Transcript

Welcome to AACC’s Expert Access Live Online program

Our topic this month is Appropriate Use of Laboratory Tests: Role of the Laboratory and Clinical Pathologist

This month's expert is Brian R. Jackson, MD, MS, from ARUP Laboratories in Salt Lake City, Utah. View the presentation and direct your questions to our online expert. AACC would like to thank Bayer HealthCare Diagnostics for making this program possible.

When doctors order the wrong test, how does it affect the lab’s bottom line?
Memphis, TN

Brian R. Jackson, MD, MS: I’ll assume that you’re talking about the laboratory closest to the patient, as opposed to a reference laboratory. I’ll also take a broad interpretation of the word “wrong.” The answer depends on a number of factors and is severely complicated by the fact that hospitals treat laboratories as an independent cost center, rather than considering laboratory expenses in the overall context of clinical care. One question is whether a test will be reimbursed by insurance. If not, the laboratory may end up footing the cost. A second question is where the test is performed. For tests performed in-house, the incremental cost of performing an unnecessary test is relatively small, since fixed costs represent a large proportion of total cost. For tests sent out, the cost is mostly variable, so that the incremental cost is larger. A third question concerns what additional tests might need to be performed as a consequence of the erroneous test. Overall, misordering of laboratory tests probably has a net negative impact on laboratory balance sheets. The impact to the hospital, however, and beyond that to the healthcare system overall is much more significant.

Several times a year we are faced with this problem. A physician will order a very expensive genetic test on an inpatient or outpatient. Reimbursement can be very poor or nonexistent for these tests and our hospital is responsible for paying for these. How do you handle these kinds of situations in your hospital?
WICHITA, KANSAS

Brian R. Jackson, MD, MS: I posed this question to our clinical pathologist who has primary responsibility over the University of Utah hospital. He prefers to work with specialist physicians and/or committees in addressing such issues. For example, at our hospital HIV phenotype and genotype testing (extremely expensive, and often not covered by insurance) is informally regulated through the ID department. In another example, our laboratory director teamed up with a gastroenterologist to give presentations on appropriate HCV diagnostic algorithms. In response to this, the volume of inappropriate HCV RIBA tests fell dramatically. I don’t believe that there is any one answer, but engagement and dialog are definitely key.

Very well conceived and presented. A list of key evidenced-based references would be very helpful. Also I would be especially interested in some references that elaborate on patient saftey and well being as it relates to the physician ordering or not ordering the right tests. I am interested in promoting resident education and evaluating competency in the areas you have discussed. Fred R Dee MD Residency Program Director University of Iowa
Iowa City, Iowa

Brian R. Jackson, MD, MS: Guidelines for appropriate utilization of individual tests tend to be scattered through the literature. One resource to keep in mind is the AHRQ-maintained National Guideline Clearinghouse (www.guideline.gov). You’ll find that diagnostic recommendations are often buried inside disease-oriented documents. In my opinion some of the best studies on suboptimal use of laboratory tests have been done through CAP’s Q-Probes program; references to these are found in my slides. The best review I’ve found is: Do We Know What Inappropriate Laboratory Utilization Is?: A Systematic Review of Laboratory Clinical Audits Carl van Walraven; C. David Naylor JAMA. 1998;280:550-558. (Also check out George Lundberg’s accompanying editorial) I’m encouraged when I hear residency program directors talk about competence in utilization of laboratory tests. I spoke recently with clinical pathologists at Emory, where the residents are compiling and disseminating an evidence base for appropriate utilization.

Thank you Prof for this presentation. Please I need more explanation about : Setting Priority for QC (Slide 5) and "Errors in Lab testing" (Slide 16) Thanks Wadid Sadek (Ph.D.)
So. Burlington, Vermont

Brian R. Jackson, MD, MS: Slide 5 is just one example of how a health care organization might set QI priorities based on risk assessment. The national patient safety dialog have centered around medication errors and wrong-site surgery, in part because it’s pretty obvious how these hurt patients. My point is that diagnostic errors such as ordering the wrong tests probably have a total impact comparable to that of medication errors and greater than that of wrong-site surgery. Slide 16 summarizes one conclusion of the van Walraven and Naylor article in JAMA; I encourage you to read the full article to see how the actual numbers and how they were derived.

Misuse of laboratory tests which is common in our area can lead to poor diagnostic decisions and challenge for the clinical laboratory is to achieve appropriate utilization so that outcomes for both the patient and the health care system will be optimized but in Africa especially Nigeria we still have a long way to go because of non-availabillity of modern equipment instead we rely on manual where most scientist still pipette blood with mouth.My question is how can we ensure that good quality result will be achieved to the benefit of treating a patient without proper laboratory abuse? Ogboi Sonny Johnbull Public health laboboratory Dept of Commnity Medidicne Ahmadu Bello University Zaria,Kaduna Nigeria AACC no.74812 ogboijb@yahoo.com
Zaria,Kaduna State ,Nigeria

Brian R. Jackson, MD, MS: Appropriateness is definitely a relative term, and is highly resource-dependent. What little I know about medicine in developing countries tells me that laboratory testing is just as important there as in developed countries. Clinical diagnosis in the absence of laboratory confirmation leads to widespread mistreatment. But guidelines definitely need to be tailored to the setting.

I like your concept of "academic detailing" to provide clinicians information. We've discussed having senior clinical lab scientists attend morning rounds to monitor and suggest/shape appropriate test ordering. Are you aware of any facility that is or has tried doing this? Your opinion on that type of "intervention"? Who do you feel is the appropriate individual to "detail" the clinicians?
Boston, MA

Brian R. Jackson, MD, MS: Attending rounds appeals to me for a several reasons – it creates a personal connection between clinicians and lab scientists, it gives lab scientists a better understanding for the environment in which these decisions are being made, and it might yield opportunities for direct education. During my residency at Dartmouth-Hitchcock I spend one month accompanying one of the ICU teams on rounds. I confess that I didn’t offer much guidance during that month – most of the decisions were therapeutic rather than diagnostic – but it was valuable education for me nonetheless. If you give this a try, I’d love to hear how it goes. But I wouldn't use this as the primary method for interacting with (detailing) clinicians. A higher-yield approach is to analyze the ordering and result data in your LIS to identify problem areas, then present focused information to particular sets of clinicians. In terms of who should detail clinicians, I think there are multiple options. Clinicians may respond better to more senior faculty, but drug companies have proven that you can effectively detail even with non-clinicians. I think that effective (and targeted) presentation of material is more important than who presents it.

Your statement that clinicians are familiar with having their utilization monitored...but not yet by pathologists, is something that has always baffled me. What factors do you see behind the fact that the lab/pathologists have not monitored clinician utilization of testing?
Boston, MA

Brian R. Jackson, MD, MS: No physician will ever invite utilization review of their work, so this has to be imposed by someone else. The big question in my mind is why hospitals have P&T committees to review medication use, but don’t have diagnostic utilization committees. I suspect that this has a lot to do with the fact that pathologists haven’t been more professionally active in this area – but this is pure speculation on my part, and I’m certain this doesn’t apply to some of you reading this. There’s a definite concern about clinicians being hostile to this sort of feedback, and I think this holds some pathologists back from engaging more actively. I’m convinced, though, that most clinicians actually welcome useful feedback.

Are clinicians open to spending more time with clinical pathologists to ensure they order the most appropriate tests? Either in educational settings or at the time of ordering tests? Should clinical pathologists work in clinicians' offices and see the patients at the same time as clinicians for immediate on-site consultation and total management of testing?
Chicago, IL

Brian R. Jackson, MD, MS: Clinicians are certainly busy – in fact, that’s the number one concern I hear from them. I don’t think it’s realistic for a pathologist to be physically present in the exam room for every patient, though it might apply in certain situations. I’ve heard, for example, of pathologists who see patients as part of coag consults. Again, I come back to the pharma industry for analogies. (Note that I recognize the negative effects of misleading drug advertising, but at the same time I’m impressed at how effective these companies are at influencing doctors’ behavior.) Unlike, say, orthopedic device reps who follow surgeons right into the OR, drug reps aren’t present when prescriptions are written. And doctors don’t tolerate big interruptions in their schedules to listen to drug reps. So the reps have learned to present extremely focused, extremely targeted educational messages. I think we in the laboratory community can learn from this approach.

Having an on site pathologist for support is a decided advantage. How would you suggest handling the I want, what I want, when I want it syndrome in a rural setting where the pathologist is on site 1 day / month.
Viroqua, WI

Brian R. Jackson, MD, MS: The previous question asked about clinicians’ time as a limited resource, and this question adds the issue of limited pathologist time. If a hospital values high quality health care and patient safety, then they need to support clinical pathology expertise. Having a clinical pathologist on-site at all times may not be feasible, particularly for very small hospitals, but having this support available by phone at all times should be a minimum standard. One day/month strikes me as awfully limited. To get to the other part of your question – it will be tough in your setting to react in real time to individual test orders. But you can provide up-front education to clinicians about known problem areas, and you can provide retrospective guidance about tests that may not have been appropriate. Some sort of hospital committee, including respected local clinicians, would be helpful for implementing this.

How can I get a copy of your presentation. This is what I have been preaching for the last few years. Thank you. Priscilla Cherry
Charlotte, NC

Brian R. Jackson, MD, MS: Email me at brian.jackson@aruplab.com.

here's a follow up to my previous question: I have been asked to create a panel of tests for one or two nurse practitioners. The panel has been described as 'Ovarian Failure Panel', consisting of FSH, LH, Testosterone, DHEA-S, Cortisol, TSH, Progesterone, Sex Hormone Binding Globulin, and Estradiol. This panel would be ordered on many of their female patients. The total cost for all of these tests is currently $630.00! I am hesitant to create such a panel and would like to educate these nurse practioners that this type of ordering pattern may be excessive. Do you have any advice on this situation? Thank you, David King, Lab Director, Pennock Hospital, Hastings, Mi. (269) 945-1212, ext. 325
hastings, michigan

Brian R. Jackson, MD, MS: [I'm only displaying the followup question here, not the original one.] If I were in your shoes, I’d start by talking with local endocrinologists, since I have no specific expertise in testing for ovarian failure. They should be able to direct you to some current literature on the topic and come up with a suggested algorithm for this situation. At that point you could write up a concise summary of the recommendations and distribute it under your name and the endocrinologists’. Hopefully this would put you in a good position to refuse to create the panel. I know this sounds like a lot of work, and depending on the magnitude of this particular ordering problem you might choose an abbreviated version of the above.

Hi, I enjoyed your discussion. How do you extract the data?
Keene, NH

Brian R. Jackson, MD, MS: Excellent question. I’m a bit spoiled in my setting because we have a data warehouse containing all lab orders and results, along with tools and IT personnel to extract and analyze it. I realize that this is beyond the resources of most hospital laboratories. But you may be able to come up with intermediate solutions. When I was a resident, I developed a way to download all microbiology results (along with patient demographics) from our LIS to a desktop database. It wasn’t a perfect system, but it allowed us to analyze the data much more easily. Bottom line: Until LIS vendors provide decent analysis tools, I think labs will need to build this capability themselves. But it can often be done more easily than you might realize.

I have spent the last several years teaching residents how to use the lab appropriately and what I have learned is that they have more sense than the attendings teaching them. But, they gets calls to write orders for the attendings and it makes it appear that they over-order. As a PhD, I have no authority to argue with the attendings. What do you suggest to try and make an impact on a group who cannot be told anything?
Philadelphia PA

Brian R. Jackson, MD, MS: This will always be frustrating. But not hopeless. First of all, you have professional expertise which gives you all the standing you need to present information to attendings. (How much authority does a drug rep have?) Whether they listen or not is up to them. Let me also suggest an alternative model. When I was a resident, our bloodbank director (James AuBuchon, now Chair of Pathology at DHMC) created a hospital transfusion committee. This included prominent faculty members from surgery and medicine. Whenever a blood product order was placed which appeared inappropriate, a path resident was paged to investigate. The resident would reiterate the guidelines to the ordering physician (usually another resident, but sometimes the attending). Sometimes the clinician would insist on ordering a product contrary to these guidelines. In these cases the lab would dispense the product, and the hospital Transfusion Committee would review the case afterwards. If the committee determined that the order was inappropriate, they would issue a formal letter to the clinician (and their attending, if applicable). Overall, even though this approach couldn’t guarantee misuse in specific situations, it was very effective at changing overall blood product utilization.

Do you think molecular diagnostics will require clinicians to collaborate more with clinical pathologists? What form do you think the collaboration will take? How can clinical pathologists prepare to take advantage of molecular diagnostics to get more involved in diagnosis and treatment decisions?
Chicago, IL

Brian R. Jackson, MD, MS: Definitely. Most of molecular diagnostics involves rapidly changing technology and rapidly changing indications, which pretty much guarantees that most physicians will fall behind in their knowledge. I think we should be creative in coming up with forms of consultation. One important form is the lab result itself, which should cover relevant ordering and interpretive info along with the result. (The American College of Medical Genetics has some guidelines here; sorry I don’t have the reference handy.) And results need to be carefully edited for clarity and conciseness – I personally favor bullet-points where possible rather than paragraphs. The second part of your question is very interesting. For genetic testing, I think laboratories can benefit greatly by partnering with genetic counselors. Beyond that I’m not sure I have any good answers.

Why do pathologists not play a more direct role in assisting clinicians with selecting and interpreting the most appropriate state-of-the-art testing strategies for most efficiently establishing/confirming/ruling-out preliminary diagnoses based on clinical presentation or previous lab data? It seems to me that there is often an excessive lag-time between the establishment of appropriate protocols by the pathology experts and the utilization of them by all clinicians. Could not computers be utilized more effectively to provide a tool that could assist clinicians in this regard?
Durham, NC

Brian R. Jackson, MD, MS: Computers definitely play a critical role, but do not replace the expertise or authority of a clinical pathologist. I see the biggest opportunities in two areas: 1. Providing up-front guidance on test ordering and interpretation. This information is ideally linked to an online test menu. 2. Allowing retrospective analysis of ordering patterns, to discover and monitor problem areas. I’ve also talked with pathologists who are enthusiastic about using CPOE to regulate utilization. This can certainly be done effectively – doctors at Brigham and Womens did this to improve utilization of abdominal xrays, for example – though I think you need to be a little careful. If a CPOE system requires docs to go through additional hoops before ordering any tests, it will generate hostility. But targeted use of CPOE could be quite powerful.