Transcript for Episode 40
Grant:
My guest today on the Beatrice Institute podcast is Dr. Dan Hall. Dan is a surgeon and an Episcopal priest here in Pittsburgh, and he wears many hats. He's an associate professor of surgery at the University of Pittsburgh School of Medicine and is on core faculty of the Center for Bioethics and Health Law here at Pitt. He also performs surgeries and conducts research at the VA health system. He's also the medical director of high-risk populations and outcomes at UPMC Wolff Center, which focuses on improving quality within the UPMC health system.
Dan received his bachelor of arts, master of divinity, and medical degrees from Yale, completed a postdoc at Duke in health and religion, where he also earned a master's degree in health sciences. Dan has wide-ranging research interests, including frailty and elective surgery, theology and medicine, as well as research ethics.
So Dan, welcome to the podcast. I'm grateful that you're joining me, but I'm especially grateful knowing that you just had surgery yesterday. So, I'm grateful that you're, you're soldiering up and doing this, even though you're recovering right now. Thanks so much for coming.
Dan:
Delighted to be here, and, when I'm gimpy, thank you for giving me an opportunity to appear, at least to myself, useful.
Grant:
Good. Good. So, to start us off, I'm hoping that you could fill out some of your biography for me. So, how did you get here to University of Pittsburgh as a surgeon, bioethicist, quality improvement guru? I'm particularly interested in how you chose this bi-vocation of priest and surgeon. So, anything you want to share in terms of your history. How did Dan Hall get from Hastings, Nebraska, to Pittsburgh?
Dan:
Yeah, well, I told that story at a conference a few years ago in greater completeness and maybe give a link that people can follow if they want to get more details than just this. But I arrived as an undergraduate having had the good fortune of getting involved with a biotech firm, thinking that I would be doing bench-side research, genetic engineering of some variety, was really interested in hard sciences. And college did what it was supposed to do. It got me interested not just in the molecules that make up the tree, but the forest as well, and took some classes in philosophy. All of that, sort of, culminated in a senior essay project that was focused on the ethics of sequencing the human genome. And at the same time, the notion of being ordained in the Episcopal Church came out of left field. And I wasn't exactly sure where all that would come together, but some advice of mentors suggested that medicine was one of the few places where you could sort of remain interested both in the forest and the larger questions of life and interacting with people at that fundamental essence of meaning and value, but also have the opportunity to delve into the molecules and the details and the empirical sciences on which that kind of knowledge is based.
And that seemed to be good advice. And I went off and did both medical and divinity school and landed here in Pittsburgh by virtue of the match process, which tries to max—it's an algorithm that maximizes happiness between training programs and trainees—and it ended up being very fortuitous in connecting me to a really strong program here in Pittsburgh that was nice enough to hire me on as faculty. And it's been a whirlwind since 1999, when I arrived. And I'm still trying to figure out what it all means. But it's kind of, it's been a very long road and a long game, but it seems to be coming together now in a way that is really exciting.
Grant:
Are you still working as a priest or is your primary vocation right now as a researcher and a surgeon?
Dan:
Well, I spend most of my time—and all of my money comes from—work as a surgeon and researcher and medical director. But my identity is priest first, who works out that vocation in the context of the hospital. And I think the way to try to frame it is in a culture in which the medicine man is the only one that continues to wear clerical garb of a big white coat and a stethoscope, like a stole around their neck, and lives in a cloistered environment of the hospital, presides over life and death, is the defacto priest of our technical, scientific society. And yet the training that most of us receive does not speak about normative goods and their discernment, even though the liturgy and the ritual of what we provide seems to promise that we might have something to offer along those lines. And I think that leads to a lot of the unusual discrepancy between the fact that people are highly critical of healthcare on one hand, yet they love their docs a whole lot. And I think it has to do with some misperceptions of expectation of what is being asked and what is being offered.
Grant:
Sure. Yeah. We'll come back to this question of what we've gained and what we lost by replacing priest with the physician. This is something that you talk about in your work a little bit as the, the physician becomes the new modern priest.
But I did want to start our conversation talking about your most recent research endeavor. You've been working with a team of researchers to use medical record data and survey data to create a measure of frailty that will hopefully help surgeons make better decisions about who to operate on and when. So, I want to start off with a little discussion about this concept of frailty. We'll get to the measurement issue next. So, just very simply, what is frailty and why is this an important concept to measure when thinking about surgery?
Dan:
So, on the one hand, it is what it, sort of, calls to mind: a little old lady on a walker, struggling to get around, thin, poor nutrition, papery skin, limited mobility. It is a concept that has been developed by our colleagues in geriatrics. And I think it has a lot of heuristic value in that it is focusing on functional performance as a measure of physiologic reserve and, thereby, is able to, sort of, integrate small deficits in multiple organ systems. Any one of which would not be worrisome, but together indicate a vulnerability to catastrophic decompensation in the setting of a stressor of some variety. And to the extent that our very deterministic way of looking at things has broken human beings up into their individual systems, it is a way of trying to pull all those systems back together and give you a sense of the organism as a physiologic whole and its sense of reserve.
So, that's, that's what frailty is. There's general agreement about that, but no one can actually agree on a definition. So, it's a bit like pornography in the 1980s. Everybody knows it when they see it, but you can't really define it, which means that there are a lot of different ways of measuring it. And depending on what it is that you're trying to measure, one may be better than the other. But what did not exist before the work that my colleagues and I did was there really wasn't a way to provide a rapid, feasible measure of frailty that could be deployed in real time, in predominantly robust populations to identify the small proportion of patients who are at increased risk by virtue of their frailty.
This is important because surgery is like running a 5k race, if not more, depending on the physiologic stress. And in the same way that, maybe, there are a few weekend warriors that would think about, you know, standing up and running a 5k without any training. But most of our sedentary older folk have not been training. And to think that one can just, sort of, land on the operating room table without adverse consequence without preparing for it in some way, means that there's a real opportunity to try to identify the folks who have a depleted physiologic reserve, which allows the opportunity for doing two things.
Number one, the more obvious, is that if they're willing to participate, you can try to build up that physiologic reserve in advance of the stress of surgery through some kind of training program. And it also gives the opportunity to step back and, sort of, say, "Really is, is what we're proposing really consistent with your goals and values? Even if you do prepare, your risks are likely to be elevated compared to the robust or normal counterparts, which comprise all of the study data and the outcomes. And it may be that it would be more prudent or more consistent with your goals and values not to swing for the fences, but to do something a little bit less aggressive that would manage the symptoms and preserve the quality of life you're enjoying now for as long as we can."
Grant:
Yeah. So, this makes me wonder. So you're, you're obviously saying that, you know, historically, surgeons would often cut open many frail patients, despite the fact that the decision to perform surgery will significantly lead to loss of life or major disability. Why is it that surgeons decide to take such risks, especially with these frail patients? Is this about the money? I would suspect the VA it's not because it's, they're salaried employees. What is it?
Dan:
So, I mean, that's, that's one way to characterize it, and I'm sure that there are and continue to be those kinds of cowboys or cowgirls out in the world operating in that way. But I think that good and wise surgeons have always done, have made an attempt of doing bedside risk assessment along these lines, whether they use the word frailty or not. We would talk about, you know, I remember in training sort of saying, "Here's this patient. Give them their descriptions: eighty-two-year-old, a laundry list of problems." And the summation is not to be touched with a ten-foot pole. I mean, the idea that that patient would be too high risk to do a surgery on. But there's a growing amount of data to suggest that this, sort of, bedside, foot-of-the-bed risk assessment is not nearly as reliable as something that's systematized and uses, uses a more, well, I guess, systematic way of assessing that risk.
That would be one reason. I think people have been trying, but they need the assistance of a systematic tool to help them do it better. It's easy to detect the person who's the eighty-year-old, who's wobbling in on a walker. And these are the folks that we have not been operating on. The systematic tool is able to identify people who have a measurable risk that is preclinical, not, available immediately to the eye. And among that crowd of patients, even if there's a general sense that they may be at elevated risk, surgeons by temperament are problem solvers. So, we can succumb to the Lake Wobegon effect where all the anesthesiologists are good looking, all surgeons are strong, and, darn it, every patient is above average. At least under my care, if we can make this happen and send them to UPMC, it's life-changing medicine. This is the way that we market, and we come surprised that these train wrecks of physiologic derangement arrive on our helipad with expectations that we're somehow going to solve the problem.
Grant:
Alright. So, let's talk a little about this tool that you created. So, moving from the concept of frailty into something actionable that you can put in front of surgeons and say, "Look, grandma, she looks okay, but if you look at the way she's scored on this tool, she's probably pretty high risk." it's something called the risk analysis index. Can you just give us a high-level, lay description of the RAI, the risk analysis index?
Dan:
So, it's a tool that started as a clinician-facing tool, but now is, essentially, one that is completed by the patient with or without the assistance of a family member in the waiting room. It asks for their age; their sex; whether they've lost weight; whether their appetite has been good; whether they have shortness of breath; whether they have a history of cancer; where they live, is it independent or some sort of assisted living environment like a nursing home; asks them whether there's been any cognitive decline; and then activities of daily living in four dimensions: toileting, eating, bathing, and getting around.
And that tool, which takes the patient, perhaps, two minutes to fill out and it takes the clinician thirty seconds to key into an online calculator, can render a score that effectively risk stratifies people. The overwhelming majority of folks, ninety percent, score in a range where the outcomes that they achieve in terms of mortality, readmission, length of stay are on par or below what the published averages are across a diversity of surgical types.
And it does that with a high degree of negative predictive value, meaning that if you score low, it's ninety-seven percent chance that you're not going to have a problem. If you score high, in that highest ten percent, or maybe that highest five percent, it's not, not a particularly sensitive or reliable predictor of mortality or frailty or whatever. But it does indicate that maybe this person deserves a second look, and it allows what we, the, the terminology we've been using is the concept of a pause. That, if you score above this threshold, you want to stop the assembly line, take that one off, take a really good look at it, all around. Maybe do additional risk assessment tools and engage in an interdisciplinary evaluation of what the benefits and burdens of this treatment, or perhaps a lesser treatment, might be. And clarify what the patient's goals and values are to make sure that the care that we're rendering is really consistent with what it is that they're hoping to achieve.
Grant:
So, you know, the way that the RAI will be most successful and impactful on patient outcomes is if it gives us additional information above what the surgeon already, sort of, knows intuitively, as you mentioned before. This is a practice that surgeons do already. So, have you tested this in the real world, such that you're able to demonstrate that, actually, it does create some added value above what the surgeon already knows intuitively.
Dan:
Yeah. So, our initial demonstration was in the Omaha VA where, on instituting a program in which this RAI score became a mandatory condition for scheduling time in the operating room. And then the chief of surgery or the chief's designee on a weekly basis would look at the upcoming couple of weeks of planned surgeries. And for those patients whose score exceeded a threshold, he would start picking up the phone and talking with the anesthesiologist or the surgeon or the critical care doc to kind of talk through the—in a very ad hoc way—the care plan and the decision-making to make sure that it all passed the sniff test. And it would also, pretty aggressively, encourage people to consult palliative care physicians before the operation to get documentation for advanced directives and also goals of care really rock solid. And what happened there, comparing in matched controls from before until after the initiation of this frailty screening program, took the mortality among frail patients from about twenty-five percent to less than eight percent. And that corresponds in a multivariable model to a threefold survival advantage, controlling for a lot of known confounders.
We also completely changed the pattern of palliative care consultation in the institution. We doubled the number of consults on surgical patients and shifted them before the operation and being called by a surgeon, rather than after the operation by some non-surgeon that was afraid that things had, sort of, gone off the rails. And preoperative palliative care consultation by a surgeon quarters the risk of mortality among that group, controlling for whether or not a patient actually has surgery or not. How that's happening, I'm not exactly sure. But one of the things that, I think, could explain that is disruption of what I've come to call "premature withdrawal of care." That surgeons, we may, I may have had a really good discussion with you, my patient, and you're aware of these risks that are elevated, but you really want to give it a shot and have, sort of, bought into a full, time-limited trial of all that intensive care medicine can provide for a period of time, but I don't document it because I'm a surgeon and my documentation is poor. And you're a World War II vet, and you're kind of independent that way, and you haven't spoken with your family. So that, after my operation, when things have gone sideways a little bit and you're in the intensive care unit and can't speak for yourself because you're on a ventilator, and your son from Phoenix arrives and says, "Dad never would have wanted to live this way." The nurses appropriately call the palliative care doc, and the palliative care doc says, "Yeah. That's interesting. We are worried about that, too, but as you'll see from my note, because my documentation is really good, we had this conversation, and your father really wanted to give this a full-court press for two weeks, three weeks. We'll reassess at that point." And I think that that affords the opportunity for appropriate rescue care in these highest risk patients.
Grant:
So, it seems like the intervention that you did was, sort of, multi-pronged, and had something to do with palliative care then also had something to do with the implementation of the index itself. Is that correct?
Dan:
Yes.
Grant:
So, what's next? How do you isolate the impact of the RAI above and beyond some of the other institutional changes that were happening at the VA at that point?
Dan:
Well ,that would be one...there are two types of people in the world. There are lumpers and there are splitters. Your question is a splitting question and I honor it. I tend to be more of a lumper. If it works, and it doesn't cost much, and it's not too terribly hard, just keep doing it.
[laughter]
Grant
Right.
Dan:
It'll improve.
I think that the way forward, I mean, we've replicated this in a quality improvement project at UPMC where, after instituting a frailty screening project, we saw that month-over-month mortality rates started to drop again, one-year, long-term mortality rates. We have less of an idea about what actually changed in their care process. And it may be more of a Hawthorne effect of just signaling to the clinicians, "We're really paying attention to this thing called frailty, which are your highest risk patients. You be the great doctors that we've hired you to be and pay particular attention to these folks." And whether they're changing their intraoperative approach, their preoperative approach, their patient-selection approach, we're not exactly sure, but it does seem to be yielding an effect.
The way to tease out what is happening would require designing a number of trials that would either continue to deploy the complex behavioral intervention in its individual parts but, perhaps, adjust in a randomized scheme which parts of that you do at different times to try to...Or just test one portion of it in a different trial each time to see what does palliative care consultation in isolation achieve among frail patients. What does an interdisciplinary review panel achieve in isolation going forward? And then, if it's an interdisciplinary review panel, one could then, sort of, think of a design that would be adaptive to try to figure out what portions of that interdisciplinary review is achieving the greatest impact.
Grant:
So, it's interesting, the difference between pure research and quality improvement research, where, in your case, the Hawthorne effect is an effect that's worth preserving. You don't necessarily need to tease that out if the whole package is the intervention. It's maybe less interesting to figure out exactly which parts work, unless, of course, they're hard to implement. But in this case, it seems as though, relatively low cost in this case.
Dan:
I think it, also, there's an analogy there to some of my previous work in religious beliefs and practice. I mean, it is, there's a substantial literature that would suggest that prayer practices are associated with a variety of improved health outcomes. Now, is that a Hawthorne effect or is that evidence of God almighty intervening in the world? I probably favor the former a little bit more than the latter. But again, if prayer can be a heuristic, and it is, in fact, defacto, the way that cultures have organized and incentivized and encouraged a set of behaviors, which may be, you know.
Herbert Benson at Harvard talks about the relaxation response and the changes in the oximetry and the like that happen that may have a salutary effect. Is it prayer or is it the relaxation response? On one level, I don't care, if your outcomes are better and if that's what you're trying to achieve. But I wouldn't, as you and I have discussed, and we may get there again, eventually, to confuse that and those sets of motivations with prayer as motivated within a religious tradition of its own integrity is to confuse apples and oranges.
Grant:
I'm realizing, as we're having this conversation, we're going to be talking about religiosity and health later, and I have a feeling I have a couple more splitting questions, but we'll see, we'll see how that goes.
[laughter]
One thing I do want to do now is, sort of, situate the RAI within a broader context of using big data predictive analytics to help surgeons and other clinicians and healthcare providers make decisions.
So, as you know and our listeners know, we're increasingly relying more and more on big data predictive analytics to help us make appropriate decisions based on these predictions that we make about what could happen. We think about innocuous things like making book recommendations. They look at your previous purchase history and get some sense of the kind of books you like and can predict what you might like in the future. We also use predictive analytics to figure out where we should place police officers in the city based on crime patterns.
But there's a little bit of a darker side of predictive analytics. We think about, maybe, Facebook and Google, who are arguably leading to polarization as they predict the clickbait that you're going to like, send it to you, and it makes you more and more angry at the people that disagree with you. At the same time, there's some concerns about how predictive analytics might make racism work. Right? If we find that African Americans are more likely to do something, we might feed them things that will make them keep doing certain things that are, maybe, are not, you know, in their best interest.
And again, these methods are increasingly being used in healthcare to predict all sorts of outcomes: hospital admissions, risk of heart attacks, readmissions. So, the question is what are some of the risks of using these, sort of, big data predictive analytics within health care?
Dan:
It's interesting. I was, just before this, on a two-hour panel with people from Amazon Web Services and Microsoft and clinical leaders in venture capital asking precisely this: what is the state, age, maturity, and risk of using AI in healthcare? And I think that the risks, there are a variety of risks.
So, one you mentioned is particularly with the machine learning algorithms. They're a black box. There are kinds where, there are types of, and the ones that I've worked with are black boxes, which at least will tell you which variables it's looking at most, which are providing the most value, which allows you to begin to have some clinical sense of what it's paying attention to.
But with any black box, garbage in is garbage out. There was a paper in Science a little while ago that was, sort of, talking about a machine learning algorithm that could predict, but could basically read x-rays for pneumonia. And it turned out that when you actually got into it—and it was very good, it had, it could do it very well—but it was using a sample of x-rays that were done in the hospital and not in the hospital. And it turned out that there was a different brand of machine, x-ray machine, in the hospital compared to out of the hospital. And that was the single greatest predictor for whether or not someone had pneumonia, because it was essentially a proxy for where they were in the hospital, rather than actually reading the visual information and the lung fields of the x-ray.
So, as we apply to the black box—and it can do amazing things—it requires prudent, wise, human guidance to kick the tires to make sure that what's going in... The uniform, the uniform comment on this panel was ninety percent of all AI applications, ninety percent of the effort is spent on data curation. That you have to clean the data to the extent that you think it's reliable and it is the type of thing that ought to be subjected to the AI algorithms going forward. And then, you get the result that, sort of, comes out.
So, one huge risk in the midst of all of this, which you were talking about, which all of these experts were acutely aware of, is that there's a need as the technology develops to build and maintain trust in the tool, which is a bit of a black box. Which will require a great degree of transparency and good faith honesty with what's moving forward. And I think that we have, there are, there are facilitators and barriers to that kind of honesty and transparency. I think that those who are engaged in healthcare are truly motivated by doing what's right and good for patients to help them, but they're also wanting to protect their intellectual property and capacity. If you were asking about the money in it in the past, and I think money does make this world go round. And I think figuring out a way to align those incentives appropriately is something that there's a lot of work underway to try to do that well.
Grant:
That's right. And these predictive analytics are also not amoral in the sense that they're, they're trained by people to spit out the results that they're trained to spit out.
Dan:
Right.
Grant:
You know, if you think about the way that Facebook identifies comments as hate speech, there's already a priori, say, for example, that anti-abortion speech is hate speech. Then, of course, the algorithm's going to tag and eliminate statements that are against abortion. So, it's already programmed with the a prioris of the creator, so to speak.
So, is the RAI immune from these concerns? Do you see any of these similar concerns, sort of, baked into your own tool?
Dan:
I don't think it's immune. Part of what has made it easily implemented and adopted is that there's only fourteen items. They're transparently made available to the clinicians involved. And, most of them, if they are answered in the riskier toggle position, could suggest a target for intervention. So, it remains very clinically approachable. Where it might be vulnerable to some of this is that one of the biggest, so the two [inaudible], the things that have the most points attached to them, are age and whether or not you have cancer. And both of those are scored, in fact, as a statistical interaction. So, you get more points if you're young and you have cancer, because just about everybody gets cancer if they live long enough, because it's an age of disease, you get fewer points. But, but you get more points as you age. So, on balance, if you have cancer, you get a bunch of points, regardless of your age, added on to things. And if you get old, you get a bunch of points added onto it, which can lead to a kind of ageist set of assumptions.
I think one of the values, however, though, of frailty is that it does rather than just saying that a patient is too old, which may indeed be the case, but does lead to a set of assumptions that I think can lead to biased heuristics. The notion of frailty encourages people to think about what the consequences of that age is in the form of functional performance and limited physiologic reserve. That's really what's at issue. And you can have decreased physiologic reserve at a young age, and it's the decreased physiologic reserve which is materially relevant to a decision about whether you want to subject someone to a stressor.
Grant:
Yeah. And I also think that an important part of the RAI is also that it is a tool meant to facilitate decision-making. It still requires prudence and wisdom to execute as opposed to replacing wisdom and prudence with determinism.
Dan:
Thank you. I mean, in fact, that's the exact point that I, sort of, made to this panel. Is that I think there's a huge opportunity to use AI and machine learning to automate the kind of drudgery that consumes a huge portion of clinicians' time. And particularly in primary care medicine where they're just constant number of goals that they need to achieve of minimizing risks of problems that haven't emerged yet.
But within surgery, this, this notion of this is identifying the one to two patients that may come to your busy clinic on a given day who you really need to slow down and put on your big boy or your big girl pants and be the doctor that you've been training a lifetime to be and engage all of that prudence gained by experience over time and really give that the full-court press. The folks that otherwise are robust and low on the RAI. They can just, sort of, they can be processed safely and appropriately in a more industrialized sort of approach.
Grant:
Yeah. I'm gonna keep my eyes peeled for future papers with the RAI. This is really interesting. And, you know, I've seen you present on this a few times, and it's really fascinating work.
So, I do want to change the subject a little bit. Talk about your dual vocation of priest and surgeon and get into some questions about religiosity and health.
If the modern age is marked by anything, it's certainly marked by specialization. Healthcare industry is deeply specialized, but the rest of our life is really quite specialized. If we have a physical problem, we go see the healthcare providers, whether that's medical doctors, surgeons, nurses. If we have spiritual problems, we go see the spiritual specialists, whether that's clergy or chaplains. If we have psychological problems, we go see psychiatrists. If we have social problems, we see the social workers.
But, at the same time, we continue to talk a lot about whole-person orientation within the healthcare system, because we recognize that our lives are not specialized in that way and that our bodies and our souls and our psyches and our life together are intimately connected. So researchers and professors all types think that clinicians should be able to engage spiritual aspects of the lives of their patients. So, to what extent do you think patients actually want physicians to engage in their spiritual lives? Or do they just want you to stay in your lane, deal with my body, I'll go talk to my priest if I need something spiritual?
Dan:
Well, so the answer to that, I think it depends on the patient. There have been a few surveys that have been done along this line. It depends on the context, as well. So, when you're talking, when a patient has a life-limiting or a life-threatening disease, a survey that was done in the late nineties, I think—maybe early two thousands? no, late nineties—would suggest that close to seventy to eighty percent of them would want their clinician, their doctor, to engage spiritual aspects of their life. Maybe even as many as fifty percent would want the surgeon not only to participate but actually initiate prayer. I would guess that that would be substantially different now because our culture has changed a lot in the last twenty years. It'd probably go down. We're more secularized.
But I think that there's, in the setting of life-limiting disease, I think there's quite an appetite for folks to engage whatever it means in their mind, to, to reach out to something called religious or spirituality. Which, I think, fundamentally, is gesturing toward having some answer to metaphysical questions of: So, I'm alive. I will not be alive forever. What will I make of the life that I have left? To what end and for what purpose am I living my life? Those are fundamental questions that have, I think, always been part of what the best doctors can help patients discern. But they are not scientific questions that can be that, that yield to placebo-controlled trials.
Grant:
For sure. And even if patients do want clinicians to engage these aspects of their health that are social and spiritual, there does seem to be significant concern that a) clinicians are not competent to have these discussions. They know very little about the religious lives of their patients. They themselves might not know much about how to have these conversations. But then there also is this interesting concern that spiritual discussions could be coercive and threaten patient autonomy, as if physicians are proselytizing to their patients.
Now, you make a very important point in a number of your papers that medicine itself, as a course of practice, you're trying to convince people to do things that you think is best for them. And you're trying to convince them that your opinion is, is what is best for them. So, the question would be how should clinicians engage these religious and spiritual questions? Is there a role just to do what we call a "spiritual assessment" and then pass the patient on to the relevant religious expert? Or is it to actually pray with them or really delve into their spiritual lives? What's the limits of that relationship?
Dan:
I think it's something that needs to be discerned in the context of the diad of the people that are engaged, the patient and the surgeon themselves, the patient and the clinician themselves. Not all clinicians are up for. It would be inappropriate for them to do it. But I don't think that there's a hard and bright line that says that you, you shouldn't engage. I wouldn't say that, wouldn't say that. I mean, I think it'd be strong to say that medicine is at root a coercive practice. I think it's a persuasive practice. I mean, I think...
Grant:
Good point.
Dan:
...that, within the bounds of prudence, we don't typically strap people to a gurney and take them where they are expressly saying they don't want to go. But we will deploy all of our unequal power dynamic. We will puff ourselves up in the white coat. We will appeal to our multiple degrees on the wall behind us and our years of experience to, sort of, say, "I hear that you really think that you want to continue smoking, but here are all the reasons why that is just a bad idea." And, and do our best to persuade them.
And I think, at times, for those kinds of health behaviors that we are so certain have negative consequences that, that we, we tread right up to the line and some of us probably go over the line where it may be unduly influential. It may not be respectful in the way that we would hope our better angels would permit us. And it might in fact become coercive. But, short of those circumstances, we are constantly engaging in a discourse with our patients to try to persuade them to recognize what's at stake and choose what's best for them, which requires discerning what is best.
That is a teleological set of arguments. It is, it is moving toward a telos, a particular goal. The more concrete and specific that goal can be, the more powerful it becomes to guide the choices about what we do. So, the example that I give when I'm teaching medical students about this and other things is I, sort of, say, "You know, look, I've got a ninety-year-old in my clinic, who's got an early right-sided colon cancer and scores high on the RAI frailty score. I think I can get him through the operation, but it's not going to be an easy recovery on the back end. And the VA wants me to have that tumor out within thirty days of its diagnosis. So, as I'm trying to book him for June 8th, I don't do my job if I don't, sort of, say, 'Hey, Mr. Smith, anything really important on your calendar coming up in the next couple of months that I need to know about as we're planning your treatment?' And he says to me, 'Yeah. My granddaughter is getting married in August. Her dad died in a car crash two years ago. I'm supposed to walk her down the aisle.' So, the concrete specific goal for him to flourish as a human being, to live the fully articulated life that he is aiming after, of which good health is a means to that end, would likely be disrupted if I took him to the operating room at the beginning of May. And the consequence of delaying—beginning of June—the consequence of delaying until after his daughter's wedding is not typically easy to quantify, but probably a risk that, you know, a conversation between the two of us, we might be able to discern." That is fundamentally a teleological, normative, goal-oriented discussion about what it means to live the good life. And, for many people, religious traditions are the primary source of how we answer those questions.
So, I think that physicians of any or no religious background can become better and more skilled in the conversation strategies and discussions required to help clarify their understanding and maybe even help their patient clarify their own self- (patient's) understanding of what it means to live the good life. And make sure that—which may be a faithful life, in a religious bent—and ensure that the treatment plan is serving that good rather than the goods that are just intrinsic to the practice of medicine, which often devolve to, sort of, all-cause life expectancy.
Grant:
Right. So, this actually brings me to an interesting issue. You wrote an article for the Journal of Health and Religion, where you express a deep concern that the clergy had been banned from hospitals, at least in some locations, at the height of the COVID crisis. Now, COVID clearly made us think hard about the relationship between physical and spiritual health, whether it was the, the closing down of public masses—most of that was related to infection concerns—whether that's barring clergy from the hospital. And I try to avoid the COVID question, but we just can't, ultimately.
[laughter]
So, in what ways do you think COVID really exposed our fraught connections between health and spirituality? Do you think that made those relationships worse? Is this an opportunity to rethink those questions? I know that the Catholic churches in Pittsburgh are, have now decided to reopen. If you've been vaccinated, there's no masks. But there was this period in time where there was a real barrier to faithful people practicing in deference to our concerns about physical health.
Dan:
Mm-hmm. I think that I was surprised in the early months of the pandemic, when I wrote that piece, about how—I had always been aware and had feared that we had divest, that the churches had divested their past claim on the lives of their members to the hospitals, right. That, that we—and I think that this started to happen a lot in the 1950s. You know, you look around here in Pittsburgh: Mercy hospitals; St. Margaret hospital; Montefiore, which is a Jewish hospital; Presbyterian hospital; Passavant hospital, which was founded after a Lutheran. Consistent with the story that I told you about coming out of antiquity, the institutions of healthcare are, have historically been, very frequently, funded, fostered, and maintained by religious communities, which in the United States context is typically Christian.
And we, back when the powers of medicine were constrained largely to fixing a broken bone and takin' out a life-threatening tumor, we typically, sort of, handed over that—because our goals and our values were consistent with the goals and values that were intrinsic to the healthcare enterprise—we divested that sort of authority and control over to the hospitals. And then, the power of medicine has expanded in a much broader way and plays to, makes plays to efficiently manage a much broader aspect of what the human condition is, about which there is normative, moral, and theological disagreement that is there. And I think we're all, kind of, a little bit confused about what all that means.
All of which is to say, I'd seen and had written about this phenomenon, but was still astonished by how completely and unquestioned it was that everyone, including pastors and priests, were barred from attending to people who were dying in hospital without the benefits afforded by last rites.
There was an interesting—and I had expected that the clergy would be much more courageous than they turned out to be or a little bit more willing to stand up and be assertive. There, there was an interesting piece in the New York Times about some priests of the Roman Catholic Church in the Boston area who took on increased personal risk and, essentially, quarantined and lived in the hospital there their whole life, so that they could attend to those who were dying with COVID. But no one was really thinking about ways of coming in and out or going in and providing these, these roles.
I was very pleased that here at UPMC, when I raised the question and concern, that there was, it wasn't a deliberate choice to exclude the clergy. They were, just, sort of saying, "We've got to, we've got to, from a public health perspective, put the brakes on visitors in the hospital, but they said, of course, clergy are essential personnel just like others. And, you know, assuming that they can pass the normal screening process and they follow appropriate personal protective equipment, which we will provide, we want to afford the opportunity to folks to get in." So, very rapidly UPMC was implementing a posture that was more permissive.
But then, I wrote that story. I shopped it at a number of national news outlets as an op-ed. It didn't get picked up. Circulated it through professional organizations. And it's still really surprising to me that, I think it is much, you either have no one being permitted in, or you have the, sort of, the, in my opinion, foolishly aggressive folks that are throwing people to the wolves to maintain it as a political point. It's, it's sad to me that it's become politicized. But…
All of which is to say, I think we have surrendered control of our bodies and their care to the healthcare institution. And the consequences of that, I think, we became aware of in a different sort of way. And whether that is altogether prudent and consistent with faithful Christian discipleship is not entirely clear to me.
I hope, I really would have, I had hoped that the clergy and the churches had been, would have been a little bit more assertive. But...
Grant:
So what's your sense, do you think that the, why do you think the clergy were not more assertive? Was it just the sense in which we just don't know the physiology enough and we're really afraid that we're going to be vectors? Was it just being exceptionally risk averse? Was it their own recognition that what's really important here are people's bodies, not their souls? What's your sense of what, sort of, prevented clergy from being more courageous in this moment?
Dan:
You know, I'm not sure. And I, and there are all a lot of different motivations at play. In the most favorable light, and I think this did, in fact, motivate some of the own things that I wrote to my own church communities about why I was in favor of shifting to a virtual environment at start and to refrain from physical contact, was that we didn't know a lot. And it's quite clear that the Gospel enjoins us to care for the least of these. And even though it might be comparatively low risk for someone of my age and vitality to go about business in a usual fashion, our churches tend to be predominantly older, with people with comorbidities, and it's set up to be a spreader event. So that, out of deference and Christian charity to those who cannot protect themselves, that we constrained voluntarily our own freedom. Right.
I reject entirely the discourse from certain quarters of the Christian church, which talks about a right to worship in person and in the flesh. That's a, that's a political context, which has no precedent any place in the Gospel or in the scriptures. So, that doesn't get me very far.
But I also think that it laid bare some weaknesses in the theological formation and the cultural confidence of clergy to actually lead. I mean, I think that with fifty years of declining enrollment in the mainline Protestant church, which no longer experiences and enjoys the kind of cultural privilege that it once had. The discourse for thirty or forty years is how to make oneself relevant. There's been more and more surrender of scriptural language and theological terms to couch them in the terms of culture and business and relevance and marketing and all that sort of stuff. I think, we just, it laid bare how much has changed and, potentially, been given away in this period of transition.
I do think that, you know, there's, there's something to be said that every five hundred years the church goes through a major transition of some variety, and I think that we're kind of in the midst of that. You and I both have read Alasdair MacIntyre. And in the same way that moral philosophy is disordered and people are using the word "justice" but meaning completely different things by it, I think that the church has also, kind of, lost its moorings. That we continue to talk about sin and redemption and salvation and justification, and reconciliation, but we're, but what we mean by those things is not easy to pin down, and we've lost the ways to talk about it.
Grant:
Yeah, this will be really interesting to see where we emerge, after COVID, in terms of our faith communities. At the very least, we're certainly not gonna be in the financial positions we were. And the financial positions we were in were not particularly good in the first place. We'll wait to see what, what happens.
Dan:
God will raise up his church. Right.
Grant:
Yes.
Dan:
Right. About this, I am clear.
Grant: That's right. And one thing that I always return to is this prediction that Joseph Ratzinger, who would become Pope Benedict XVI, is that the future of the church would be small, vital communities of exceptionally committed believers. That will be the roots that will emerge from, sort of, the fallen tree. And to the extent that COVID forced out folks that were sort of on the margins, that were minimally committed, and what remains is the small, vital community could actually lead to a resurgence of the church in its power—and I mean spiritual power.
Dan:
There's all sorts of horticultural language and metaphor pruning the scriptures. Right? So, this has accelerated a process of pruning that was already well underway.
Grant:
Absolutely. So, last set of questions, and I'll keep it somewhat brief. I want to talk a little bit about this concept of religiousness, religiosity, and how that relates to health. You've noted in one of your early papers that quote unquote "being religious" can add two to three years to your life expectancy, similar to physical exercise.
So, this is an interesting finding, but the implications are a little hard to wrap your head around. What's the mechanism? I guess that's a huge question, but also a basic question. What's the causal mechanism that drives this relationship between religiousness and health outcomes?
Dan:
Okay. So let's, sort of, back up a little bit and put that paper into context a little bit. The, there has been longstanding an increasing interest in trying to look at the empirical associations between religious belief and practice and various health outcomes. People will talk about religiosity, not a term that I'm particularly fond of, I like to talk about religiousness as practiced within communities. I think that there's, that people also want to talk about spirituality. Spirituality is a part of religion, I think. I think there's no way to undermine it or to phrase it in any other way.
Bob Bellah's book, Habits of the Heart, told a very interesting story about a nurse named Sheila, who was bold enough to name her own religion after herself. She was trying to sort of, you know, "I'm not religious; I'm spiritual. I have my own little religion; it's called Sheilaism." And some basic tenets around those, those facts that we, there is something about the human condition that leads us to searching for meaning and value and an interpretive framework. That leads us towards some kind of sense of: to what for and what purpose am I living my life? And religious systems focus that.
So, when you look and you measure people who do and do not go to church or some religious activity on a regular basis, those that regularly go to a church or religious service live longer than others. And you can control for their physical activity. Some people thought that that was just, you know, well, you're, you're, self-selecting the people who actually are healthy enough to get in a car or walk down to church. Well, you can control for that. And you can control for depression and you can control for obesity and a whole bunch of other things. And, still, this effect is observed.
I suspect that, the two mechanisms that I would propose as being the most likely to answer your question: one is materialistic; the other one is more theological and spiritual.
The materialistic explanation is that we are fundamentally interconnected, dependent social beings. And medicine has an anthropology that sees the human being as an isolated individual. And to the extent that we continue to try to speak of the health of the individual in isolation from the wider community and their network of interdependence, it does not capture the fullness of what that health is. And to the extent that regular participation in one of the most powerful social organizers of interdependent relationships, namely religion or Christian churches in the United States context, that it is functioning as a proxy for people who are living more interdependent, interconnected lives. Right. And that that has a salutary health benefit.
Now, why might it have a salutary health benefit? Here's the theological perspective. Again, if from a Christian perspective, we take God to be revealed most clearly not in isolation but in relationship, particularly the relationship of love shared between Father, Son, and Holy Spirit, one substance, three persons. It means that at the root and the absolute essential nature of all that is, seen and unseen, is relationship. And if you are following after a notion, an ideal, of health which uproots the individual out of that relationship, either to other human beings or, also, to God, because it is agnostic, in that act you are denying important aspects of what it means to be a human being. And that that would have a health consequence is not surprising.
Grant:
Right. So, I've read a number of papers from folks that self-identify as religious, especially Christians. Researchers who study this religion–health phenomenon. And you've weighed into these debates yourself. So, one hallmark of American religion is to understand religious faith therapeutically. So, we see religion as worthwhile to the extent that it can offer us other material benefits like friendship, morality, health, but the metaphysical claims are less important. And many theologians would argue that this instrumentalizes and ultimately undermines religion itself.
So, I guess the question is to what extent are Christian researchers, perhaps unwittingly and ironically, contributing to a, sort of, secularization of their own religions while they try to highlight these instrumental benefits of religion? Is it well-meaning, but perhaps a self-defeating project for religious people to demonstrate the material benefits of religious practice?
Dan:
Well, I think the proof is in the pudding. I think the risk is definitely there. I think that those who have, you know, started this field and got it going did us a great service in that they were able to garner enough of a certain kind of data to get their foot in the door and keep it open. To sort of say, "Look, there's something going on here that, no matter how you look at it and how you slice it, deserves further consideration." To ignore something that has an effect size similar to regular exercise, the kind of thing that we coercively or persuasively engage with our patients to convince them to do all the time, is something that deserves, absolutely, further consideration.
But I think that, you know, there are some in that camp that will, sort of, suggest that it would be malpractice not to prescribe religion in the same sort of therapeutic way. I think that that, I think that's misguided for the theological reasons that you made. It becomes fundamentally idolatrous, right? What are you, are you trying to be faithful in your attendance at church and your participation in religious community? And are you trying to live your life in accordance to something that one might call discipleship in a Christian idiom? Or are you doing it because you want to live long and prosper? You know. So, we've got all sorts of health and wealth. God and Jesus wants me to have a Rolex and a Mercedes. Nah, I don't think so. But there's a tradition in Christian, some kinds of Christian health-and-wealth gospels that are teaching that. And I think these people may be unwittingly falling into that for any variety of reasons, but I have no doubt that their motivations are nothing but good. They're attempting to highlight, in their own way, using whatever media capacity they have, to shine a light on one of the advantages of, of living into the purpose for which, as a Christian, I think we were all called.
Grant:
That's right. So Dan, we have reached and maybe even exceeded our time limit today, but I wanted to thank you for joining me. This is a really fun and lively conversation, especially as you're recovering from surgery. I'm grateful that you took some time to chat. So, hopefully we'll see each other again, soon, in person.
Dan:
In the flesh. Yeah.
Grant:
That's right. And again, thanks so much for joining me today. I really appreciate it.
Dan:
Most welcome. Thanks for the conversation.
Grant:
All right. Take care.
Dan:
Take care.