April 30, 2008
The Practice of Laboratory Medicine in the Twenty First Century – Achieving Relevance in a Tidal Wave of Change
“The only way to predict the future is to create the future”
Urgency to Reconsider Path Forward for Laboratory Medicine
- Unabated increase in health care costs
- Compounding increase in new diagnostic technology
- Heightened interest in safety and quality of medical care
- Increased attention to decision making grounded in evidence based medicine
National Health Care Spending
- 1960 -- $28 billion
- 1970 -- $75 billion
- 1980 -- $255 billion
- 1990 -- $717 billion
- 2000 -- $1.36 trillion
- 2006 -- $2.16 trillion
National Health Care Spending
- 1960 – 5.2% GNP
- 1970 – 7.2%
- 1980 – 9.1%
- 1990 – 12.4%
- 2000 – 13.8%
- 2006 – 16.5%
- 2015 – $4 trillion (20%)
- 2030 – 25%
- Access -- 45 million uninsured
- Infant mortality – 28th
- Adult longevity – good but no cigar
- Balance between preventive and end of life care out of kilter
RAND Health Study (2003)
- 4612 adults in 12 cities
- Evaluated 439 quality of care indicators for 30 different medical conditions
- Good methodology for indicators – literature reviews and Delphi techniques
- Good methodology for evaluation – duplicate chart audits by trained nurses, supplemented by interviews
RAND Health Study
- Acute and chronic conditions; preventive care
- Screening, diagnosis, treatment, follow-up
- Participants received 55% of recommended care
- $30 to $50 billion
- 10 billion tests/year
- 33 tests/ American
- Disproportionate punch – 70 to 80% of medical decision making
New Technology -- FDA Vantage Point
- Not an explosive growth but slow and steady incremental growth (5.1% per year)
- Improvements in existing technology (35% re-investment rate)
- Cutting edge new technology
- Like compound interest, growth although not exponential will be surprising
Laboratory Technology – Impact on Efficiency (costs)
- Biomarkers as tools for personalized medicine
- Biomarkers as tools for improved diagnosis of infectious diseases
- Improved diagnostic device performance for core functions – better design, reagents, instruments, software
- Novel new technology
- Tailored but old fashioned choices
- Treatment selection*
- Not your father’s Oldsmobile
- When drug becomes linked to diagnostic; drug effectiveness becomes a prisoner to the diagnostic (unsettling to colleagues in drugs)
Biomarkers for Infectious Diseases
- New technology to detect microorganisms
- Profound impact on public health (ordinary, emerging, and biothreat pathogens)
- Improved quality of testing challenging traditional gold standards
Improved Diagnostic Device Performance
- Design and manufacturing of routine devices have improved
- Impact on central lab test
- Impact on point of care tests
- Impact on home use tests
- Improved access, speed and performance
Novel New Technologies
- Continuous glucose meters
- Breath tests
- Artificial intelligence or neural networks
- MNR – light spectroscopy
- Brave new world
- Challenge of menu selection
- Interpretative challenge
- Potential to be a cornerstone in change in medicine
- Lower costs – faster better decisions
- Improve quality – faster better decisions
- Lead to better outcomes
- Not new issue (CAP 1946)
- Not new issue to medicine -- JCAHO
- Institute of Medicine – 1999
- To Err is Human
- 98,000 deaths per year
- One million injuries
- Galvanized health care community and patient advocacy groups
Laboratory Interest -- dual
- Errors in testing – impact on patients
- Use of testing to measure quality of cares – defining, capturing, minimizing health care system errors
Errors in Testing
- Small but growing literature
- Lack of consistent language and methodologies
- Variable, method dependent reports on error rate
Errors in Testing
- Complaint based data -- .05%
- Total system audits -- .47%
- Process reviews (blood bank) -- 5%
- Process reviews – blinded secondary evaluation (anatomic pathology) – 4%
Errors in Testing
- Preanalytical – 41% to 68%
- Analytical – 4% to 13%
- Postanalytical – 18% to 55%
Errors in Testing
- Numbers are low – percentage (.05%)
- Real world of test use may mitigate
- Numbers are high – volume effect (500,000 errors; 50,000 adverse events/year)
- Not competitive with best in other industries, not six sigma – airlines
Errors in Testing -- Analytical
- Analytical performance overall is robust
- Labs have most control
- Labs have most regulation
Errors in Testing -- Analytical
- Begs the issue of inherent quality of tests
- Good and getting better BUT
- Major problem in lack of standardization and traceability (commutability)
Three Outlines for Addressing Errors
- Novis – 2004
- Elin – 2004
- Howanitz – 2005
- All rely on standard quality tools: from proper collection and labeling to timely reporting
- Laboratory errors may be defined as any defect from ordering tests to reporting results and appropriately interpreting and reacting to these
- “Pre-pre” and “post-post” analytical error (Laposata 2007)
- Perception misuse of existing technology (overuse and under use)
- Problem in ordering and interpreting tests
- Well established perception with existing tests
- Likelihood to blossom with wild new tests
Van Walraven (1998) – Do We Know Inappropriate Lab Use?
- Review of 137 papers (44 used)
- Implicit criteria – 11
- Explicit criteria – 33
- Range 4.5 to 90%
- Need more vigorous methodology
- Effect of false positive – Ulysses syndrome
- Delays in actual disease detection
- Emotional/work impact
- Iatrogenic anemia
- Effect of false negative
- Incorrect diagnosis without corrective follow-up
- Quality and economic costs of wrong conclusions
Broadest View -- Kind
- Laboratorians like to tend their own gardens
- Focus on technology of new (and old tests) and not use
- Connectivity between pathologists and user (both pre and post analytical) in general is suboptimal
- Laboratory medicine is poorly taught – clinicians don’t understand language of labs
Broadest View -- Blunt
- Abrogation of responsibility, broken chain from laboratory to patient
- Converted lab work into commodities not services
- Generate mentality labs are like cafeterias, subject to sub-contract, consolidation, use of prepackaged foods, subject to competitive bidding
- Laboratory-clinical connections poorly tended
RAND Health Study (2003)
- Not exotic tests
- Theophylline – 62%
- Potassium – 22%
- FOBT – 72%
- HBA1C – 42%
- Lipid tests – 42%
Evidence Based Medicine
- Sackett definition (2001): The conscientious, explicit and judicious use of current best evidence in making decisions about the care of patients
Missing Ingredient – test value to system
- Evidence based medicine
- Lab practice is highly empirical
- Not new concept
Fryback and Thornbury (1991) – The Efficacy of Dx Imaging
- Technical efficacy (correct analytical signal)
- Diagnostic accuracy efficacy (correct clinical signal)
- Diagnostic thinking efficacy (impact on decision making
Fryback and Thornbury (1991) – The Efficacy of Dx Imaging
- Therapeutic efficacy (impact on therapy)
- Patient outcome efficacy (impact on outcome)
- Societal efficacy (public health good)
Marchevsky (2005) – General Processes
- Formulation of specific diagnostic question
- Search for specific scientific literature
- Critical appraisal of the evidence
- Incorporation of “best evidence” from several “reliable sources” with personal clinical experience for development of rules
- Evaluation of effectiveness/efficiency of recommendations
Knowledge is Power
- Who will make decisions
- Third party payers
- Architects of practice guidelines
- Health care providers
- Patients (advocates)
Challenge to Physicians
- Keep up with technology
- Become educators and innovators in use of new technologies
- Create strategic alliances to facilitate use of lab services
- Alternative option is to loose the farm*
Lewandrowski (2003) –Managing Utilization of New Dx Tests
- Best interest of patient
- Actively solicit opinions of others
- Project a team-oriented approach
- Develop organizational skills
Lewandrowski – Managing Utilization of New Dx Tests
- Expand fund of laboratory knowledge – watch for changes in use of old tests, nuances provided by new tests
- Maintain persistence, execution and follow-through
- Be prepared to negotiate deals and seek win-win arrangements
- Be humble. Admit when you are wrong or lack knowledge
Preparing For the Charge
- Institutions will need to be ready with programming
- Educational programs will need to be attentive to menu of educational needs
- Individuals will have to do what physicians are supposed to do – be perpetual learners and also leaders
The Straight Facts
- IVD industry slow to promote value base for competitive reimbursement
- Laboratorians slow to lead in use of lab tests and to demonstrate how lab tests can impact patient safety, total costs, and health care outcomes
- Physicians have used lab tests empirically with tolerance for too little evidence
Lundberg (1998) -- The Need for an Outcomes Research Agenda
- “The real reasons for laboratory testing should be to improve the outcome for the patient’s or the public’s health. The time has come to develop a solid research agenda to measure and monitor these vitally important outcomes and to change practices when indicated by the results.”
Sodeman (2007) -- Taking Ownership of New Technologies
- “Our future has never looked better, but I do intend to convey some urgency. Our growth potential is intimately linked to new tools whose patient care applications are well within our grasp. “The market moves” as investors like to say, and when it moves it creates momentum… We need to reach out now and take ownership of these tools.”
Sodeman (2007) -- Taking Ownership of New Technologies
- Ensure we drive technology rather than watch technology drive us as health care providers, laboratorians
- “I shall be telling this with a sigh Somewhere age and ages hence; Two road diverged in a wood and I --- I took the one less traveled by And that has made all the difference”
The Practice of Laboratory Medicine in the 21 st Century:
Achieving Relevance in a Tidal Wave of Change
Moderator: Jill Marion
April 30, 2008
1:00 pm EDT
Coordinator: Please stand by for today’s conference call. We will begin in just a couple of minutes. You will hear music until the conference call begins. Again, please continue to stand by. We will begin today’s conference in just a couple of minutes. Thank you.
Welcome everyone and thank you for standing by. At this time I would like to inform all participants that you’ll be able to listen only until the question and answer session of today’s conference call. Today’s conference is being recorded. If you have any objections, you may disconnect at this time.
I would now like to turn the conference over to our first speaker, Miss Jill Marion. Thank you ma’am. You may begin.
Jill Marion:Welcome to the Practice of Laboratory Medicine in the 21st Century: Achieving Relevance in a Tidal Wave of Change presented by our guest speaker today, Dr. Steven Gutman. My name is Jill Marion and I will be your moderator.
I’m a biomedical engineer with the Food and Drug Administration Office of Surveillance and Biometrics. I’m also the lead of the FDA’s lab net program. We’re glad you could join us today.
We’re expecting over 170 people on the call. Today’s program is expected to last one hour and 30 minutes which includes our feature presentation and a ten minute question and answers session at the end of the program.
The presentation slides can be found on the registration site. For directions on how to access these or for any other problems you have on this call, please call the toll free MedSun line at 800-859-9821. And again that number is 800-859-9821.
After the presentation, an electronic certificate of participation can be accessed through the registration site. I would like to begin today’s program by introducing Marilyn Flack, the director of FDA’s Patient Safety staff, who would like to make a few introductory remarks. Marilyn?
Marilyn Flack: Thanks, Jill. Well welcome to everyone on the call today. We’re so excited that so many MedSun hospital laboratory clinicians and professionals are interested in learning more about FDA’s perspective on the future of laboratory products.
There are some on the phone today who may not be as familiar with the MedSun project and one of its subnetworks, LabNet, as some of the others on the phone. The LabNet team is sponsoring this conference today, but I’d like to give a quick overview about MedSun.
First, FDA is divided into several (sectors), each of which focuses on a type of FDA regulative product. Our center -- the Center for Devices and Radiological Health -- clears medial devices for marketing, monitors those devices once they’re marketed to detect problems that occur once they’re in widespread clinical use, then work with manufacturers and the clinical community to improve medical devices.
FDA learn about problems with marketed devices through a variety of mechanisms. The Medical Product Safety Network or MedSun is the mechanism our Center uses to obtain knowledge about the problems experienced by the users of these devices in the clinical community.
This is a targeted surveillance system, and we currently have 350 hospitals from across the country participating in the program. The sites use secure on-line reporting tools to report problems to us. (The sites also) participate in focus groups, surveys, research, and other means of learning about and understanding problems with devices.
MedSun’s working to create a learning environment where FDA and the clinical community may interact to discover, understand, and solve problems with the use of medical devices. A key component of MedSun is the feedback we strive to provide to our sites about problems we’re learning about in the network.
One of the reasons I think MedSun works so well is that it’s permitted all of us who care about improving safety with medical devices to see that we can accomplish much if we all work together.
MedSun’s a high priority for the FDA Commissioner and for the Secretary for the Department of Health and Human Services. In fact, MedSun’s part of our Commissioner’s strategic plan for improving patient safety nationwide.
So we’re constantly seeking ways to improve MedSun, and one of our goals for the next few years is to target specific areas of the hospital to learn more about device problems in those areas. We call these our subnetworks. LabNet is one of these subnetworks. Currently there are 44 labs from our MedSun hospitals participating and also the very large NIH laboratory, which makes 45 participants.
Lab directors and (laboratorians) make up our reporters for LabNet. Today we’re very pleased that our FDA colleague, Dr. Steven Gutman, (is with us and) will be talking about the many changes affecting how labs function in today’s ever-evolving healthcare environment.
So again, thank you for joining this audioconference. And if you’re currently one of our LabNet reporters, keep up the great work in helping us solve lab device problems. And if you’re not yet one of our reporters, please think about joining. Through LabNet you have the opportunity to bring your problems and questions with the use of in vitro diagnostic devices and point-of-care devices to the attention of FDA. This is your chance to have your voices heard.
(You) will be provided throughout this session (with) the phone number you (can) call if you’d like to have more information about how to join LabNet. So thank you very much and I’ll turn it back to Jill.
Jill Marion:Thank you, Marilyn. I’d now like to introduce today’s’ guest speaker, Dr. Steven Gutman, director of FDA’s Office of In Vitro Diagnostics. Dr. Gutman is a board certified pathologist and holds a master’s in business administration. He has 16 years of experience as an FDA regulator in the area of in vitro diagnostic devices and 15 years of practice experience managing clinical laboratories of all sizes. Dr. Gutman you may go ahead and begin.
Steve Gutman: Good day. Mario Plebani, an Italian chemist who has created a literature niche on the quality of laboratory testing, introduced the (aphorism) that the only way to predict the future is to create the future. And I think that MedSun in general and LabNet in particular are in a unique position to be the eyes and ears for FDA and help FDA remain an informed stakeholder in the healthcare community in trying to create the future. So this is an opportune group for me to be talking to.
There are four reasons why this is an opportune time to be talking about laboratory relevance, and there are four reasons I believe the call for a certain urgency to consider or reconsider the path forward for laboratory medicine. And those four reasons are the unabated increase that we seem to be experiencing in healthcare costs, the compounding increases in new diagnostic technology, the heightened interest in safety and quality of medical care and the intersection of those issues with laboratory testing, and last but not least -- last but perhaps most -- the increased attention to decision making in and outside of labs grounded in evidence-based medicine.
So to begin with national healthcare spending, in the year 1960, the year I entered middle school, we were spending $28 billion a year on healthcare. Since that point in time there has been an exponential - essentially a 75 fold increase in that spending. We are currently spending in the neighborhood of $2.2 trillion per year on healthcare. And what’s particularly disturbing in looking at those numbers is that it’s not the absolute growth, but the relative growth (in terms of our) gross national product.
In 1960 we were spending essentially 5% of our gross national product on healthcare. In 2006 we were spending 16.5% of our gross national product on healthcare. And if you follow current trajectories, in my lifetime, certainly by 2015, we will be spending 20% of gross national product on healthcare. And in my children’s lifetime in 2030, we (are) projected to be spending 25% - perhaps 30% or more -- of our gross national product on healthcare. We are a very rich country but we’re not that rich.
And a logical question to ask, given (these) incredible, actually world record breaking expenditures, (is) what are we getting for our money? And the metrics that you can look at -- the standard metrics -- are not terribly reassuring.
In terms of access, we currently have in the neighborhood of 45 million Americans who in fact are uninsured. In terms of infant mortality, we are 28th among countries. In terms of adult longevity we’re higher but we’re still not exactly at the top. So we’re good, but we don’t get a cigar. And in terms of the balance between how we spend healthcare dollars and preventive medicine in standard care (and) in end of life care, there seems to perhaps be an imbalance in the way we direct our expenditures.
The definitive statement on quality might actually have been written about five years ago in a somewhat unheralded article in the New England Journal of Medicine by the RAND Group.
And the RAND Group actually studied healthcare in 4600 adults in a dozen cities, evaluated 439 quality of care indicators for 30 different medical conditions, used good methodologies for those indicators -- using a combination of literature reviews and then expert (Delphi) techniques to pick the indicators and to pick the end points to show the indicators were being met -- applied a remarkable methodology for evaluation. They essentially had duplicate chart reviews performed by well trained nurses and nurse practitioners supplemented by specific interviews of healthcare providers.
And in looking at acute and chronic conditions and looking at preventive care and looking at the screening, diagnosis and treatment, and follow up, the RAND study concluded that this represented a sampling which suggests that acceptable standards of care were met in about 55% of study cases.
Now in terms of the lab industry, we’re actually a small part of that big picture -- a small part of a colorful mosaic. We actually are pocket change in terms of healthcare costs -- about $30 to $50 billion. Less than 2% of the total healthcare spending is actually directed towards in vitro diagnostic testing.
Although we do get a lot for our money because that $30 to $50 billion does pay for 10 billion tests per year. The average American has 33 tests performed per year. We are the most tested people on God’s earth. And in terms of (being) ubiquitous medical products, I would argue that there is probably no one on this line who has either not been tested or does not know someone who has had a lab test in the last year and that we probably rival Band-Aids for being a product that is known to virtually all Americans.
And there is a disproportionate punch in terms of the impact of that testing on healthcare decision making and healthcare treatments, because in fact it is estimated that 70 to 80% of medical decision making is not based on a history, is not based on a physical exam, but is based on results of lab tests.
From the FDA vantage point, the second operative and powerful force is the new technology itself. And I would predict not an explosive growth but a slow and steady incremental growth. And the estimate you can find in the literature on the Internet would suggest about a 5.1% growth in technology in the (unintelligible) industry per year.
It is a little known fact that this is an industry that is very entrepreneurial and that actually spends a very high percentage of revenues generated on test development, redevelopment (and) improvement. There’s actually estimated to be a 35% reinvestment rate. That actually is higher than the reinvestment rate for drugs. And the result is that cutting new technology -- cutting edge new technology -- does appear quite regularly in FDA (shop) as people are trying to bring new products to market.
And so I - although I don’t believe this will be explosive, I don’t think it will be (logorhythmic) or exponential. I think it’ll be a little bit like compound interest and a well-managed financial portfolio -- that at the end of the day, there will be a surprising increase in the menu of tests - menu of technologies available, which is overall good news, particularly good news of (unintelligible) if used properly.
And I take that technology and would arbitrarily divide it into four areas. These - this is somewhat of an artifice, but I would look at these as biomarkers, as tools for personalized medicine, as biomarkers is tools for improved diagnosis of infectious diseases, as improvements in simple diagnostic device performance -- because the core functions themselves are better so the existing products get more accurate and more reliable -- and then I - last but most intriguing is a novel new technology.
In the area of personalized medicine, this refers to the - this advent of tailored choices that based on molecular diagnostics based on genomics, based on proteomics, based on new technology. And although the principle is new, the choices for testing of course are not new. And we’re looking at trying to determine the best or the most appropriate or the quickest diagnosis. We’re trying to understand the prognosis of the disease, we’re trying to monitor the course of the disease and perhaps of particular importance we’re trying to use new diagnostics to help us use either old or new therapies to better affect.
At FDA if there was one epiphany in the last four or five years it has been the unveiling of the critical path in 2004, in which it was pointed out that diagnostic tests, particularly what we might refer to as biomarkers, were not your father’s Oldsmobile. Biomarkers were not (only) important to make traditional diagnoses, but that biomarkers actually might also be valuable in helping to remove obstacles in the critical path of drug development.
The notion was that biomarkers might be a better way to select candidate drugs for further study than the old fashioned used of animal models in empiric or small patient studies, and that biomarkers might allow for more streamlined or more efficient studies.
And in my view, that is probably true. It’s not entirely tested, but it’s probably true. But when a biomarker starts to become determinate in the use of the drug and when the biomarker in fact determines who might benefit from a drug or who might better not be given a drug because they’re likely to show an adverse reaction, I would argue that the drug becomes inextricably linked to the diagnostic -- that the drug effectiveness becomes a hostage, becomes a prisoner, becomes a slave to the diagnostic, and the diagnostic becomes an equal partner in importance - or the diagnostic becomes an equal partner in importance to the drug. My colleagues in drugs might find that view unsettling, but it certainly is my view.
Biomarkers for infectious diseases are not exactly personalized medicine. They actually, I would argue, are public medicine or public health medicine or tests used to make decisions not only about people but about surveillance in epidemiology in populations in general.
That is not to suggest that they don’t have a profound potential impact. And the biomarkers for infectious disease may change the way we practice current infectious disease, and microbiology may improve our ability to detect emerging infectious diseases like avian flu, and may actually be essential as we fight the war on bioterrorism and look to head off the potential threat of biothreat pathogens.
What’s interesting about the new biomarkers for infectious diseases are that they actually challenge the scientific paradigms that’s historically been used to estimate performance for tests for infectious disease, and that they may be better than the gold standards or the reference standards used. And trying to demonstrate that may be quiet challenging scientifically and also from a regulatory perspective.
The third general category would be sort of an incremental improvement in diagnostic device performance for run of the mill tests that now are the mainstream and life’s blood of the average lab. That’s because the design and manufacturing (of) routine devices has improved in everything -- the hardware is better, the software is better, the agents are better, the design is better.
And the results of those quality improvements show in better tests for use in central labs, in better tests and more portable tests for use for point-of-care testing, and for potential, frankly, for tests to be made simple and easy and reliable so that they may be robust enough to be used, when appropriate in patients’ homes. And the result is an improvement in access, improvement in speed and improvement in performance.
And then last, and as I said, most intriguing, are the novel new technologies we’re starting to see. We’re starting to see continuous glucose meters, which at the least may refine the way we manage diabetes, and at the most may at some point allow some patients to toss their meters for at least some or many of their finger sticks.
We’re starting to see breath tests move from traditional use for GI diseases to non-traditional use for metabolic and other non-GI diseases. We’re starting to see artificial intelligence or neuro-networks taking multiple signals and generating more refined diagnostic choices. And we’re also starting to see interesting new in vivo technology -- the light spectroscopy MNR technology is looking at prostate cancer, looking at pancreas cancer.
I would say we’re not close to (operating in a world) where a patient can simply step into a large black box or a large sort of photo outfit and have a biosensor detect their state of health or pathology, but we’re clearly starting to make the first steps in that direction.
And with this growing arsenal of fascinating improved old tests and this growing arsenal of intriguing and challenging new tests, we are seeing increased dilemmas, increased challenges in menu selection -- what tests should actually be offered to the average patient -- and also interpretive challenges -- if you have an improved old test or if you have a fundamentally new test, what does that mean in terms of what decisions you make about a patient or what you actually tell the patient?
I would say there’s a potential, there’s an opportunity for laboratory medicine to be a cornerstone in change in medicine to lower cost, to improve quality and to lead to better outcomes if this new technology is harnessed and used appropriately.
The third powerful moving force is the heightened interest in 2008 in patient safety. And for laboratories that’s not a terribly new issue. In fact, when the College of American Pathologists was first formed in 1946, one of the first items on its agenda -- one of the first items they began to address -- was the issue of laboratory safety.
It’s not new to medicine. The Joint Commission has for literally decades been monitoring healthcare indices and trying to ensure quality in a wide variety of healthcare settings, but it does have a new sense of urgency and a new sense of drama as a result of the Institute of Medicine’s 1999 report called To Err is Human. And in that report there was an observation that simple and avoidable medical errors could be responsible for as many as 98,000 deaths per year and 1 million injuries per year. And that report, whatever else it did or (did) not do, clearly put a new spotlight on the issue of patient safety.
And although those may be overestimates -- some folks would suggest that (these) are overestimates of the real numbers -- that report galvanized the healthcare community, it galvanized patient advocacy groups, and it did interest almost anyone was involved in the trying to of patients.
The laboratory interest in patient safety is a two-edged sword. We are very interested in errors in testing itself and the impact it might have on patients, but we actually have a more global interest in that lab tests can be used in many cases as surrogate markers for treatment, and for right treatment decisions or wrong (treatment) decisions or to assess what’s going on with a particular patient. So laboratory testing actually can be used and is increasingly being used as a measure (and) as a monitor of the quality of care being delivered to a particular patient.
In terms of laboratory testing errors, there is a small but growing literature. It is a colorful literature with a lack of consistent language and methodologies and with variable method-dependent reports on the error rate. So if you look at very passive reporting systems that are based say only on complaints, you may see errors as low as 0.05%. If you look at more systemic audits, you may see error rates that might approach 0.47%. And if you actually do very active process reviews, you may see errors as high as 4 or 5%.
And those errors have been well studied in multiple publications, and it’s now well known that the single most important source of error in lab testing actually occurs in the pre-analytical phase of collection -- labeling, processing, bringing the right samples to the lab at the right time and the right way -- that the second most common source of error is post-analytical, and that in fact the part of the lab processing that we have most control over, we in the lab, which is the analytical part of testing, actually contributes the smallest component to total errors.
It’s worth noting that when you look at errors you can look at it as the glass half full. You can say, “Well gee,” particularly in complaint-based systems, the percentage is very low, perhaps as low as 0.05%, that in the real world of test use, sometimes an error will simply be a draw. You’ll switch a sample and it’ll be from one normal calcium to another normal calcium, so fortunately there’s no harm done. In other cases, it will lead to absurd results. The doctor’ll take a look at the results and will say, “Oh, darn, the lab screwed up again. I got to reorder it.” So you might view this as not so bad.
If you want to look at the glass half empty, the fact is that 0.05%, when multiplied times 10 billion tests per year, is actually a real number of tests like 500,000, and that even if only 10% of those were to result in adverse events, it means that the laboratory is responsible for perhaps as many as 50,000 adverse events per year.
And if you want to look at the glass half empty and you compare what’s going on in the laboratory with what’s going on in the airline industry or other state of the art industries, that we’ve not actually reached the Six Sigma level yet.
In terms of errors in testing, there is a unique opportunity for people participating in this call who are part of MedSun and part of LabNet to actually inform FDA. It is very hard for us in the pre-market program to predict with absolute precision how tests will perform from the prototypic studies we’ve seen in a submission, to the real world where there’ll be thousands of labs with perhaps tens of thousands of operators with perhaps hundreds of thousands -- perhaps even millions -- of patients. And so the information that you provide us about test performance, especially test performance gone wrong, will incredibly rich and help our program.
In general it is our impression based on preliminary experiences with LabNet that labs do have control, that labs are regulated, that labs are in general happy with the products that they’re getting to run their labs, and that responses fortunately are usually responsive to the complaints that are raised and that things in general are good. That’s fortunately or unfortunately not always the case, and we do appreciate getting the signals when that’s not the case.
It’s worth noting that in spite of the strength, what I would argue is the robustness of the analytical part of the testing lifecycle does beg the issue of the quality of the tests themselves. And as I’ve noted, because of the technology changes, I do believe that (they) are getting better. But I also believe that there remains a problem and lack of standardization of tests and a problem of traceability and a problem of commutability -- being able to move from lab to lab and expect to see numbers that actually resemble each other.
And you don’t have to look very far. You can look at common tests like (troponin) or like (PSA) or like (D-dimers) and you can have a heart attack simply by moving from one lab to another because the cut points are different and the standards are different and the decisions are different from assay to assay. And that in certain ways is rooted in the regulatory laws that have us make many determinations based on substantial equivalents, not based on traceability and not based on de novo safety and effectiveness.
There are outlines for addressing laboratory error. (In the accompanying slides) I’ve actually given the names and the authors and the date. They all rely on standard quality tools and they all focus on the big picture, which is everything from proper collections of sample, proper labeling of the sample, and timely reporting of the sample.
If you want to take the broadest view of what’s going on here, however you can’t just look at the lab. You can’t just look at the test. You actually have to look functionally at what is going on. Laboratory errors and the broadest (definition) is any defect from ordering the test -- actually picking the test for that particular patient -- to reporting the results and then properly interpreting the test as that test should be interpreted in that particular patient.
And actually (Mike Laposata of Massachusetts General Hospital) is so hung up with this concept of looking at the early and the end stage of the lifecycle he’s actually given them names. And he talks about ordering tests as being pre-pre-analytical and interpreting tests as being post-post-analytical.
And in the broadest view, there is certainly a perception that there may be some, may be a lot of misuse of existing technology, both overuse and underuse, that there may be problems in ordering and interpreting tests, and that this is true for existing tests, and this is likely to be true - perhaps it’s likely to be worse for very novel new tests where there is no history of test use.
And in fact about a decade ago, a scientist named (van Walraven) did a systemic review looking at inappropriate laboratory test use, tried to define the highest quality papers on inappropriate lab use, and came to the conclusion that these papers perhaps, because of (a methodologic) problem, show wide ranging differences but inappropriate use. The best case scenario was a 4.5% and a worse case scenario perhaps as high as 90%.
And this is serious business from the standpoint of lab safety and lab quality, because overuse will tend to result in the appearance of false positive results. False positive results will often end up in repeat testing, extending testing, information to patients that may make them worry, may have impacts on them in terms of work, may cause them to be drawn so much they become anemic. And the wandering around trying to solve a false positive lab tests actually has a name. It’s actually appeared in the laboratory literature. It’s called the Ulysses Syndrome.
And there are also problems with underuse -- effective false negative results -- because underuse won’t - if you fail to order a test, you won’t get a positive result. And that is that you may end up with an incorrect diagnosis without a correct follow up, and the quality and economic cost of wrong conclusions needs to be considered.
To be kind in the broadest view, laboratories - laboratorians tend to - like to tend their own gardens. They like to focus on the dazzling technology of new and improved old tests more than on the use of those tests, that the connectivity between pathologists, lab managers, and other laboratory professionals and users in general tends to be suboptimal, and laboratory medicine in our medical schools is poorly taught and clinicians often don’t understand the language of laboratory. That’s physicians, that’s nurse practitioners, that’s physician’s assistants. They often don’t actually get the lab lingo.
To be blunt in the broadest view, there may be some abrogation of responsibility and actually a broken chain in what should be a seamless flow from the lab to the patient. That may have resulted in lab work being converted into something that is being viewed by many as commodities and services, creating a mentality that labs are like cafeterias -- subject to contract, consolidation to use of pre-packaged foods or to competitive bidding. And to be blunt, in some cases laboratory clinical connections are poorly tended.
And if you go back to the RAND study and look, the RAND study would certainly suggest that there’s a use problem. And it’s not for exotic uses. They found that misuse (range) was as low as 22% for potassium. That’s a rather disturbing analyzed (unintelligible) underused to as high as 72% for fecal occult blood when you can’t imagine fecal occult blood is underused because it’s not a popular test to collect, perhaps not a popular test to perform. It’s a rather important one, however.
And then of course the last issue -- I would say the issue that is likely either the sun shining or the cloud over the whole field -- is a movement towards evidence-based medicine. And evidence-based medicine has been around now for more than a decade but a particularly nice definition popped up in 2001, which is the “conscientious, explicit, and judicious use of current best evidence in making decisions about the care of patients.”
And I would argue that evidence-based medicine in all too many cases is a missing ingredient in the life of the lab. And that lab practice, in spite of its underpinnings in the work done by (Rocha) tests -- (Rochatensky) and (unintelligible) the fact that the pathology and autopsy in my entirely biased view as a pathologist -- is the grounding for much of modern medicine that in the current practice of medicine we tend to do things in a somewhat empiric manner and that we have been late comers to the party of evidence-based medicine.
And that’s not because evidence-based medicine is a new concept. You can actually go back to 1991 and see in the imaging literature what should perfectly parallel the lab approach towards evidence-based medicine. One has to look at what these imaging folks call technical efficacy. That’s for us a correct analytical signal.
You have to look at diagnostic accuracy. That for us is the correct clinical signal. You have to look at diagnostic thinking efficacy-- the impact on decision making. You have to look at therapeutic efficacy -- that is the impact on therapy.
Where the rubber really hits the road is that you need to look at patient outcome -- impact on outcome. And then I think certainly beyond probably much government work is to look at the societal efficacy and to look at the public health (good). That’s very complicated and high level but probably worth pursuing.
And evidence-based medicine is not a foreign process. Marchesky, a pathologist actually in 2005, had a nice set of general processes in which he described what I would say (are) the key features of evidence-based medicine, which would be formulation of specific diagnostic questions, search for specific scientific literature, then a critical appraisal of that literature, incorporation of best evidence from several reliable sources with personal clinical experience in order to develop rules of behavior, and then some kind of evaluation process to see if what you’ve done, what you’ve recommended actually makes sense and is doing what it’s purported to do.
And I would argue that evidence-based medicine is important because knowledge is power and that there is a wild variety of stakeholders who are trying to make their mark on the lifeblood of labs, and I would perhaps say the lifeblood of hospitals, is laboratory information. So certainly third party payers have a vested interest.
The architects of practice guidelines have an interest, the average healthcare provider has an interest in 2008. Patients are becoming more and more advocates for themselves. They are working as individual advocates, they are working collectively with advocacy groups, and they are showing up in their physician’s offices with information sometimes correct, sometimes not so correct, from the internet.
And of course pathologists and members of the lab team they lead, and frankly members of the medical community who are purveyors of the lab information, also have an important role and a place to play in this game.
The challenge to physicians and to healthcare providers in general is to keep up with the technology -- to become educators and innovators in the use of new technology to create strategic alliances to facilitate the use of good lab services.
And it would be my argument that we on both sides of the fence -- within the lab and the lab users -- need to be a party to this and that the alternative option is to lose the farm. The alternative option is to have scientists turn over responsibility for technology to administrators, to have MDs, to have (MTs), to have RNs, to have various other types of clinical scientists turn their work over to MBAs. I don’t think that - I may have an MBA, but I don’t think that’s where I want decision making to start and end.
And again, I think that there’s a unique possibility for MedSun as part of a network that shares information, that provides FDA with signals and that is a place that we can do risk communication, can in small and perhaps in larger ways contribute to this - answering this challenge.
(Lewandrowski), a pathologist again at Mass General, actually talks about the common sense of managing utilization of new diagnostic tests. And it actually are fairly (unintelligible), fairly pedestrian, fairly important tenants that he ascribes to this task. So he talks about keeping in mind the best interest of the patient, he talks about actively soliciting the opinions of others, all opinions count -- projecting a team oriented approach, and the need for organizational skills as you’re mapping the use of lab tests.
He talks about expanding the fund of lab knowledge, he talks about persistence, execution, and follow-through in designing controls and designing menus and designing interpretive guidelines for tests. He actually talks about deal-making -- about how as the fiscal environment gets tighter you might make tradeoffs between what you’re offering, when you’re offering, and how you’re offering it. And God forbid, particularly for us who are physicians, being humble -- admitting when we’re wrong.
In terms of preparing for this change, institutions will need to be ready with programming. Educational programs will need to be attentive to educational needs and individuals will have to do it. What physicians are supposed to be leading all of us to do in the healthcare profession, and that is to be perpetual learners and also be leaders or at least be outspoken followers if we’re not leaders.
And the straight facts from my perspective, which is probably a biased perspective, is that the IVD industry has been dazzling in promoting technology but perhaps slow to promote the value base for competitive reimbursement, that laboratorians are - have been willing to play with new technology but slow to lead in the use of lab tests and to demonstrate how lab tests can actually impact patient care, can actually - more money spent in the lab might actually mean less money spent in the hospital or the outpatient clinic and that the ultimate healthcare outcomes have been unattended too long, and that physicians have used - physicians again on both side of the aisle -- both in and out side of labs -- (have used) lab tests empirically with tolerance I think for too little evidence.
Again, about a decade ago, Dr. Lundberg, who at that time was the Editor in Chief of JAMA -- now he’s the editor of MedScape, an information internet system -- published an article in which he talked about the need for an outcomes research agenda.
And to quote directly from Dr. Lundberg, “The real reasons for lab testing should be to improve the outcome for the patients or the public health. The time has come to develop a solid research agenda to measure and monitor these vitally important outcomes and to change practices when indicated by the results.”
And I would argue unfortunately that ten years later these words still resonate with vitality. I would probably change it from a solid research agenda to a solid practice agenda, because the technology is out there and it’s in use and we need to get a better hold of it.
At the end of last year, Dr. Sodeman -- Tom Sodeman -- who was the incoming President of the College of American Pathologists, described what he viewed as a need for the laboratory community to take ownership of new technology. And again I’m quoting him verbatim -- “Our future has never looked better, but I do intend to convey some urgency. Our growth potential is intimately linked to new tools, whose station, care, applications are well within our grasp. The market moves as investors like to say, and when it moves to creates momentum. We need to reach out now and take ownership of those tools.” And the point of that is that we do want to ensure we drive technology rather than watch technology drive us.
So there is here an undercurrent call to arms, a call to activism. It is my view that FDA is an important stakeholder, and in part because we hold regulatory tools and we have risk communication systems. That’s the good news. We have the tools. But actually you on the line -- you who represent MedSun and you who represent LabNet -- you actually hold the cards, because you actually understand in the real world how things work.
And certainly on the part of my colleagues who are operating out of the Office of In Vitro Diagnostics and the Office of Surveillance and Biometrics, we’re interested in the bad news when things don’t work well, but sometimes we’re interested in the good news as well -- to understand what is working well and what is good about the menu of tests - the (dazzling) menu of tests that you are using and will continue to use.
There’s a lot of poetry written on little bit medicine and biomarkers. A little bit less on regulation. And I’ll give the final word actually to Robert Frost, because I think there is a Frost-like view to the place where we are at now in our professional lives in our healthcare system.
And so Frost said, “I shall be telling this with a sigh somewhere age and ages hence. Two roads diverge in a wood, and I, I took the one less traveled by. And that has made all the difference.” And I urge you all to take the one less traveled by. And thank you.
Jill Marion:Thank you Dr. Gutman. I will now turn the program over to our coordinator, for our question and answer session.
Coordinator: Thank you. At this time if you would like to ask a question, please press “star one” on your touchtone phone. Please unmute your phone line and record your name clearly when prompted. To withdraw your question, you may press “star two.” And once again if there are any questions or comments, pleas press “star one” on your touchtone phones at this time. One moment while we wait for our first question.
(Participant): Yes, I have a more specific question in terms of some of the newer tests coming out for molecular testing. A lot of these tests - instead of saying use an FDA approved DNA extraction machine and an FDA approved thermocycler, these approvals are coming through so that you have to use a specific brand of extraction and a specific brand of thermocycler.
So it turns out in terms of trying to keep health costs low, you end up with having to purchase three or four machines (that do the) exact same function in order to stay under (IBV) label. Can you please comment on that sort of contradictory force that’s going on there?
Steve Gutman: Yes, I can do that. We obviously can only comment on systems that we’ve seen. So if, you know, we’re from Missouri, we are driven by data, so our clearance or approvals are very holistic. They look at the whole system with all of the instructions and all of the instruments and reagents.
It is actually possible for companies to get more general clearances by bringing products through with appropriate specifications, but the drawback is they actually have to be cleared or approved instruments or at least FDA compliant extraction or instruments or (unintelligible) in some cases reagents in order to be compliant, because if there aren’t FDA kosher materials, then you don’t have any assurance that products are being made under the quality system regs and you actually have no assurance that the actual end product will work.
So what you can do to help FDA is to encourage people, one, to be compliant in terms of meeting quality systems regs and making their products according to good manufacturing practices. You may have small leverage, medium leverage, or large leverage. The more labs that ask the more likelihood that companies will try to become FDA complaint.
And that you also understand that if you actually can - you can wonder. You can modify lab testing and use alternative techniques, but you need to constantly then question what the impact on the validation of the assay is. You have to take responsibility for the non- quality system compliant product and you have an increased opportunity for flexibility and also an increased liability.
(Participant): Thank you.
Coordinator: Thank you. Again at this time as a reminder, if there are any questions or comments, please press “star one” on your touchtone phone. Again, it is “star one” on your touchtone phones if you have any questions or comments you’d like to make at this time. Okay, you do have a question. Your line is open. You may ask your question.
(Participant): What technologies do you see coming out for home care and home health that are used in the hospital?
Steve Gutman: Yes. Well I can’t talk bout products that are specifically in the pipeline, so I can’t talk about products that we’re seeing either in early protocol review or that haven’t been cleared or approved. I can make a sort of general statement -- a general view about what’s going on with home healthcare.
It’s my view that the technologies are becoming increasingly robust so that the option, the opportunity to create a broader range of tests that would actually work in the hands of lay users is possible.
There are two things that I think will actually make this not an incredibly high growth area. In spite of the fact it’s a possible area, because sometimes it’s cheaper -- certainly the access is easier, there’s a powerful convenience factor. And the two things are that FDA’s (requires that) the home product has to have the same performance or equivalent performance in the hands of lay users that you would see in the lab. So the thing has to actually work at home in an uncontrolled environment.
We require that there be labeling so that the patient actually be able to understand not only how to run the test but how to use the test. And then the third, which I think is the most powerful constraint on home use, is that the benefit of having the test has to outweigh the risk.
So glucose meters would be a wonderful example of where there actually is probably some degradation performance in the home environment, but the benefit of being able to make real time (unintelligible) decisions outweighs the colorful background of device failures, operator failures, failure of the strange environments.
There are other tests, tests for influenza might come along for example that might actually be technically feasible to do at home like a pregnancy test, but FDA would probably not think it’s a really great idea for people to start making self diagnosis of influenza without a history of physical exam -- a chest X-ray or a (gram stain) and so there would probably be some reticence to clear a test where it might actually be unsafe as opposed to a safe and effective alternative.
The other thing that is a very powerful modulator on frankly the market itself and probably has cooled the market a little bit is the discovery that people who don’t have training -- people outside of the healthcare network -- don’t really like to stick themselves and actually don’t do a very good job of doing finger sticks at home. And so products that have - that actually do seem to work reasonably well at home are just not best sellers because of the sort of finger stick features.
So I think there will be incremental changes in the menu of tests at home over time. I think it will be slow. I think (there) will be places where convenience or access really does make great sense and I expect it to be somewhat restrained.
Coordinator: Thank you. Our next question has come in. Your line is open. You may ask your question.
(Participant): Thank you. We are a community hospital and one of the challenges that I see is with new advancements. Do you see that there would be replacement of traditional testing due to the addition of new technologies? And then how would you address to the physician changes in their practice in order to incorporate this new technology?
Steve Gutman: Yes, well the history of lab medicine is that there’s not a test that isn’t cherished and easily goes away. So old tests die hard. And then some of the old tests die hard and then they come back again. So if you look at C-reactive protein, for example, it was a cherished test went out of favor and then it popped back up on the screen and has all kinds of new intended uses. So it actually is hard to displace standard of care current technology.
But you know, that’s why I made such a powerful picture of evidence- based medicine. I think that the way to change practice is to go through continuing medical education, continuing nursing education, going through the continuing education programs to work through practice standard, to work through third party payers because they have a lot of punch in terms of what patients will say to their doctors, what doctors may order.
But I think a change in the behavior of doctors is very difficult. I think one of the challenges to the new technology when FDA clears it, when CMS decides to pay for it, is the new technology that is really new doesn’t come necessarily with all of the answers and doesn’t always come without post-market surprises. Again, one of the reasons for the LabNet and the MedSun program is to look at those things that we were just not smart enough to ask up front.
But I’m really upbeat -- actually, perversely upbeat -- that the economic pressures that are being brought to bear will actually force at some point more intelligent decision making, because we can’t afford to just be throwing money out on bad or useless (tests). Many people have already started to figure that out, so it’s truly a question of implementing that into practice.
(Participant): Thank you.
Coordinator: I’m showing no further questions from the phone lines. I’ll turn it back to Miss Marion for any closing remarks.
Jill Marion:Thank you for participating on today’s call. Transcripts are available for today’s presentation. Please call our MedSun toll free 800 number for more information. Once again the number is 800-859-9821. An electronic certificate of participation can be accessed through the registration site.
You must first register and complete a brief evaluation in order to receive the certificate, but we appreciate and welcome your feedback in order to continually improve our educational offerings.
Marilyn Flack: I just want to make sure that Jill gives you the number one more time. It’s the same number for those who might want to join LabNet. So Jill give them the number one more time.
Jill Marion:800-859-9821. Also check with your MedSun Representative for upcoming LabNet programs and particular education offerings planned for the fall, which will enable you to receive continuing education credit.
If you would like to find out how your MedSun Hospital can join LabNet or if you have any other questions, please call our toll free MedSun number, which once again is 800-859-9821. This concludes the Practice of Laboratory Medicine in the 21st Century: Achieving Relevance in a Tidal Wave of Change. Thank you again for your participation.
Coordinator: Thank you. That concludes our call for today. You may now disconnect.