A Conversation with Brent C. James, MD
Pay for performance is not enough. Fundamental changes in the fundamental relationship have to be taken into account.
Brent C. James, MD, is chief quality officer and executive director of the Institute for Health Care Delivery Research at Intermountain Healthcare, an integrated health system in Salt Lake City. He is a member of the National Academy of Science’s Institute of Medicine, and has participated in many of that organization’s seminal works on quality and patient safety.
Known internationally for his work to improve clinical quality, patient safety, and the infrastructure that underlies successful improvement efforts (such as culture change, data systems, payment methods, and management roles), James holds faculty appointments in the University of Utah School of Medicine’s Family Medicine and Biomedical Informatics department, the Harvard School of Public Health’s Health Policy and Management department, and the University of Sydney, Australia, School of Public Health.
Intermountain Healthcare is a system of 23 hospitals, almost 100 clinics, more than 550 physicians, and an HMO/PPO insurance plan jointly responsible for more than 500,000 covered lives, serving patients in Utah, Idaho, and adjacent states.
Through the Intermountain Advanced Training Program in Clinical Practice Improvement (ATP), James has trained more than 2,200 senior physician, nursing, and administrative executives, drawn from around the world, in clinical management methods, with proven results. His designs have resulted in nearly 30 quality-training programs now running in six countries.
James holds bachelor of science degrees in computer science and medical biology, an MD with residency training in general surgery and oncology, and a master of statistics degree. He is a trustee of several not-for-profit organizations dedicated to clinical improvement.
James was also featured in a recent New York Times Magazine article, “Making Health Care Better,” which can be viewed here: http://bit.ly/4wrLpN.
We asked him for his perspective on accountability and incentive programs, how the measurement of quality has evolved in recent years, and the potential value of electronic medical records.
MANAGED CARE: Current discussions about health care are focusing on insurance reform. But the single most important issue — and the one to which you’ve dedicated your career — is the relationship between physician and patient. How has that been evolving, and where are we headed?
JAMES: The first thing we have to remember is that there is a sea change happening in the medical field. It’s the first time in a hundred years we’ve seen a thing like this, not only a shift in how we think about being physicians but in how we practice — what it means to be a physician. It has to do with how the current system falls short of theoretical potential. We do have to be pretty careful about these terms. It’s theoretical potential — that is, nobody’s really got it all figured out yet. So when you examine what is happening in the practice of medicine today, you discover five major areas of concern:
First, there is massive variation in the quality of care; that is, we want to try to make sure that everyone is getting good care.
Second, we want to make certain that we as professionals are able to assess the appropriateness of care, because we have appallingly high rates of inappropriateness of care, which in itself is professionally challenging.
Third, we have high rates of injury or death.
Fourth, we suffer from a truly striking inability to demonstrate that we are doing things that actually work. That is not controversial: Everyone agrees that there are certain things that ought to be done but that we are failing to execute.
And finally, fifth, there is very good evidence that half of all health care dollars are wasted.
MC: Pay for performance has been described as a panacea for our fragmented health system. How can we avoid pay-for-performance initiatives that result in physicians and other providers simply gaming the system — providing results that meet the data expectations of payers without actually improving the quality of individual care?
JAMES: Well, these data are emerging and people are beginning to try to find out why things are happening the way they are. The primary cause is the head-on collision of two factors.
The first is the way we think about practicing the whole class of medicine, the idea that physicians are independent experts who stand up on their own feet and say “in my experience” and have that resolve all decision-making. This has worked for a very long time, physicians thinking that way.
The second factor is that, just in the course of my experience, there has been an exponential increase in the amount of information and data available to individual physicians. The practice of medicine has become so massively complex as to accelerate clinical uncertainty to the degree that there has been a head-on collision in practice styles.
If we are all on our own, what it does is create a massive amount of treatment variation, which has created a collision in practice styles. The evidence of this is so overwhelming that the profession has started to shift to a form of practice where more and more decisions are being placed with a professional foundation with some form of coordinator. This involves several problems.
For example, CMS now provides data on a huge number of antibiotics and their effectiveness. It is impossible to be aware of all the data in the time necessary to make a diagnosis. So you can get good care but you can’t get top quality care based on all the available data.
MC: You are describing a very specific dynamic of what is basically too much available data for a person to grasp. How do you adjust for that dynamic? Are there existing incentives to avoid the misuse of information?
JAMES: No, there are none. What we have right now that works directly against good care is how you set up the delivery of care. Within Intermountain we have done a lot of original work on this. We emphasize evidence-based care, but individuals come to us with different anatomy, with different expressions of disease, with unexpected response to treatment, and with personal preferences.
So when we put together an evidence-based protocol, it’s not just that we allow or even encourage specificity in following specific evidence-based guidelines. It’s that we demand that our physicians rationalize their treatment decisions in a very detailed way.
We frequently get cases that fall outside of the parameters outlined in the CMS pneumonia protocols, for example. Because that is the case, we find that the guidelines can breed a high level of cynicism among the medical staff. They live this experience every day.
But the people back in Washington, D.C., don’t see that detail, so you start to get reports that this is causing actively bad care. The simple truth has always been that the definition of insanity is doing the same thing again and again and expecting different results. In my career this is the third time we have been through this kind of quality emphasis.
The first time was a quality data reporting initiative that led to a controversial public report by CMS, then known as the Health Care Financing Administration, in 1987. The idea was that if people stayed away from the bad hospitals, not as many people would die.
But subsequently we found that several years after the HCFA data was published, hospitals experienced only a 1 percent fall in patient population for the procedures they performed badly.
The second time was the Joint Commission’s Oryx initiative, an ongoing effort to tie outcomes to accreditation that started more than 10 years ago.
Such programs hold promise for improving the quality of care when physicians are rewarded appropriately to do what’s right. It is a very important concept, but this isn’t the first time we’ve tried. It’s the third. We need to learn from the past and be careful.
MC: What are the lessons learned from the past and what are their implications for this third initiative?
JAMES: In essence, it comes down to this: The closer you are to the front lines, the better you can deal with improving quality. All our protocols are very detailed. But when I start to see 100 percent compliance, I know we have a problem. So I don’t ever set that expectation.
If I do, then it is used as a tool to examine variance and I carefully examine outliers, which creates an opportunity in itself. When we start running our reports, there are always significant outliers, and that provides a quality improvement opportunity.
When we get outliers, it provides precise information, and it can be any outlier, a hospital or a physician. What we have found is that about half the time in this very carefully developed system the cause was in the measurement system, not in the care, not in things we had anticipated.
What you do with this is, you clean up your measurement system. What you are saying to physicians is that you’re not playing “gotcha,” you’re saying “we’re in this together, trying to find clinical truth together.” We all have the same expectation of care and we’re honest about it.
It is complex and it is hard to figure out how all it all fits together. The first thing is to step back and look at an entire system of care, using data systems. Then what becomes clear is that if you improve things, you reduce the cost of care.
MC: So what you are describing is bottom-up accountability, adjusting measurement systems based on the evidence, the actual evidence experienced by physician?
JAMES: That’s right. This is complex enough to measure quality accurately. So let me talk about how I would do it if I were setting up my own system.
The first thing I would do is build up my measurement system so that I would be setting goals for performance that wouldn’t be geared toward failure, but toward success, with lots of slack for reality.
I would do it so that it depends on the process 85 percent of the time and not 100 percent. The reason is that I know that the reality is that reaching 100 percent will almost certainly produce some degree of bad care, some portion of the care that is dangerous to the patient.
MC: Why is that?
JAMES: It comes back to doctors gaming the system. When you create an incentive around an achievement of 100 percent, the real danger will be in gaming the data system. For example, if you create a system whereby the idea is for a nurse or doctor to check off on a sheet that a patient had received instruction in how and when to take his or her medications, then people will always check off those boxes.
But we know that 100 percent is not realistic. They were following the letter of the law but not the spirit of the law. What we need is a cultural shift, an educational initiative that encourages nurses and doctors to take the time to explain to patients the importance of the medications.
That is best achieved when a medical practice has a financial interest in outcomes. When outcomes evidence demonstrates that there has been an improvement in compliance, then physicians should be rewarded. They should be allowed to share in the reduced overall costs that result from improved compliance.
We need to learn to measure our outcomes precisely enough to be able to reward practices when their performance improves outcomes.
MC: We are still left with the issue of encouraging evidence-based medicine. Incentive distribution occurs at the group level, but how do you encourage the use of evidence-based medicine on the individual physician level? Isn’t that the bottom line?
JAMES: That happens, really, on two levels. The profession is shifting in the direction of evidence-based utilization. Best practices produce the best outcomes, and individual physicians are aware of this.
And physicians are highly motivated to achieve the best outcomes. The second level is to clearly demonstrate to physicians in a practice that producing the best outcomes also produces significantly lower costs of operation.
Intermountain does that. It applies evidence-based guidelines and then it clearly demonstrates to payers that it is those guidelines that are producing improved outcomes and at the same time creating savings.
But the system works best like this: We apply the evidence-based best practices and get better outcomes and lower costs and then we get to keep the savings. That’s where the true incentive comes in. This is a form of pay-for-performance that is a subset of the way P4P has been functioning within our system. It is the system I have been advocating.
The basic idea is that when you find better care and clearly demonstrate that care, the savings that accrue from these better outcomes should be shared with the caregiver. A portion should go to the caregiver and a portion should go to the purchaser.
Think of it this way. I practice surgical oncology. We deal with life and death, and there’s a profound psychological need to believe you’re helping others. It hurts if you believe you’re harming someone. You don’t have to wait for the data to come get you.
It’s a heavy load to bear. You have to believe that you’re doing good. You already have an environment wherein the caregivers are heavily socialized to believe in each other and to watch each other.
Now add the element of shared financial incentives for the implementation of evidence-based care. The result can only be a system that rewards compliance with the best possible medicine.
Right now we are spending $2.4 trillion on health care. We need to understand that if we don’t make fundamental changes in the way we incentivize the practice of medicine at its most basic level — in the patient-doctor relationship — no amount of top-to-bottom system of evidence-based guidance will produce lasting and truly meaningful quality improvement.
MC: Thank you for speaking with us.
“There is very good evidence that half of all health care dollars are wasted.”
“We apply the evidence-based best practices and get better outcomes and lower costs and then we get to keep the savings. That is where the true incentive comes in.”