Robert Kisabeth, M.D., Medical Director

Diagnostic Insight, 1997 Interview

Robert Kisabeth, M.D., took an interesting route from Newport, Tennessee where he grew up to the position he now holds as Medical Director of the Mayo Medical Laboratories at the Mayo Clinic in Rochester, Minnesota. While his interest in mathematics could have led him in a vastly different direction, he is extremely satisfied with the path he chose. "I was a physics major at Union College in Schenectady, New York, and at the end of my sophomore year I decided that although mathematics was fascinating, it was not as fascinating as people were," he explained. "I was trying to imagine how I was going to earn a living as a physicist, so I decided to carry my interest in science into medicine." He attended medical school at the University of Tennessee in Memphis from 1968-1971. Working his way through medical school as a blood collector and later as a bench tech at a laboratory in Methodist Hospitals of Memphis gave him a view of what the laboratory could mean to the care of people with many different types of diseases. "The question was whether to become a general internist, a general practitioner, or a pathologist, and since I had a privileged view of lab medicine, I decided to go into pathology and laboratory medicine," Kisabeth said. After completing his residency, he worked in a large general pathology practice – the Duckworth Pathology Group – until 1989. At that time, despite some trepidation about moving to "the tundra", Kisabeth and his wife moved to Minnesota where he began a position at the Mayo Clinic as a pathologist. Two years later, he was named Medical Director of Mayo Medical Laboratories. "We've been absolutely delighted from the day we arrived – we couldn't ask for a finer life," Kisabeth said of his move north. "It's a wonderful group of people with whom to practice medicine – it has been a totally positive experience. " Kisabeth and his wife have two daughters, aged 14 and 16. In addition to his duties at the laboratory, Kisabeth also serves as Chairman of the Mayo Clinic's European Task Force, which evaluates opportunities for Mayo to constructively serve the interests of European communities in aiding sick people. While his work involves "an extraordinary amount of travel, "which limits his free time, Kisabeth enjoys building furniture and cabinetry when he has a few spare moments. Regarding his hectic work and travel schedule, he said, "My vocation has turned out to be my avocation. But I have no complaints; I don't work for a living -- I just practice laboratory medicine." In the following interview, Kisabeth shares his insight about changes in the delivery of health care, current laboratory practices, reimbursement concerns, and other issues involving diagnostic testing.

ROBERT BAUER: In your 20 years as a physician and most recently as medical director of the Mayo Medical Laboratories, what do you feel have been the most important changes in the delivery of health care?

KISABETH: Most physicians my age have been part of a health care system that has been inexorably and continuously commoditized over the years. The commoditization can actually be traced back about 25 years when insurers, HCFA, and the American Medical Association set out to design an "improved" means for medical billing. As you know, they developed the RBRVS (CPT codes), which codifies all tangible procedures of health care. The problem with this approach is that it is based on relative cost, not on relative value. Why? Because procedural costs are easier to define and tangible costs are easy to characterize. The problem is, however, that this system places little value on the important and valuable intangible aspects of medicine, such as researching patient histories, physical patient examinations, and knowledge of the patient and the patient's environment, etc. The best-reimbursed things in health care are procedures. As a result, the entire health care system shifted its focus to tangible procedures. Medical specialists' revenue in the last 20 years has been associated with procedures. Medical students have gravitated toward procedure-intensive specialties. The laboratory industry has also focused for the last 20 years on revenue-associated procedures–not on the contribution of laboratory tests to the improved management of patients' illnesses.

The pitfalls of managed care and capitated reimbursement

ROBERT BAUER: Do you see managed care and capitated reimbursement as improving or exacerbating this situation?

KISABETH: Theoretically, there are good things associated with capitation. . .that expertise in health care would focus on preventing illnesses and thus costs, thereby reducing waste but we have not seen what we might have imagined would occur. There is a conflict of interest in the system today. When you add profit initiatives to capitation, you end up with a provider of services looking at short-term objectives– quarterly or annually. Chronic heart problems and chronic inflammatory diseases, like rheumatoid arthritis, really have to be looked at in the longer term to appropriately define the financial context in which to apply controls. But as providers bid on capitated contracts, their intent as it relates to the treatment of hypertension might be, for example, to minimize the number of visits and the overall cost of caring for the patient over a one to two year period.

ROBERT BAUER: It is a common perception that this conflict of interest can be mediated by establishing a metric to compare the outcomes of different providers. What is your perspective on this?

Kisabeth: It is true that some have suggested that this conflict of interest can be countered by the creation of outcomes measurement that would in some way be published, making clear to the purchaser, patient, and employees which plan works best. The problem is that one cannot get a realistic view of one organization vs. another in treating the patient unless one looks at the treatment of patients over a 20-year period. For example, in hypertension, does the provider that enforces compliance of therapeutics lessen the incidence of heart failure over one that does not enforce compliance? You can point out all the normal blood pressures you want, but to truly determine if one provider is outperforming an other will require a study period long enough to measure the impact on heart failure. I picked this example because it is very real; the 3.5% of patients with high blood pressure consume about 40% of the hypertension-related expenditures. Population-based reimbursement is simply another way to focus on costs, but it has a very serious and unintended consequence: providers have an incentive to keep sick people out of the system.

ROBERT BAUER: If the current approach is flawed, what do you suggest?

KISABETH: In this country, unfortunately, we use specific diagnoses, meaning ICD9 codes, to categorize our patients. And, let's face it, in many cases, we use diagnostic tests not to evaluate myocardial infarction but to confirm a presentation of it for reimbursement purposes. This is clearly the wrong orientation. It would be more effective if payers paid per illness or per episode of care. An illness basis would work best for long-term care such as diabetes while episodes of care would be best for acute illness such as chest pain. This approach would allow the organizations that provide care to better focus on applying resources to the patient. Some countries use Reed codes, which are not diagnoses, but define a patient's problem by complaint. Patients would rather pay based on how they present to the doctor–not on their diagnosis. Being paid for the diagnosis makes no sense at all. We're not caring for a diagnosis; we're caring for a patient who presented with a specific set of problems. An example of this problem is with laboratory tests that are paid only when certain diagnoses are present. At Mayo, we see many patients who do not have a diagnosis, yet they often require an extraordinary amount of time and expertise. In a CPT-based system, time and expertise have little value. The incentives are all for a quick, even premature, diagnosis.

Establish clinical and analytical objectives first – then minimize the cost of the procedure

ROBERT BAUER: Our observation is that many laboratorians really aren't active in the clinical process. They are more focused on the operations of the laboratory as a plant. Why is this?

KISABETH: I think this is a consequence of the commoditization of health care. As administrators look for ways to save money in health care, laboratorians have responded by reducing the cost in the lab or reducing the number of labs–again, thinking that the task was to minimize the cost of the procedure. The real issue is how can the laboratory be used more effectively in reducing the overall cost of care? The point here is not to look at the laboratory as an isolated venue for improvement, but to look at the medical process or, more specifically, the disorganization of the process. The big gains are in overcoming entropy and organizing the process. Unfortunately, many of the incentives established by insurers actually inhibit this approach. The improvements will come from integrating the care process, not propagating artificial boundaries. We've prided ourselves on minimizing the cost of analysis rather than on optimizing the cost of analysis. Minimizing has become the watch word of laboratorians. At Mayo, we prefer the term optimizing–first establishing very carefully the clinical and analytical objectives, and then minimizing cost within that context. That context might also include, for instance, frequency of delivery. What we have done for so many years is maximize batch size. It's ludicrous really. In an overly simplistic search for economies of scale, many laboratorians have minimized the cost of the procedure without considering the context. Batches are now beginning to dissolve at Mayo and almost all tests are done on an ASAP basis.

ROBERT BAUER: With the commonly espoused "hub and spoke" concept, many believe that laboratory tests that are not STAT should be done in remote centers for greater economies of scale. How does this fit with your philosophy of optimizing test cost?

KISABETH: That to me is an arcane notion. If one looks at the case mix at any hospital in this country, one finds that today's inpatients have increased substantially in severity and complexity, and the kind of laboratory support needed is extraordinary compared to what it was 20 years ago. Our task for patients in the hospital is not only to maintain them until we can move them to an unoccupied bed but to diagnose them and treat them as quickly as possible. I think that the notion of having only STAT tests within the hospital is an ill-conceived notion that reflects a lack of understanding and familiarity with real group medical practice–and this decreasing familiarity is largely a product of having practiced in a boutique environment.

ROBERT BAUER: Considering where insurance is today–managed care, integrated health care delivery systems–what are the implications for the future?

Kisabeth: I think patients and purchasers will ultimately decide whether our notions make any abiding sense. When we survey patients, their desire more than anything else is that we, in the health care system, communicate better. One of the most consistent complaints patients have is that, as they move from one specialist to another and from one care setting to another, there is virtually no communication. In response to this patient mandate, we will see the integration of the provision of medical services. And the best place for integration to occur, because it requires the ongoing communication and working together of health care professionals of all kinds, is within the community. Care plans, decisions regarding formularies, and the diagnostic test and therapeutic modalities that work and do not work for patients will be decided on a community basis. This is in the best interest of quality care, and the best interest of cost management. Another reason the community is important is that it's difficult to deliver care without first caring. This may seem corny on the surface, but it is very true. And it's probable that such caring can occur only at a community level. So if we can move technology as near as possible to a person's hometown, not only will the care be less expensive, but it could be perceived by patients as far more acceptable because it will be delivered in a traditional and caring manner, by people who actually know the patient– strange as that may seem in today's environment. Along with this, we will see a reappraisal of medical technology with regard to whether it really benefits patient care. I think no segment of medicine will face more of a reassessment than the laboratory/diagnostics industry. One of the greatest opportunities for laboratory medicine to pursue in the future is a larger role in the treatment of sickness. Wouldn't it be a tremendous cost and quality benefit if we could tell patients more consistently, through good laboratory testing, whether their treatment had been effective, and whether that treatment–which is probably very expensive–could indeed be stopped?

I think another opportunity for the diagnostics industry will lie in ascertaining whether a metabolic pathway requires any therapeutic intervention.

Diagnostic products should be sold as sources of a reduction in overall cost

ROBERT BAUER: Many insurers do not reimburse a new and experimental technology without years of established use or extensive data, and diagnostic products generally do not generate the financial returns to substantiate such a clinical analysis. What does this mean for the future of laboratory technology?

KISABETH: Diagnostic products don't earn that return because they've been sold as sources of incremental revenue instead of sources of a reduction in overall cost. I think absolutely that payers will pay for it, but we're going to have to show them that it's more valuable to do our tests and reduce downstream costs and enhance quality of care. The only way we're going to do that is to demonstrate validity, and if it takes years to do that, that's what we're going to have to do. But in most cases it doesn't take years– it just takes effort.

ROBERT BAUER: What are some diagnostic tests that you consider very valuable?

KISABETH: Most of our laboratory tests are very valuable, when properly used. I'm not one to decrease testing–I don't think that's the key to reducing overall health care costs–but I do think there are tests that we should not be doing, that we could better spend that money on other tests, procedures, drugs, or health care services. Today, many tests that we perform will not improve patient care if not adequately controlled. These tests, in fact, can have serious potential for causing harm. For example, within the laboratory community we have normative behavior standards of quality control. These standards, however, are based on the operations of the test procedures, and not on their downstream clinical impact. Let's take PSA, for example. If the mean for this assay on a given day, on a Levy Jennings plot, were to move 0.7 standard deviation, many would say this falls within acceptable standards. We often don't consider that such a deviation–in their community, for example–would result in more than nine men being selected for biopsy and ultra sound rather than 2.5 who should have been selected. So we must seriously question the basis on which we set our standards. Are our standards in the context of our operations, or in the context of the patient's care and the overall cost of that care?

Interview conducted by Robert Bauer, president of CaseBauer & Associates, a business development and research firm operating in 14 countries.

CaseBauer has 5 practices:
Market Modeling, Strategy / Business Development, Benchmarking, Market Research and Business Services.