, about nuance and context, and how that relates to algorithmania. It comes back to metacognition, gestalt vs algorithmic thinking. Dr. Ricardo Nuila, MD is an Associate Professor of Medicine at Baylor College of Medicine and sees patients at Ben Taub hospital in Houston, Texas as part of the Harris Health System. Ben Taub is a safety-net hospital where the focus in on a given individual’s healthcare problem. He is also a writer, and his pieces on health disparities, politics and medicine, and art and medicine have been published in VQR, The Atlantic, The New York Times Sunday Review, The New Yorker, and The New England Journal of Medicine.

Transcript

[00:00:00] Christine Ko: Welcome back to SEE HEAR FEEL. Today, I have the pleasure of being with Dr. Ricardo Nuila. Dr. Ricardo Nuila is an Associate Professor of Medicine at Baylor College of Medicine and sees patients at Ben Taub Hospital in Houston, Texas as part of the Harris Health System. Ben Taub is a safety net hospital where the focus is on a given individual's health care problem. He is also a writer and his pieces on health disparities, politics and medicine, and art and medicine have been published in VQR, The Atlantic, The New York Times Sunday Review, The New Yorker, and the New England Journal of Medicine.

[00:00:37] Welcome to Ricardo. 

[00:00:39] Ricardo Nuila: Thank you so much, Christine, for having me on the program. This is an honor to be here. 

[00:00:44] Christine Ko: Thank you. Could you first share a personal anecdote?

[00:00:47] Ricardo Nuila: Sure. Within the first couple of months of my internship at Ben Taub hospital, one of the patients I had not met yet that was on my list... I could see during a chart review that he was very sick. He had been sick for many months going in and out of the ICU with a metastatic cancer. I went in, and I I talked with him. He was Spanish speaking only, and I speak Spanish. We started to discuss his illness. I was going through the history. I was an intern, so I was nervous, learning my way; and going through his history, verifying the stuff in the chart, him telling me all of what he'd been through going in and out of the hospital, the ICU, his family, how he felt like he was a burden on his family. I said to him, it just came out of nowhere. I felt that he was suffering so much. I said, you know that it's okay that if you die, and we talked about it. He changed his code status in that conversation, and I put the order in the chart, and I informed the the attending. And literally 10 minutes later, I get a page from the nurse that he passed, he died. So I was stuck thinking to myself, wow, it's hard not to feel that conversation had something to do with his passing. Just the strength of the human mind, what it does, how it affects the body, how it affects somebody's hospitalization. Some people can will themselves to live longer. Some people can will themselves to live shorter. It's one of those mysteries that sticks with me.

[00:02:21] Christine Ko: I I think it is really hard to quantitate things like will or spirit or whatever. 

[00:02:28] Ricardo Nuila: Yeah. If we can't quantify it, then it's not something that is readily discussed or readily taught. And I understand that, but I also think it misses part of the bigger picture. I feel like we're arriving at a crossroads in medicine right now, because I feel that medicine exists between science and art. And when I say art, what I mean is really like more humanities driven sort of field that involves philosophy, that involves really giving advice to somebody who needs advice. How do you help solve that problem?

[00:03:02] For really good reasons, that pendulum has swung toward the science part of medicine over the last hundred years. The crossroads is, what do human beings really want? We're almost getting to the point where it's like all of the services can be given by machines, with the impending AI revolution, but is that still medicine? I don't know. I think that people really do want a human being right there with them to help them solve their problems and to think through their problems. And so I feel like that crossroads is really about the pendulum swinging the other way, from science. It's swung for a good reason towards science, but I think it needs to fall somewhere in the middle between the two, and that's what our profession really is.

[00:03:48] I'm at Baylor College of Medicine. I direct the Humanities Expression Arts Lab, which tries to integrate the arts and humanities into medical education. I think we just have a lot to learn from art. I say that as a person who is a hospitalist, who is very much like thinking about my daily routines and how busy I am. And I don't have time for inefficiencies. I think that arts and humanities help me be much more efficient in the way that I connect with people, the way I communicate, in the structure that I bring to the bigger picture for a patient when I talk with them. That's what we're trying to do, help that be a part of medical education. 

[00:04:25] Christine Ko: Yes. The pace of medicine and new knowledge, I think it's just so great that the humane aspects of medicine are being left behind.

[00:04:34] Ricardo Nuila: Yeah. We're driving toward a world where it's just the numbers, just like the milligrams or whatever, that's going to be what really determines the care. But so many of us who have practiced think we need to go to the other direction. We need to actually make it so that a human being is more available at the bedside. And that is a real contributing factor toward the goals, which is to help that person heal and also construct healthier lives for themselves. 

[00:05:02] Christine Ko: Yes. What you just said also relates to being able to really talk to patients and tell them the truth or not. In an article that you wrote, and I'll put a link to it in the show notes, you quote a line from Emily Dickinson, one of her poems saying, Tell all the truth, but tell it slant. And then you wrote," Most doctors not only think it's okay to stretch the truth for their patients, many consider it part of our unwritten professional code." Can you talk about that a little bit?

[00:05:32] Ricardo Nuila: Yeah. I think "tell the truth, but tell it slant" really means that you're accurate, but you have to give the message with with context in mind, you know. We do have a responsibility to maintain accuracy, but it just requires a little bit more in depth thinking. It requires recognizing context and matching that accuracy to the context. Does that make sense?

[00:06:03] Christine Ko: I think so. I think what you're saying is that, when you say it's part of our unwritten professional code, you're not saying that doctors lie or that we do lie or that we should lie. But what you're saying is that...

[00:06:13] Ricardo Nuila: The message needs to be given in with a thought of how it will land. [Yes]. And also with the thought of how it will give further impacts. Like, the horror story is that you hear somebody say, you have stage four cancer, and you use this lingo that just makes the patient scared and not really understand the illness. Telling slant might be starting off with, what do you know about cancer? You've seen on a cat scan that somebody has a large mass that looks very much like cancer. You start talking about cancer, but you say, we don't know until we get a biopsy, we won't know for sure. That's the truth, but in our mind, we're already thinking this is very likely cancer. But in order to help the patient accept the news over time, you introduce the idea of cancer and you don't give them the full brunt of, this is stage four. Does that make sense? I think that's accuracy, but I think it's measured accuracy.

[00:07:23] Christine Ko: There's this component of, that we just don't know. Doctors, we do learn I think to prevent needless anxiety because that is not helpful for anyone. 

[00:07:35] Ricardo Nuila: It's not helpful, and we don't want to cause that. The line from the poem, Tell the truth, you do tell the truth, but tell it slant, it just gives an added layer of thinking about the truth rather than just relaying it the way a computer might. I think our job, if we do it well, is to contextualize. 

[00:07:52] Christine Ko: Yes, absolutely. 

[00:07:54] Ricardo Nuila: It's so difficult because I've had the experience where the numbers have just given false impressions. It's really hard to contextualize numbers sometimes because 6 percent might be a lot to somebody, 6 percent might be very little to somebody. I've had the situation where a type of cancer had a five year survival that was high, but the rest of the picture of a patient had so many comorbidities, that influences the way that you talk to them. You want to give them options and with all these comorbidities, you could consider no treatment or hospice care. I've had the situation where I've said that, and then an oncologist comes and says, Your survival rate is 25 percent. And the son is, wait a minute, and you're talking about hospice? So that's why I think it's like the context. Some of what contextualizing is taking responsibility for your words. Ultimately I have to be responsible for what I said. A computer doesn't have responsibility for their words. We're at the crossroads because we're getting to the point where this information will be given without responsibility. I think that's it's still a very important part of our field.

[00:09:04] Christine Ko: Yes, I agree. 

[00:09:05] Ricardo Nuila: We have so much information that we have to deal with it. We have to find ways to filter through all of the loads of data, loads of information. Yeah.

[00:09:16] Christine Ko: That's a good segue when you say loads of information because I think algorithms help us sift through loads of information. And so they are very useful that way. They can send us down the wrong path, but they're useful. But you have this term, at least I read it through your work. "Algorithmania". Can you talk about algorithmania, and how we would avoid that?

[00:09:37] Ricardo Nuila: Yeah. Algorithmania. And first of all congratulations on pronouncing it the way that I think about because a lot of people get tripped up with it. Algorithmania is when the tool of algorithms becomes the practice instead of just the tool. We have wonderful tools. Algorithms are incredible tools. But when we forget that it's a tool, to me that's algorithmania. The tool becomes the care instead of the human being reconciling all of those things, those contexts and putting it all together. That's the allure of technology in medicine, which is that you can just yield yourself to the tools rather than the practice, the cognitive and personal sort of amalgamation of all this. 

[00:10:24] That's what I feel like medicine is. It's like a very human, very cognitive process. Keep algorithms as a tool.

[00:10:33] Christine Ko: Love it. I love your explanation. To me, algorithmic thinking or algorithmania is part of slow thinking or System 2 thinking. And when you're talking about, you put it in the context, like how we should be choosing the number one thing to then go down the algorithm for? And that's gestalt, right? I think it's that combination of taking in the whole, but also then honing in on what is really the most important thing and hopefully being right about that. Really, that is something else that's really hard to quantitate, like how much of the whole versus then how quickly can you really hone in? 

[00:11:11] Ricardo Nuila: It's difficult. Technology is giving us a ton of tools, a ton of tools, but we have to learn how to apply them efficiently. If you have to repair stuff around your house, there's a lot of different screws. If you don't have a Phillips head screwdriver, you will spend a lot of inefficient time on the house when really what you need is the Phillips head. 

[00:11:33] So again this allure of technology. Are you applying the right tool? I think in order for us to know if we're applying the right tool, that's where the cognitive function comes in. That's something that I think we need to prepare doctors for, using the right tool for what I'm trying to arrive at. I think that that's where the humanities comes in, building that mind to know the right tool to apply. 

[00:11:56] Christine Ko: Yes, I agree. I love this conversation with Dr. Ricardo Nuila. He was just talking about really knowing the right tool to apply and how algorithms are a tool. I actually just finished writing a review article on cognitive bias and de-biasing; how algorithms, fast and frugal trees, narratives, and different tools can help debias us. So this conversation was fascinating to me, just a slightly different twist on how to think about these concepts. 

[00:12:31] I'm going to stop this conversation here. Dr. Nuila and I go into disaster syndrome, how that relates to burnout, how Dr. Nuila avoids burnout. We also talk more about the Harris Health System and Ben Taub Hospital, a publicly funded healthcare system, and how that's a hint of optimism that perhaps the rest of healthcare can really learn from. So please tune in for part two next week. Thanks.