Interview with Dr. Brent James
Hedrick Smith: We're in Salt Lake City talking to Dr. Brent James with the Intermountain Health Care System. So tell me, what's this chart about?
Dr. Brent James: Several years ago Dr. Kim Bateman visited a number of our rural hospitals. His intent was to tell them about state of the art community-acquired pneumonia treatment. He gave a series of lectures and recommended a choice of antibiotics, as the most effective antibiotics for treating these patients.
He then came to me and said, "Did it have any impact? Did it make any change?" Well, we tried to measure two things. Number one, did they change their use of initial antibiotics? And then across all patients, regardless of whether we did or not, did it change survival? Did fewer patients die?
We've properly balanced the patients' severity of illness, their age, their gender, and whether or not they followed Dr. Bateman's recommendations. What it shows us is that at the ten hospitals where Dr. Bateman gave his lecture, about 29 patients lived who otherwise would've died, because they followed Dr. Bateman's advice relative to choice of initial antibiotics.
Hedrick Smith: So what you're doing is comparing patients who did get the antibiotics that he recommended and those who didn't get it?
Dr. Brent James: Well, it's actually, more conservative than that. We used a group of hospitals where he didn't give his lecture and we didn't distinguish whether they followed his advice or not; was there a difference in mortality rate that we can directly associate with Kim doing his academic detailing.
Hedrick Smith: But you don't know whether or not the doctors used or didn't use his method?
Dr. Brent James: Well, we actually do. We tracked whether the physicians followed his recommendations. We're basing this on whether or not those physicians chose to use the antibiotics that the protocol recommended that Kim shared. Previously, we wererunning at about 22 percent antibiotic compliance. Which increased to 40 percent, which is a significant jump. That's the shift that produced the [results] that we saw just a minute ago.
Hedrick Smith: So you're saving 50 to 70 lives a year?
Dr. Brent James: Well, for pneumonia. Of course there are many other areas where we apply this as well. It also reduces the cost of care. In conjunction with this, if we went back to the original study, the cost of care fell by about 12 percent. The main reason is if you use the right antibiotics most patients don't suffer major complications. If they don't suffer the complications they don't die. But more than that, if they don't suffer the complications we don't have to treat the complications. But we so often fail in our treatment. So it turns out you got a better medical outcome, and at the same time the cost dropped.
Hedrick Smith: Dr. Bateman was telling us that it took about four years to get full compliance at LDS Hospital, which is surprising given the fact that they seemed to be doing better at the outset.
Dr. Brent James: That's the nature of the medical profession. It has to do with levels of evidence in convincing physicians, independent physicians, that this really is best treatment. But over time the evidence builds and the evidence builds. There are other studies that suggest other treatments and so there's room for argument.
Hedrick Smith: Why are doctors so hard to persuade to change?
Dr. Brent James: Well, now you're going to hear my personal belief on this. But I think that many would agree with it. The fact is that we're just moving into a modern era. Medicine can adequately be described as a craft. The idea is, we call it a physician/patient relationship that I handcraft a solution to your problem. That in some sense makes me what I call radically autonomous. But what really counts is my opinion, my experience about what's best for you. When we started to measure variation from physician to physician to physician, that promise that we'd made that you could expect best possible results because I handcraft your solution, turned out to be demonstrably false. That's when medicine started to make that transition. Which means for a core part of this standardized care, that we come to a standardized approach and we all share it, as physicians. I have to be able to adjust to your particular circumstances. So I can't use it as a cookbook particularly, and I can't mandate it. But I can say, Can we start from a common protocol around which you can freely vary? And that's what this is. That's what produces fifty lives per year saved in pneumonia; coming to a common, best approach.
Hedrick Smith: What should ordinary consumers know about the American Health Care System? What's wrong with the American heath care system today?
Dr. Brent James: The most important thing is what is right about the American health care system. We achieve miracles on a fairly routine basis - even compared to twenty years ago when I graduated from medical school - that said we could be much better. In particular, when patients come into American hospitals, they are at risk for some fairly serious complications that arise directly from the treatment process. We call those patient safety factors and we could do much better on patient safety than we have in the past.
Patient safety means the common complications. Top of the list is adverse drug events, usually a drug overdose. Second on the list is hospital acquired infections. You can go into that place and come out with a really nasty infection that can take your life. Third is an esoteric thing called venous thrombo embolism. Fourth is [tucuberitis] ulcers, bed sores. Fifth is the use of restraints, chemical or physical restraints to hold a patient in bed, usually. Sixth is problems with blood transfusions.
Most Americans understand that their own physician is a good person, but they get caught in the complexity. As a human being, you will make mistakes. Regardless of how careful you are, regardless of how conscientious you are, you will make mistakes.
Hedrick Smith: Do we have a health system in America today?
Dr. Brent James: We have a health system that performs far below its potential capability. It is a system. It is not as good a system as it ought to be.
Hedrick Smith: The implication is there is something sort of large that we are not doing, not just individual performance here. What is going on?
Dr. Brent James: We have a system that is immensely complex. And we expect human beings to be perfect within it. Sorry, human beings just aren't perfect. They will never be perfect. And so you have to build, build parts of your routine process in place to account for, to anticipate and remove the effect of those known human imperfections.
Hedrick Smith: You make it sound chaotic.
Dr. Brent James: At one level the health system is chaotic. It is so difficult to coordinate care from outpatient to inpatient care, across five different specialists for one patient. Across all their nurses and therapists and the pharmacy and the clinical lab. It is so very complex that chaos may be the right term for it at one level.
Hedrick Smith: One of the things that has interested us is to see how very different participants in the health care system are to let quality and performance information out. Health plans, a lot of them don't want to do it except on a very minimal basis. Doctors dislike it terribly. Hospitals dislike it. What is going on? Is there [a] kind of secrecy about quality?
Dr. Brent James: I think the reason that quality doesn't get its due attention in the American health care system is based upon two main things. The first is tradition. Traditionally what I did, by definition, was quality, as a physician. And the fact that you challenge my quality in some sense is challenging my competence, my professional competence, challenging me personally. The second part is that we understand how difficult it is to actually measure it, so it feels very threatening to many physicians and many nurses.
Hedrick Smith: What do we as patients, as consumers actually know? In terms of our knowledge, what other comparison would you make? I mean do we know, do we really know a lot about the quality of the care we are getting?
Dr. Brent James: We do know a lot about the quality of care that we are getting as a country. We have large national studies that show how the population of the United States as a whole does with health care. We have more studies that show how a state does, or even an individual system down to the level of hospitals [and] in some case[s], individual physicians or nurses or clinics. And we know that we do far better than we have done in the past. Yes, health care is a miracle in what it can achieve but we haven't begun to achieve what we could achieve if we were to systematically apply the knowledge routinely available to us today.
When we carefully ask our patient[s] what is important to them when they see a physician or a nurse, what they routinely say is a caring and concerned clinician. Interestingly they don't look directly at medical outcomes. They ask that physician or nurse to tell them what to expect in the context of that individual interaction. That is their expectation. But the fact is that patients look at what we call service quality. At the interaction. At the physician/patient relationship. That is where patients look.
Hedrick Smith: Is it accurate to say that I, as a patient or as a consumer, know more about the performance of my toaster than I do about the performance of my doctor?
Dr. Brent James: That is accurate to say. To the extent that our health care system is functioning correctly, though, there are those who are carefully watching over the performance of that system to make sure that it does well, and saying that we can do much better than we currently do.
Hedrick Smith: Is there a big gap between best practice in American medicine today and average practice? For instance?
Dr. Brent James: Dr. Jack Winberg at Dartmouth University first carefully measured variation in physician practice. He looked at the rates at which people were hospitalized with particular conditions. He showed more than 30-fold variation from community to community in the United States at hospitalization rates or, let's say, for hysterectomy or for congestive heart failure.
So one community would have thirty times more patients hospitalized for congestive heart failure than another, for example. Others followed up on that. Dr. Bob Brook, in Los Angeles at the Rand Corporation, looked at inappropriate care. He demonstrated two things. The first was that inappropriate care does not explain the variation in rates. Far more important, he showed that for some surgical procedures, [for] as much as 40% of all patients who received that surgery, the risk outweighed the benefit. And what it suggests is that no, we are not performing as well as we can as a system.
Even more important [is] the Institute of Medicine Report that suggests 44,000 to 96,000 preventable deaths per year. I think that that represents a fairly significant gap between where we are and where we can be. When we discussed that on the internal committee, we didn't use the figure of 44,000 to 96,000 deaths per year. Most of us seated around the table believed that the real mortality rate was on the order of 120,000 to 180,000 preventable deaths in the United States.
Hedrick Smith: Is there something in the culture of doctors that makes it hard for them to change and adopt new methods?
Dr. Brent James: You know, these days we use standardized approaches to manufacture a car, to build that toaster that you were talking about. That is not true in medicine. I am a skilled craftsman that hand-fits a solution for every patient that I see. I have apprentices learning the art from me as I do it - my residents and house staff. The reason that we never made the transition was complexity. The processes we used were so complex that people couldn't plan them in advance so I was forced to hand fit them and as part of that, it all came back to my judgment.
Based upon my prior experience, I would try to get [the] best care for you personally. The shift has been -- is that we have learned how to deal with complexity, that is, new theory that has occurred in about the last five or ten years inside medicine, where a group of physicians and nurses designs a standardized process. Then can vary around it. Yes, you are a unique individual. No two human beings are quite the same, but I can use a common baseline process across a group of physicians and then customize it to your particular needs. That is what is making this change occur. That is what is suddenly bringing that new level of medicine in reach. That is what we never had before.
Hedrick Smith: Are you suggesting that medicine can be kind of mass produced, the way cars can be mass produced?
Dr. Brent James: I am suggesting that medicine can be mass produced to a degree. The fact is, again, that you are different from any other individual, but it doesn't mean that 80 or 90% of what we do for you shouldn't be the same.
Hedrick Smith: And how do doctors feel about delivering mass produced medicine?
Dr. Brent James: When you approach the physician with this idea of a standardized approach, they feel like they are losing control. They feel like they are losing their professional autonomy. They particularly want to see the data to prove that it actually works before they are willing to sacrifice the main foundation for what they judged to be good care in the past. It is hard to make that shift in thinking.
The doctor was trained that it depended upon their professional judgment and then experienced that through a career of practicing medicine. That is the whole model that they have in their head about how they approach this topic and we are coming to them and saying, well, wait a minute. We really need to get together with a bunch of other doctors, hammer out a common baseline approach. Yes, you still need to be free to vary around that as you customize it for every individual patient. We certainly can't replace your clinical judgment.
But if you can come up to a common baseline approach, it means fewer errors. It means robust computer systems that can support you. It means that the technical staff can be much more efficient in their work in support of you. It means a system that functions much, much better for our patients.
Hedrick Smith: How do we get to a situation where, as you said earlier, that for a hysterectomy or for some other condition in one part of the country you have 30 times more people hospitalized than you do in another? How can anybody be arguing that one or the other of those is the right way to go?
Dr. Brent James: When Jack Winberg published his data showing 30-fold variations for some conditions from community to community across the United States, it proved to us that not all physicians could be right. It proved to us that we needed to come together at a level that we had never attempted before, as a group. It showed to us that this idea of the autonomous physician standing alone against the forces of chaos really wasn't the best model for the current day.
Hedrick Smith: How quick are health plans and providers to make public really significant performance data on health? How well are we informed as Americans?
Dr. Brent James: We generally don't know the outcomes that the health care system produces. It is not just the consumers, though; the physicians don't know their own outcomes. The hospitals don't know their own outcomes. Even when they do measure them, they tend to measure them inaccurately, which is one of the reasons for their fear. Physicians today have the possibility of being able to accurately measure their outcomes. They didn't have that before and that is one of the things that our system really needs to focus on. We need to be able to measure better.
Hedrick Smith: Isn't it something more than that? Part of it is technical capability but aren't you talking about a mind shift?
Dr. Brent James: Yes, we are. In order to make this work, physicians need to make a major shift in how they see themselves. The fact is that in the past I was an autonomous individual, accountable only to God and myself. I would tell you how good I was by my recall of how well I did for my patients. The difference is, today, we are measuring it. And we are discovering that we are not nearly so good as we thought we were relative to the outcomes that we got for our patients. And that, of course, opens doors for major improvements.
Hedrick Smith: Essentially doctors were not comparing their results. They were not sitting down and figuring out what worked best. Is that right?
Dr. Brent James: Doctors didn't compare their results in the way that they should have compared their results. We shared opinion back and forth and we learned from each other. It was collegial; it was helpful. But we didn't share data. We didn't share good outcomes data on comparable patients. We didn't share the processes we used to achieve those results at the same level.
In one sense medicine still is in the horse and buggy era. We have got new techniques, new approaches that will allow us to move beyond the craft [of] medicine that describes my education and of all of those medical progenitors that came before me. Doctors need to compare those data and identify best processes. If we are able to do that the evidence suggest that we will be able to achieve levels of medical outcomes for our patients that far surpass anything available to us today.
Hedrick Smith: How important are information systems to this whole new approach?
Dr. Brent James: Information systems are critical. Medicine is inherently an information science. In general, the better information I have, the better diagnosis I can make, the better treatments I can offer, the better treatments I can deliver and the better outcomes I can achieve. The main information systems that are critical to me are outcome systems. Outcomes information is important not just for the physician at the front line, it is also essential to organize the health system as a whole. So think of information for patient care on the one hand, and on the other hand, information for management. It turns out it is the same information set, properly organized. On the other hand we can use those data to plan how we structure [the doctors] clinic and how we train his nurses.
Hedrick Smith: Are you saying that to improve health quality we need to have data driven medicine?
Dr. Brent James: It is impossible to improve without data. You manage what you measure. If you don't measure it, you can't manage it. Data are absolutely essential. The life blood, the nervous system of how we deliver care, both for an individual physician, for a hospital or a clinic, and up to the entire system, within the country, it all turns on data. Good data. Accurate data. Timely data that is in front of you when you need it, properly formatted so you can tell exactly what you need to do.
A classic example: I trained in general surgery and used to run into a very difficult problem. I'd be seeing my patients at the hospital and discover that one of them had developed an infection. Well as I began to prepare to treat that infection, I was theoretically supposed to know something about a very difficult topic. Even for a physician. It is called the epidemiology of infection. It is a specialty. It is that complex. And as a poor, dumb surgeon, I didn't know it as well as the infectious disease people did. I really didn't have an opportunity, usually, to consult with them, though. I had to act too fast. It would take too long. It would cost too much.
Today in one of our hospitals what I could do is call up the computer program on a computer terminal next to the patient's bed. It does that infectious disease consult for me in two or three seconds. What it gives me on the screen is a list of the most likely causes of bacteria that could be causing your infection, down the screen in probability order, customized for you personally. With each one of them it lists the antibiotics that are best to treat that particular infectious agent.
If I page ahead one screen, it shows them as entire antibiotic regimens. And I can look at one of those and say, yes, that is the right one in my experience. I want number two on the list. Hit a return. That computer will place the order with the pharmacy to get the medication up to the floor to start your treatment. It will automatically alert the nurses to start the IV, usually, to start the antibiotic. It will automatically order any necessary testing. And just the ability to have that list in front of me, you see, that summarization of data, customized to our circumstances, - the physician and the patient at that particular locale [is] massively helpful. That is data at work.
Now take those same data and start to track infection rates across all of IHC, across all twenty-three of our hospitals, across a hundred plus clinics, almost two thousand physician practices. I can use those data to determine where I put in systems, how I modify those systems, how I direct patients in some circumstances to try to optimize the care experience for everybody for whom we are responsible across the entire system.
Hedrick Smith: Do you know how much of doctors' performance is based on their opinion as opposed to data?
Dr. Brent James: We know that about 20% or less of what we do in routine medical practice has a foundation in published scientific research. About eighty to ninety percent of what happens in routine clinical practice is based upon congealed experience. For example, if you read most major medical textbooks, you'd read not scientific research, [but] the opinions of that senior professor who wrote the textbook.
Hedrick Smith: Might be very knowledgeable opinions but opinions nonetheless?
Dr. Brent James: Those professors who wrote the text books would be very, very knowledgeable, very experienced, caring. The fact is, though, that they're often wrong, as we discover when we finally generate the data. That's one of the big challenges we face as a profession. Physicians can legitimately hold differences of opinion about what's best because we dont have good science. Even when we do have good science the research shows there's a lag of as long as 15 years in getting well-established truths from scientific research in true team practice across the country. Although some communities will adopt it fast, some communities are much, much slower in adopting those best practices.
Is it any surprise that there's massive variation in how we deliver care to patients across the United States? No. Is there any surprise that there is massive variation in the measured outcomes of the care we deliver? Not particularly. Is it a fault of the physicians and nurses? No. It's the fault of the system. We need to sit down and figure out a system that deals with these realities.
Hedrick Smith: How successful have you been in spreading the examples that you have derived, the protocols that you have developed within your own system?
Dr. Brent James: Within Intermountain Health Care I regard myself as a poor, dumb researcher, basically, sharing the good results of others and trying to facilitate those results. The fact is, it's worked. The fact is I've seen improvements in care in which I've played a role - some small role that went far beyond what I ever would have been able to accomplish in a lifetime of practicing medicine. It's simply the funnest, most worthwhile thing in which I've ever participated.
Hedrick Smith: How does it make you feel when you see the results?
Dr. Brent James: When I see those results, it makes me feel like a real physician. All the reasons that you came into this profession, believing that you could really do some good, believing that you could really help some people.
Now I've taken a step back. I spent the early part of my life doing that one on one with individual patients and got interested in the research side. But I've had far more impact on a far broader scale. I've seen it help my own family members in terms of the level of care that they received. The real question is how good can we be for our patients. That's a challenge.
|Home | Helping You | Information and Resources | Program Content |
About Us | Join The Discussion