Curtis Hanson, M.D., is a professor of Laboratory Medicine and Pathology at the Mayo Clinic College of Medicine, and a consultant in the Division of Hematopathology at Mayo Clinic in Rochester, Minnesota.
Contact us: MMLHotTopics@mayo.edu.
Curtis Hanson, M.D., presents the first in our series "Laboratory Test Utilization Strategies." The series addresses strategies for test utilization in the clinical laboratory. Dr. Hanson will discuss the need for appropriate laboratory test utilization in today’s clinical practice and offer management tools for constructing and implementing a utilization program.
Our speaker for this program is Dr. Curt Hanson, Professor of Laboratory Medicine and Pathology, Mayo Clinic College of Medicine, and Consultant in the Division of Hematopathology, at Mayo Clinic in Rochester, Minnesota. Dr. Hanson presents an update on test utilization strategies in the clinical laboratory.Welcome to Mayo Medical Laboratories Hot Topics. These presentations provide short discussion of current topics and may be helpful to you in your practice.
Thank you, Heidi, for that introduction. As Heidi mentioned, my talk today is entitled “Laboratory Test Utilization Strategies”. This is the first of a 3-part series regarding test utilization and the role of pathology and the clinical laboratories. My colleagues, Dr. Amy Saenger and Dr. Bobbi Pritt, will be presenting parts 2 and 3 at a later time.
I have no conflicts or financial disclosures to make.
Laboratory test utilization can be defined as a strategy for performing appropriate laboratory and pathology testing with the goal of providing high-quality, cost-effective patient care. I’ve highlighted some of the key words to keep in mind as we address this topic. It is important to determine whether the motives behind a utilization strategy are financially or medically driven. If the focus is solely on money, then that test utilization effort will have, at best, mediocre success. However if good medical practice is the focus, then test utilization will have longevity and will likely be successful in our changing health care environment.
Health care costs in the United States are significant and are escalating. It is estimated that laboratory testing accounts for 60 to 70 billion dollars per year, which is about 4% of the overall health care spend in this country. Laboratory costs, however, are increasing at an annual rate of 20 to 25% with much of that increase due to significant growth in the arena of molecular testing and genetic studies. Some experts predict that the cost of molecular and genetic studies may soon account for 25% of total laboratory costs.
At Mayo Clinic, the value equation has always been defined across our practice as outcomes plus service plus safety divided by the cost over time. How does that apply to a clinical laboratory test utilization strategy? I think we in the laboratory can equate this to the right test being done on the right patient at the right time and done right while being good stewards of laboratory resources. In today’s environment, not only is this applicable to test utilization but this is really the laboratory’s formula for being a partner within an ACO-type reimbursement environment.
So let’s ask the question, “Do we have laboratory utilization issues?”
As an example, I want to share with you an anecdotal case that I recently saw as an hematopathologist here at Mayo Clinic. Patients frequently come to Mayo for evaluation and bring their slides and their outside workup with them. An evaluation of the bone marrow slides from this patient showed a straight forward diagnosis of primary myelofibrosis, which is a type of a myeloproliferative neoplasm. These are all the various studies that this patient brought with him from that particular bone marrow study, but once you look at the bone marrow morphology, only these 4 studies were necessary in providing the appropriate workup for that patient. All the remaining studies are unnecessary and clearly added a significant financial burden to the health care system in this case.
So why do we have utilization issues? I will talk in more detail about the first 2 items in subsequent slides. However, other factors also contribute to the inappropriate use of laboratory testing. Clearly, our laboratory systems and processes are geared to providing every test as fast as somebody wants it without asking questions as to whether it’s appropriate or not. There is also the tendency to overbundle test panels and include assays that are not necessarily needed for that patient at that given time. There are also wrongly aligned incentives in our current fee-for-service reimbursement environment. The change to value-based payments or bundled payments will clearly have an effect on utilization going forward. Finally, we can’t forget the effect that societal and patient demands have on the system, especially as patients become more informed about their health care via the internet.
The realities of today’s clinical practices are the major factor in driving laboratory utilization issues. There are clearly various levels of understanding by clinicians on how to use today’s laboratory assays. And the growth of these complicated and complex assays will continue to put further strain on what our clinicians need to know. It’s also important to understand that patients with particular diseases may not see that disease subspecialist or may see a specialist with only limited experience with that particular disease process. More importantly, laboratories must understand that clinical knowledge at the time of test ordering is incomplete. Because of that, the clinician, who may be considering multiple diagnostic possibilities, is compelled to order everything as it may be the only chance to get that information. It’s also important to know that initial laboratory and pathology studies can help narrow the diagnostic choices in testing needs. But, if laboratories don’t have a review process in place to help narrow down those testing choices, clinicians have no choice but to order excess testing.
I think we all acknowledge that there is a knowledge gap between what clinicians know and the depth of knowledge that is necessary in ordering laboratory testing. There’s also a growing gap between what we know scientifically versus what actually can happen at the therapeutic level. Importantly, laboratories usually do not provide guidance for the appropriate use of assays and, indeed, commonly just list the various tests that are available that a clinician must choose from. As we add new tests to our menus, there is typically not an effort to understand how these tests should be utilized in the context of other existing assays.
So what do we do next?
It is critical to understand what I call the test life cycle – in which all actions or steps must be centered on the patient. When we look at this drawing, it’s important to understand that the test life cycle begins at the point where the doctor requests the test, all the way through the laboratory, and finally to the point when the doctor actually acts on the results. Frequently in the laboratory, we only think about those steps that are located within the laboratory walls and we ignore those things that occur at the level of the clinician. However, if we’re going to have an impact, we must understand the questions at all points around that patient. We need to make sure we understand what the clinical need is. Does the laboratory stop and ask why is that test being done all the while realizing that our bosses are asking what’s our margin or that we’re an ACO and how are we going to participate? It’s important that our reports are clear as we send back results to our clinical colleagues. Is the pathologist involved in making sure that the medical side is being appropriately represented within the laboratory environment? And at the end point, does the physician get the right information and, importantly, does the doctor know what to do with the information that we have provided?
There are a variety of test utilization management tools that can be used in a utilization program. Providing clinician education is the necessary first step, but it’s important to realize that it probably has little lasting impact in and of itself to change physician behavior. Clearly as a laboratory, we need to get in the habit of obsoleting certain tests that no longer have clinical value. As examples, bleeding times, band counts, and most erythrocyte sedimentation rates are all tests that have probably outlived most of their utility. I think it’s important that laboratories and pathologists provide some type of a gatekeeper role and identify tests that require laboratory intervention and review. It is clearly appropriate that we identify those tests that are ordered too frequently, especially for the hospitalized patient. Clearly, in the hospital environment when multiple physicians are seeing an individual patient, you run the risk of the same tests being ordered by multiple physicians or tests that really should be ordered only once during hospitalization being ordered more frequently. It’s also very important that pathologists and laboratories insert themselves into the creation and review of admission and treatment templates and look for redundancies in testing and the frequency in which tests get performed.
Our IT systems can be a very important tool in test utilization management. Order entry pop-ups and online decision support tools are important in providing immediate guidance to the ordering physician. We know from our colleagues in Transfusion Medicine that providing profiling or report cards as feedback on blood product utilization is an important tool that can also be applied in a more general test utilization management strategy. And clearly, establishing a utilization review process for sendout tests is important, especially as those expenses continue to grow, and complex molecular and genetic tests get requested. Some practices have restricted ordering of expensive sendout tests to certain specialists or have established medical criteria necessary for ordering particular tests. It is certainly important that a laboratory review process be inserted for certain tests to validate their medical necessity and to ensure that proper payment for these tests will occur.
Creating laboratory guidelines and algorithms will be the focus of the remaining part of this presentation and will be discussed further in the subsequent two presentations by my colleagues, Dr. Saenger and Dr. Pritt. It’s important to realize that there are really 4 ways to develop and use testing algorithms. First are those that are IT-driven where clinical input and information are required at the time of test ordering to drive appropriate testing. Second are the laboratory-driven algorithms, ie, where clinicians order a testing cascade and initial laboratory results drive subsequent test selection. Thus, the laboratory owns the entire cascade or the algorithm being performed. Finally, there are those algorithms that are generally lower volume but high cost where individual health care professionals provide their input into test selection. It may be pathologist-driven where the review of pathology findings determines the next steps in testing, or it may be a genetic counselor or other laboratory staff that review test requests and use their expertise and laboratory knowledge to drive the right testing strategy.
Let’s look at an example of a laboratory-driven process. At Mayo Clinic we use a triage approach in the flow cytometric evaluation of hematologic disease where we use a minimal number of antibodies to screen for B-cell clonality, an increase in blasts, and/or an abnormal T-cell phenotype. The results from this triage, together with the clinical history provided and morphologic appearance of the slide, are used to determine whether or not to stop the analysis or to add on more disease-specific antibodies. As a result of this approach, approximately 80% of our Mayo Medical Laboratories’ requests actually stop at this triage step. If we compare this approach to what many laboratories use, which is a shotgun approach with 20 to even over 30 antibodies in any given case, you can clearly see the advantage of using a step‑wise or triage-type process.
This slide was shared with me by Dr. Mike Astion and colleagues from the University of Washington and Seattle Children’s Hospital and data that he acquired in working with CareCore National. This slide shows how different clinical laboratories bill for the CPT code 88185, which corresponds to the number of antibodies used in a flow cytometry study. You can clearly see there is a huge range in the number of antibodies that are used by flow cytometry laboratories. The cost drivers are obviously at the right-hand side of the graph and are the source of what really adds to the overall cost of flow cytometry in this country. You can also see where our practice at Mayo falls in relation to others. We have been using this approach for approximately 17 years now, and clearly it has worked well within our large and varied clinical practice. I am certain it can work in other practices if used.
We have had a long history of utilization control for flow cytometry and immunohistochemical and cytochemical stains in our hematopathology practice here at Mayo. Three years ago we observed that we were seeing many unnecessary requests for cytogenetic, FISH, and molecular assays. So, as part of a utilization review process, we gathered data from our practice, analyzed utilization patterns, and compared the results to what we knew medically and what was in the literature. From there, we worked with our clinical colleagues in developing guidelines and algorithms toward the appropriate use of complex laboratory testing in the context of hematologic disease and have since successfully implemented many of these into our clinical and hematopathology practices.
Here is an example of 1 of the many algorithms that we have instituted. This, in particular, relates to the bone marrow evaluation of a possible myeloproliferative neoplasm. In this pathology-driven algorithm, the morphologic features help determine what the subsequent, most cost-effective laboratory testing should be. We have found this approach to be very satisfying professionally for both pathologists as well as our clinical colleagues, and yet has been very effective in controlling for appropriate use of expensive and sophisticated laboratory testing.
So what should we do when thinking about beginning a laboratory utilization program? I think it’s important that you first ask yourself some critical questions. Do you believe that you have a utilization problem? If you can’t answer this basic core question, you can almost guarantee that your program won’t get off the ground. Do you have an organization that wants to respond? If you are the only individual that’s interested in a utilization program, again you can guarantee that it will be a tough row to hoe. Next, do you have a laboratory practice committee to answer medical questions and provide the necessary clinical leadership? In other words, test utilization success cannot come solely from the laboratory. The clinical practice has to be engaged and has to take this as seriously as we do in the laboratory. Next, are there opportunities to dialogue between the laboratory and clinician? Are there ongoing meetings similar to the transfusion medicine clinical meetings or other types of laboratory clinical practice meetings that will allow you to have a test utilization dialogue? Are you prepared to answer the financial questions that will arise? You have to acknowledge that we are currently in a fee-for-service environment, and until substantial volumes switch to value-based payments, people will question the financial impact of going to an algorithmic or utilization control approach. And, most importantly, do you have a local champion who will be able to lead the talk about utilization issues?
Again, if we go back to the test life cycle that I showed earlier, let’s think about what we can do at each step around this circle. At the time of the physician order, does the clinician have algorithms and guidelines to use? Have we as a laboratory unbundled our tests that would make it a more adaptable to a utilization testing strategy? Are we reviewing requisitions, standing orders, and looking at the redundancy and frequency of testing? Once it gets to the laboratory, do we have a review process for select tests? Do we have a test formulary or some kind of a review process for sendout testing? Do we have a send and hold process where we can hold a specimen until an initial test gets done or until a utilization decision is made? The laboratory owns mountains of data. Use that data as a friend to see how we are doing and to identify other utilization opportunities. Be sure that the reports that go back to the clinician are integrated and clear and answer the questions at hand. And, finally, auditing results of our program and strategy is important to see if we are achieving what we intended from the start.
I would like close this talk by challenging us in the laboratory community to think differently about our roles in the clinical laboratory. We have always been proud of providing high-quality, cost-effective care with a focus of providing service to the patient and clinician – this has been our long-standing patient care paradigm. However, this paradigm needs to change. We in the laboratory have been put in a box and it is now just assumed that we will provide these to the clinical practice. Our new laboratory paradigm demands that the clinical laboratory also provide utilization management, clinical effectiveness, and data integration as part of their expected performance. So we must begin thinking about our clinical role with this new paradigm in mind. It is the integration of all these components that will really define whether we can bring value from the laboratory to our medical practice.
Thank you for the opportunity to speak to you today and hopefully I have been able to introduce you to the concept of laboratory test utilization and the role that pathology and the clinical laboratories can have in this effort. In part 2 of this series, Dr. Amy Saenger will continue this discussion and will talk about test utilization in the Clinical Chemistry Laboratory.