Podcast Episode Transcript: Standardized care or clinical flexibility? Finding the right balance

With L. Gordon Moore, MD

Gordon Moore: Welcome to 3M’s Inside Angle podcast. This is your host, Gordon Moore, and with me today I have two guests. The first is Dr. Elvin Geng, who is the professor of medicine in the Division of Infectious Diseases and the director of the Center for Dissemination and Implementation at the Institute for Public Health, both at Washington University in St. Louis. And also Dr. Michael Lipsky, who is a retired former political scientist mostly with a career at MIT and retired grants officer at the Ford Foundation and is also the author of several books, Street-level Bureaucracy: Dilemmas of the Individual in Public Services published in 1980 by the Russell Sage Foundation and reissued with a new introduction, additional chapter 2010. He’s also the author of Protest in City Politics and Nonprofits for Hire: The Welfare State in the Age of Contracting. Welcome, doctors Geng and Lipsky.

Dr. Lipsky: Thank you. Glad to be here.

Gordon Moore: So let me state the premise for the conversation today. I read a piece in the New England Journal, a viewpoint by Dr. Geng, where you were talking about working in an urgent care center. And you were describing the complexity of working with individuals whose needs were not strictly clinical but had a direct impact on your ability to work with that individual and help them meet their clinical needs. And you described in that the concept of being a street-level bureaucrat where your decisions and how you chose to approach that would impact the likelihood of that person getting what they needed. And I’m curious. Where did this all come from, and why did you decide to write this up?

Dr. Geng: Yeah. Thanks. First of all, it’s just a great pleasure to be in this conversation with Dr. Lipsky. And I think some years ago—and I don’t remember exactly where or how—I came across Dr. Lipsky’s idea of street-level bureaucracy. And it struck me at the time—I think I was a resident in internal medicine at Columbia in New York City—that these kinds of discretionary decisions were made by physicians sort of constantly every day and many times by sort of relatively inexperienced physicians, of which I was one, residents.

And it just struck me that these decisions oftentimes have enormous consequences, but yet they sort of fly under the radar. They’re not taught about in medical school, that we physicians make these decisions without a tremendous amount of internal discussion with each other. And, in fact, I think a lot of people think of them as sort of nuisances in their sort of overall quest to deliver clinical care, but I was fascinated because it seemed to be sort of ubiquitous and all around us and not always visible or interrogated. So it’s been on my mind for a long time, but this particular encounter got me thinking about it again and led to this piece, which I hope will provoke some conversation about this phenomena and how we can best use its presence to deliver the kind of health care we all want to deliver.

Gordon Moore: Well, tell me—maybe use this case or a different one, if you want, that describes the kind of decisions you’re talking about.

Dr. Geng: Yeah. I think these happen all the time, right? So when you’re in clinic and a patient is 15 minutes late for their visit and the front desk calls you, and you’re waiting in the back, the physician can say, “It’s too late. Have the patient come back another day.” Or the physician can say, “I’ve got 15 minutes left in the encounter. Let’s see what we can accomplish. We’ll give it a crack.” In general, no one will say whether or not that’s the right decision or wrong decision, but it could have enormous consequences for a patient who may have struggled to get to clinic, maybe spent money on transportation and may or may not have an issue that could turn out to be more urgent than is immediately apparent. And so I think this ranges to more acute issues and clinical ones, but they really do seem to be ubiquitous.

I ended up as a part of medical teams that took care of patients who were candidates for surgery. And this is another sort of decision point in medical care that often is very much up to the discretion of an individual whether or not a particular surgical procedure would be offered. And, again, decisions that have enormous consequences for people but which we, I think, collectively don’t have in as much focus as we could.

Gordon Moore: But isn’t that just the day-to-day work of a clinician, making a clinical judgment based on the scenario you have in front of you?

Dr. Geng: I think that’s absolutely right; it is the day-to-day work. And I think this piece kind of is meant to kind of bring that out, right: it’s the day-to-day work. And if it is the day-to-day work, then then let’s shine a light on it. Let’s make that kind of work sort of more visible to ourselves, first and foremost, but also collectively. And I think maybe that’s the first step in a process by which we can do that day-to-day work better.

Gordon Moore: Well, let me let me pivot to Dr. Lipsky. And I’m thinking that this sort of lived experience of I am sitting in the back, I get a call, patient’s late, I’m deciding which way to go is a lived experience in lots of different ways, as Dr. Geng is saying, but you took this to a level of deep analysis and formality in your description of street-level bureaucracy. So I wonder if you could tell us more about that.

Dr. Lipsky: Sure, Gordon. Let’s go back to the origin story. I was about to review a book on the police, and I realized that the problems of the police were very similar to the problems of social workers in that they had a tremendous number of rules that they needed to implement and kind of infinite amount of observations they could make to do the job as best they could. But their resources to achieve those objectives are limited. The result of this is that they have to make decisions and find rubrics that allow them [social workers] to make the decisions about their patients or their clients or the people that they may encounter in the street very quickly and using good judgment and the resources at hand.

The number of decisions that Dr. Geng and his colleagues have to make are arguably infinite, so we develop both rules and private mechanisms to process the work in an efficient way and one that is satisfactory to all concerned. I started with police and social workers, and then I realized that there are an enormous number of people working in the public sector who have this—these circumstances. They are authorized to use discretion. They have to implement these rules, and they don’t have the infinite capacity to do that.

The implication of all this is that they make decisions about what they will do and what they won’t do, which directions to go continually, and in such a way as to essentially make the policy that their organizations are going to carry out. So when Dr. Geng decides to spend a little bit more time with a patient, he’s essentially saying this hospital—this emergency room is going to be a—is going to go in this direction. But you can’t always go in that direction, or else you’ll run out of time and resources.

So what I say essentially in Street-Level Bureaucracy is that the decisions that people make under these circumstances of uncertainty with regard to patients or clients or whatever becomes the policy that we are implementing in these various fields—in education, in health, in policing, in social work, and so on. I have more to say, but let me stop there and let people catch up.

Gordon Moore: And when you hear that, Dr. Geng, is that—were these the thoughts that were going through your head as you were thinking about the circumstance that led to the article and having read his book?

Dr. Geng: Yeah. I think that—in that particular moment, which in many ways was sort of very mundane, right, like—as Dr. Lipsky has said, these kinds of decisions confront practitioners virtually constantly. I think the connection to the ideas that were just discussed I think sort of came to me a little bit later as I sort of was reflecting on it. But I think the lived experience of sort of standing at that fulcrum of the system and the people is an interesting one, and I think interesting for a number of reasons.

One is that the sort of nature of decision making is very interesting at that fulcrum but also just that the formalization of this concept in Street-Level Bureaucracy, to me, was so eye-opening because it helped to render something that was ubiquitous but invisible sort of seen. And I think that’s oftentimes what we write for is to try to bring the complexities of life and of practice and of medicine in this case to the conscious sort of level and provoke some conversation about it.

But I agree that there—as Dr. Lipsky said—the discretion that people can apply is not infinite, right? On another day, if there were five people in line, I could not have made the decisions that I made. It was just that on that particular day, there weren’t any other patients there, and so taking a little bit of extra time became possible.

Dr. Lipsky: One of the keys, Gordon to understand what my thinking is is that we expect street-level bureaucrats to follow the rules but also to be open to the possibility that they could do more or that there are exceptions to the rules. If you look at the professional documents of the various job categories, you see that everyone should be open to the possibility that they can be heard beyond just being processed in a bureaucratic manner. And this is a strength of our public services.

For example, a kind of a—an example that’s simplistic is that you want the police to stop people who are speeding, but if there’s a pregnant woman in the car, you want the police to have a different view and escort the car to the hospital. Being open to these possibilities, as Dr. Geng was in the case of the woman he describes in his piece, is the essence of good professionalism. And so how we structure our public services so as to be as efficient as possible but also open to the possibility that there are exceptions and that people deserve to be heard is the challenge. And it’s an ongoing challenge that all the rulemaking that you might want to see isn’t going to fully solve.

Dr. Geng: I really like that point. I think in my sort of career in medicine, which is not that long, I have seen this movement towards increasing protocolization, standardization, right? And we have a lot of practices now that are—for which each hospital or each unit has sort of written rules about. And in general, I think those are good, right? We want to eliminate variability in the bad direction. But on the other hand, as you say, it is—we will never be able to write rules for every situation, and at the end of the day, some provider is going to have to apply some discretion some of the time. And so that—I think that remains true and should remain a part of how we think about making medicine and systems sort of human, right, because that’s what that discretion will allow for.

Gordon Moore: It’s such an interesting fulcrum right there for me in that we’ve had this long history of the art of medicine. We do what we think is right based on our training and what we perceive at the moment. And this move more recently for evidence-based medicine, the use of guidelines in care. And I struggle at times with a disease-oriented guideline in a person with multiple conditions where there is a clash between the guideline recommendations based on the different conditions and recognize that I need to adjudicate that clash, and that’s my role. And so that, to me, is the—I need permission to step outside the rules and do what’s right for that person at the time, say this speeding car goes to the hospital fast. And I’m wondering if there’s—how we describe that and how we allow for permission without giving up standardization.

Dr. Lipsky: Yeah, that’s the essential question. Well, I think the first answer or the first thing you would say is to recognize the problem and to find a way to authorize discretionary actions that go outside of the boxes that the system set up in order to be most efficient. We haven’t yet mentioned payment systems, Gordon, as the driver of many of these decisions. But I think it’s a good entry point into the discussion because, as I understand it—and I’m, of course, not a doctor—is that the pressure to bill properly and to take as little time as possible, while giving full attention to the presenting problem creates a series of boxes to check that are very powerful relative to the instinct of the physician that something is wrong here. I need to spend a little bit more time.

One sympathizes with a very productive professional who is processing a lot of people on a kind of a mass basis. But if he or she doesn’t have authorization to take a look and see what might be a problem outside of the boxes that are being checked, I think you’ve got a kind of standardization that ultimately leads to missing a lot of possibilities and at the same time degrading the job of the physician who was trained to be on the lookout for all sorts of things outside of the things that are most likely to be presented.

Gordon Moore: That’s a perfect segue, I think, to what was actually happening in the story that you wrote. So you were—Dr. Geng wrote you were going to go into a room to see a patient for a particular clinical scenario. Just give me—give the audience a taste of what actually happened.

Dr. Geng: Yeah. So recollecting this now from retrospect. As a physician, when you hear about a case, the cogs, the machinery starts turning, and of course, my machinery is first and foremost clinical. So I’m thinking about what kind of problem could be going on, what kind of pneumonia this could be, having heard perhaps that that was the sort of main issue from the triage nurse. And then when—you bring that assumptions and those questions to the encounter, and then you find the patient is actually thinking and prioritizing something completely different.

And it took me a moment to try to pivot, right, because I wanted to chase down this pneumonia issue. I needed to talk to the patient about pneumonia. I wanted to listen to her breathing. I might want to order a chest X-ray or some other things, but to do any of those things, I’d have to have a conversation about the pneumonia. And after a few attempts here, began to realize that I wasn’t going to get very far in this conversation about the pneumonia unless I made at least a gesture towards addressing the issue that she wanted to talk about.

And I think that the real lesson for me there was that even though it started as a gesture, because I was still convinced that the problem was the pneumonia, at the end of the day, turns out that the problem that she had brought and was prioritizing was the real issue at hand. And it’s always—clinical encounters are always complicated, right? You enter them with a set of assumptions, and oftentimes you find that those assumptions actually were not the right ones. And it’s important to be able to pivot in the moment and react in a slightly different way.

Gordon Moore: And the issue at the moment that she was bringing was around housing and saying, “I need a clinical diagnosis so that I could be prioritized on a housing list. Otherwise, I could be homeless.”

Dr. Geng: Yes, exactly.

Gordon Moore: You wrote in your piece—and I’m going to quote it here—”Is housing a doctor’s problem?” I think not, but it had everything to do with that encounter, and it has so much to do with a person’s probability of health and wellness, wellbeing, and prevention of disease. All of a sudden, it is.

Dr. Geng: Yeah. I have two comments about that. Number one is that the health care system in America does give physicians a certain width of discretion, right, that other providers in the health care system don’t have. Nurses, obviously, do so much the heavy lifting in health care, but the system doesn’t give them as much sort of berth as physicians. And so sometimes in a system that doesn’t always work, there are opportunities to use that that width in a way that has nothing to do with anything clinical but rather just to kind of push the system. Maybe, perhaps, you could say metaphorically treat the system in a minor way to help patients get what they need.

Dr. Lipsky: One of the things, Dr. Geng, that I thought about when I read the wonderful piece was that this was a woman who had a certain aggression. She presented with a certain aggression and positivity. She knew what she wanted. She wasn’t afraid to speak up. I think about the passive patients who have equally important issues but who are for a whole bunch of reasons are not inclined to be as assertive as the patient that you described. There’s a range of potential discretionary behavior to try to tease out what’s going on with people who are afraid of the doctor, intimidated by the doctor, or not inclined to be assertive but who probably deserve as much attention as the people who are assertive and able to speak out.

Dr. Geng: Absolutely. One of the things I think, in medicine that remains this timeless element of it, which is really one of the sort of most interesting aspects of practicing medicine is this notion of a relationship. And I myself have been struck by in my primary care practice or my sort of—when I was in San Francisco, that sometimes it took months or years to sort of discover really what the basis of a particular problem might be. And I think that there is still, despite all of our technologies, just no substitute for that, that relationship, and so relationship building continues to be an important element of the practice of medicine and probably always will be despite our most advanced technologies.

Gordon Moore: When you described the scenario in the article and also listening to you now, it makes me think about the degree to which you were starting to put your own presumptions aside as you were engaging with this individual and then beginning to listen more and figure out through that listening what’s the needful intervention at the moment and had to bring that to bear. And that’s interesting that you would do that. Again, to extend the quote that I started with before, when you asked, “Is housing a doctor’s problem?” you said, “Perhaps not. But the consequences of poor housing, they’re no different from those of non-adherence of the wrong antiretroviral treatment, so therefore, we can’t afford to ignore them, even if we can’t single handedly remedy the problem.”

I think that’s super insightful, and that gets at what you described as the power that a physician may have and be able to extend beyond their typical sphere of influence to influence other parts. And that sounds like a street-level bureaucrat is making policy at moment to moment and is adjudicating that policy based on what a person needs.

Dr. Lipsky: Right. So another way—let’s move to the police for just a second. Is it that it’s public security or comfort with the level of security that we want the police to deliver? And that is different from saying that police work is the sum total of what every police officer does. So I think you’re obviously introducing, Gordon, the question about whether we want to think about what the doctor’s responsibility is or whether the doctor is a link in a health policy regime contributing to the best outcomes for the population.

Gordon Moore: Yeah. I think of it ultimately that a doctor, as any other member of the health care delivery team, might recognize an issue, may have different sources of influence, and might be able to act on those things. I think of the—most people who are interacting with the health care system have needs that are met by multiple people, so there is—it’s a team sport. And as Dr. Geng mentioned earlier, on a different day, he might have had five people stacked up in the waiting room and wouldn’t have time to pivot and do this work.

But what I kind of want to get into, Dr. Geng, some of your work in Dissemination and Implementation, thinking if this is a core concept, this idea that we need to think about the adjudication of policy at the moment-to-moment work, how do we do that in a way that expands beyond just the moment-to-moment opportunity? Is there a way that we build this into the way we deliver care? Would that empower us, and how would we do this better?

Dr. Geng: Yeah. So a couple of comments on that. So as you mentioned, in my kind of day job, I’m mostly a researcher, particularly now, and I do dissemination and implementation research. A lot of that work is focused on systems and how systems function, how systems should function. But the issues of discretion are all sort of where the cracks are in the edifice of the systems, but I think they interact in two ways. So first, this system can put more or less sort of pressure on these discretionary decisions.

And I want to, if you will, indulge a quick anecdote from the time that I was working in China maybe now 15 or even more years ago working with HIV treatment in China, where there was a very much, in many cases, kind of like a fee-for-service type approach. And I worked with a very good physician, a really nice guy, astute guy who was the director of the unit. And I’ll try to be quick about this.

But we would get sort of leadership who would come to the hospital, and one day the leadership would say—from some other higher bureaucratic level would say to the physicians, “This hospital is losing money. You have to make sure the patients pay for their services.” And then the next day, someone else would come and say, “This hospital’s mission is to treat patients with AIDS when no one else will. You should treat these patients like your brother or your sister and take good care of them.” The people who work at the hospital are faced with these completely diametrically opposite directives.

So I remember one day this person came in with an AIDS-related complication. And this friend of mine called the family, and I was just sort of observing. I was doing other things, and I noticed this conversation was going on. He called the brother, and he said, “Listen, if you want your brother to be treated here, you have to bring us this much money. That’s the price of admission.” And the guy said, “We just don’t have it. We spent all the money. We went to three hospitals before we got here.” And he said, “No, I’m sorry. This is the way it is. You have to pay this amount of fee up front, and this is what it says on the admission criteria. If you don’t do it, we can’t accept your brother.”

And they went back and forth like this for a long time. And then, finally, the patient said, “Okay. We can’t do it,” and so went upstairs, packed the bags, and began to leave the hospital, at which point the physician, who was my friend, called him back over and said, “Listen, don’t worry about it. We’ll sort out the money later. You can stay, and we’ll treat your brother.”

And it was a harrowing experience to watch. But it made me realize that this is a man who is being—the system is telling him totally different things, and he has to figure out how to navigate that and meet sort of the needs of both masters. And I think—in this case, I’m glad he did the right thing and made that decision. You can imagine another person would have made the opposite decision to serve the imperative to make—to not lose money at that hospital, which would have cost this gentleman perhaps his life. So systems do interact with the kind of decisions and discretion and the consequences that come from it.

Let me say something briefly about some work I’m doing with colleagues in Zambia with CIDRZ, which is an organization, an NGO that helps provide care and treatment for HIV in Zambia. And we are sort of testing in the field the hypothesis that discretionary awareness and the right discretionary sort of decisions from a sort of patient-centered point of view can be taught in a public health system. And we have—there’s a few premises for this. One is that I think that most health care workers, be they doctors or nurses, sort of want to do things that are helpful deep down inside. They have a pool of internal motivation to do those things that help people.

At the same time, they operate within systems, and the systems force them to sort of make decisions, make heuristic decisions. There’s more work than can possibly be done, and so some—these decisions have to be made but that the—through some training and mentorship, that these kinds of opportunities can be surfaced and can improve health care services. And in addition, our hypothesis is that it can improve the experience of being a health care worker. Yes, that you can get more satisfaction out of your work and more morale.

So let me stop there. That was quite long-winded, but I hear Dr. Lipsky and would love your thoughts on this.

Dr. Lipsky: Oh, I love that anecdote. One of the defining characteristics for me is that street-level bureaucrats want to do a good job in some way. So want to do a good job, but I—qualified by—that by saying in some way because they don’t get to do their ideal job. They structure their job in their—both in their minds and also at work in such a way as to do a satisfactory job according to the inner light that they have come with and experience.

And so one thing that an organization can do in which street-level bureaucrats are performing is help create an environment in which what it means to do a good job includes the kind of responsiveness that you’re talking about. And they can do that either systematically by creating rules that are consistent with a humane perspective, if you will, and they can also do that by giving more discretionary power to people on the assumption that they will use that power for good.

Gordon Moore: One of the things I hear often from colleagues in practice and read about in various articles about burnout is that there’s the sense of physicians and other clinicians that doing a good job is being defined externally with strong external rewards and incentives that can be—cross purposes with what they believe is doing the right thing for the person. And I think that’s inherent in your—Dr. Lipsky, your model of street-level bureaucracy.

Dr. Geng: Yeah. Absolutely. So in this day and age, when quality improvement has become very, very de rigueur in health care delivery, I think individual providers often feel sort of caught between a rock and a hard place, right? So I can think of many examples. You’re supposed to check an LDL in all your patients, right? And if you look in a hospital and you say, “Okay. Well, this person—this physician didn’t check an LDL in X, Y, and Z people,” and you go back to that physician, you say, “You get a—you’re going to get dinged for not checking an LDL.” Given the complexity of health care, there often are good reasons that no LDL was checked, right?

And it could be that the patient had other pressing issues that were life threatening in the weeks to months from that visit, and so checking an LDL would be sort of—maybe aside from the point. Maybe the patient has a phobia for having blood draws and will only allow one, and so you have to choose which ones to prioritize. Many of these considerations are idiosyncratic and personal. And so the rules, if too rigid, produce contradictions, and the contradictions produce burnout.

Gordon Moore: So I’m thinking of recommendations or statements either of you could make to policymakers, to people who are putting together programs to improve health care outcomes, cost, and quality. What would you recommend to them? How do they this issue of street-level bureaucracy and do something different that makes the world a better place?

Dr. Lipsky: I’ll start by just suggesting that all of the energy is being put on systematization and cost control that brings—that supposedly brings efficiency and, arguably, effectiveness, but there’s very few, virtually no pushback around the consequences of constraining discretion professionally. So I would—I’d like to see a lot more conversation about this and focus on the consequences of forcing this conformity to the cost containment standards and other kinds of rules that we experience in medicine. And I could make comparable comments in the education field and other fields as well.

Gordon Moore: What do you think, Dr. Geng? Any comments, recommendations?

Dr. Geng: Yeah. I think that—one of the things that I like about the concept of street-level bureaucracy and of sort of making the discretion visible is that at the end of the day, these are still human systems and not mechanical systems, right? And so I think acknowledging that and working to create systems that both sort of ensure some level of function but make the most out of what people can add to that would be important. And that, I think, comes from maybe managerial respect for the experience of frontline health care workers. I think too often policymaking at the macroscopic level sort of is doing maybe what Dr. Lipsky was suggesting, like looking at the bottom line, cost, containment, efficiency, and loses sight of what people actually do in that system to make it work and to make it work better.

But I also think that perhaps, in addition to the groundbreaking conceptualizations that Dr. Lipsky has put forward, there could be more of an empiric research agenda around this in health care and some more attention to it, scientific attention to it. And focus on it might help us also devise systems that make the most of it.

Dr. Lipsky: I think, also, we can talk about systems of care or systems of policy in any of these areas. It doesn’t have to be the doctor who discovers that this patient urgently needs this form to be signed. Nor, one supposes, does the system require that the doctor do it rather than a qualified nurse practitioner. So if the team—and I think this is an idea I’ve probably gotten from you in conversations, Gordon. If the team was responsible for probing areas of potential interest outside the presenting clinical perspective, one could see—the system would become responsive but perhaps at little cost to effectiveness since the doctor is now confident that the patient’s needs are going to be met by people other than him or herself.

Gordon Moore: And I’m going to close by making a statement around the way we measure quality in health care in the United States by looking at smaller and smaller parts of what’s happening between a person and their clinician or clinicians, measuring discrete processes of care around disease management, where I think that level of bureaucratic rulemaking is dysfunctional and has added to the stress and strain and cost of health care delivery and as a prime example of where we need to back off. We need to look at larger measures that get closer to what people value in health care and create a little bit of space for the clinicians to make the right call to not say in every instance do we need to do an LDL. And I can be the judge of that, and we can look at better markers of how am I performing as a clinician for people.

And with that, I’m going to look forward to potential future conversations with the both of you. And I want to thank Dr. Elvin Geng and Dr. Michael Lipsky for your time today.

Dr. Lipsky: It’s my pleasure. I’ve enjoyed this immensely.

Dr. Geng: Thank you

Go to Podcast