iPathology

Thoughts on Pathology and Laboratory Medicine as a Career

As the Fall rolls on, senior medical students will be choosing their fields and applying for residency programs. It's an exciting and critically important time in their lives. The choices made between now and the Match day will determine the course of their lives.

What about Pathology as a career? Ever thought about it? What do Pathologists do?

Pathology is a very broad field that is commonly said to bridge the divide between basic science and clinical medicine. In fact, it's far more than that. Pathology is an incredibly dynamic area. This may come as a surprise to those who think of pathology as all about looking at slides and doing autopsies. While those are important and essential for patient care, which is the field that is taking the new genomics and applying to medicine? Or the new proteomics? The answer is these are being applied in pathology, particularly in the field known as clinical pathology. That's the field that includes the medical direction of clinical laboratories. It's also called laboratory medicine. A clinical pathologist may practice some anatomic pathology, but most clinical pathologists have a specialty interest in one of the areas of the clinical laboratories such as clinical chemistry, molecular diagnostics, hematology, cytogenetics, coagulation, blood banking, immunology, or microbiology. That's a lot of areas each of which has plenty of depth. Some of them involve primarily laboratory work, others, like coagulation and blood banking involve patient care to one degree or another. Molecular diagnostics is said to be the bridge between clinical pathology and anatomic pathology. Increasingly, tumors are being studied at a molecular level, providing data that complements morphology and offers information for diagnosis and sometimes prognosis. This is where the fruits of the human genome project are being applied. It's a very exciting field that is at the cutting edge of applying science to patient care.

What's the training like to become a pathologist? In the U.S., most people take 4 years of residency in anatomic and clinical pathology in accredited programs and are then eligible to apply to take the American Board of Pathology exams. Subspecialty training is available in academic medical centers around the country. To my mind, what distinguishes programs is usually not the anatomic pathology which tends to be of a uniformly good standard, but the level of involvement with clinical pathology. Does the program have full-time clinical pathologists? Who runs the clinical laboratories? Here are some questions to consider if looking at a pathology program: how many anatomic and clinical pathologists are directly involved in the teaching program? Is there a real interest in clinical pathology or is a part-time activity? What are the opportunities to be involved in cutting edge research? On anatomic pathology rotations, do residents get to see the cases from start to finish? Are the "most interesting" cases reserved for fellows? (A red flag if you're a resident). What is the activity in molecular pathology? If it doesn't exist, I'd look at other programs: it's not a good idea to begin a 30-40 year career in pathology today and not know about molecular pathology.

What are the job prospects like? What do pathologists earn? I'll talk about these more in later note.

For more information see the American Society for Investigative Pathology website.

Survey Sounds the Alarm on Physician Labor Trends

A recent survey conducted by Merritt Hawkins and Associates of physicians aged 50-65 reveals several striking and alarming findings. But before looking at the survey findings, let's remind ourselves of a few important points. First, this group of physicians is presumably at the peak of their careers in terms of clinical experience. Second, nearly one-half of all U.S. physicians are over 50 years of age, so it's a very large group of professionals. Third, the U.S. population is rapidly "graying" and the need for physicians will never have been greater than it will be in the coming decades. We face a physician shortage that may range from 100,000-200,000 physicians by 2020.

So here are some of the findings that are very troubling.

1. In the next 1-3 years, nearly one-half of physicians surveyed plan to either retire, take a new (non-clinical) job, or cutback significantly on their clinical work.

2. Only 10% agree that their work is "very satisfying".

3. 18% have already closed their doors to new patients.

4. A clear majority (57%) would not recommend a career in medicine to their own children.

Unfortunately, there are no specific data on pathologists in the survey, but in my personal experience, I'd guess that the degree of job satisfaction among pathologists is significantly higher than among primary care physicians. I think this is attributable to many reasons among which I'd cite more regular hours, less paperwork, a more structured working environment, and possibly greater intellectual stimulation given the dynamic developments in laboratory testing.

If the findings of this survey are accurate, the U.S. is going to face a crisis in physician health care delivery in the next few years. The reasons for the high level of dissatisfaction have to be addressed.

The Merritt Hawkins report can be accessed here.

Should HPV Testing Replace Cytology for Screening for Cervical Cancer?

This week's New England Journal of Medicine (357:1579-88, 2007) reports a very important study that looks at the performance of HPV molecular testing for screening for grade 2 and 3 (i.e., high-grade) cervical intra-epithelial neoplasia. The population consisted of over 10,000 women aged 30-69 in Canada with good access to health care. Sensitivities for HPV and cytology were 94.6% and 55.4%, respectively. Specificities were 94.1% and 96.8%.

Cervical cancer screening by the use of the Pap smear has been one of the best examples of a successful cancer screening program in history. With current practice, it's common that HPV testing is used to clarify cases of ASCUS (identified by cytology/cytopathology) rather than as a primary screening technique. This study (and others) suggests that HPV testing could be a primary screening modality, although more data are needed to reach that conclusion with certainty.

From a laboratory perspective, screening programs need high sensitivity, and the difference between molecular HPV testing and cytology is striking. On the other hand, there is variation in the number of HPV types that can be detected by different molecular methods and the method used in this study may not be the most sensitive. Different assays have different sensitivities and specificities (see J Clin Micro 45:2130-7, 2007 for example). This study demonstrates that in vitro molecular testing can, in certain situations, be a viable alternative to microscopy. This last point emphasizes the need for pathologists to be closely involved with molecular pathology and its implementation in hospital laboratories or they risk being marginalized in the changing diagnostic landscape. For many cytology technicians and cytopatholgists, Pap testing (or one of its liquid counterparts) is the major professional activity of the daily workload. A widespread change from morphology to in vitro testing for HPV could pose the possibility of sudden and substantial disruptions to professional activities.

Microsoft and Health Care Software

Writing in the Wall St. Journal on October 7, 2007, Bill Gates rightly points out the significant deficiencies with the state of health care informatics and its adverse effect on the quality of patient care. What might a company like Microsoft offer to this field? I submit that Microsoft is exactly what we need to make some sense out of the mess we face in health care informatics. We need a common software product that is as ubiquitous as Word or PowerPoint. We need someone with deep pockets like Microsoft to make the initial investment, to implement standards that have been agreed upon such as HL7, and to introduce some meaningful competition into a field that needs more of a driving sense of the importance and urgent need for innovation. Microsoft has been criticized by many for lacking a spirit of innovation in its flagship products, Windows and Office, but the truth is that they work as expected for most users, and it would be difficult to imagine how global businesses could function without them. If only the same could be said of health care software. Microsoft's entry to this field is to be welcomed.

Board Recertification Should be All or None

American specialty Boards play an essential role in setting standards for basic medical competence for graduates of residency programs. Historically, passing a Board exam was the final milestone to beginning a lifelong career in a medical specialty. In the last 10 to 20 years, however, Boards began to issue certificates which were time-limited, generally to 10 years. Under this scheme, a two-tier system of physician certification is established, namely that candidates who have passed the board exam more recently hold a time-limited certificate whereas physicians who took and passed the boards years or even decades earlier enjoyed the privilege of lifelong board certification. This dichotomy is not only anomalous but clearly unfair to more recent graduates of residency programs. There is no reason to believe, for example, that more recent graduates of residency programs are any less up to date with current medical practice than are their more senior colleagues. In fact, one could well believe that the opposite is true. For many physicians, there is a tendency toward specialization and therefore an increasing distance from the knowledge base of some areas of practice.

Given the rapid advances in biomedical research, it is probably a good idea to require periodic recertification of all physicians in their specialty area, and 10 years seems to be an appropriate interval. Ensuring that physicians are up to date with their field is consistent with the current trends towards emphasis on quality in the delivery of healthcare. This means that for a typical physician, he or she would be required to re-certify about three or four times in the course of a career. For those physicians who hold lifelong certificates, recertification in pathology is currently optional. Whether it remains that way is uncertain. It is entirely possible that hospital staffing committees or payers will require that all practicing physicians undergo periodic recertification if it is available. One point seems clear: if physician recertification is going to be required, it should be required for all physicians.

The Mislabeling Problem

One of the most common problems affecting patient safety in the clinical laboratory is the receipt of samples that are mislabeled. Despite all the advances in analytical performance, informatics and so on, it seems that getting the right name on the sample in a consistent fashion still eludes the healthcare enterprise. Why is this? Frequently the problem begins at the time of specimen collection when the phlebotomist is required to match a sample with the patient's name. This kind of task invariably poses the risk of human error. So many problems can arise in this situation including carelessness, haste, language difficulties between the patient and the phlebotomist etc. Even the use of bar coding of patient wristbands for automated reading by label printers does not seem to totally eliminate the risk. In the laboratory itself, there are ample opportunities for specimens and patient names to become decoupled. Surgical pathology slides can be labeled with the wrong patient identifier with potentially serious results for patient safety. Perhaps the area that poses the greatest immediate risk is the blood bank. There are few procedures that the clinical laboratory can perform that have greater potential to immediately jeopardize patient health than administration of an incorrect blood unit. It seems that if a choice of labs is to be made for implementing newer methods to eliminate the mislabeling problem, the blood bank should be among the highest priorities.

The Effect of Disease Prevalence on the Predictive Value of Diagnostic Tests

This is an updated version of a note I first wrote in 1999. Because it has been cited in several publications, I am posting it again at the ipathology.com website. 

Standard Metrics of Diagnostic Tests 

When considering the diagnostic utility of tests to distinguish persons with disease from normal persons, reference is frequently made to common metrics, namely, sensitivity, specificity, positive predictive value, and negative predictive value of a test. Traditionally, these concepts are explained with a simple matrix to define the terms true positive (TP), false positive (FP), true negative (TN), and false negative (FN) as follows. 

 

	Test Positive	Test Negative
Disease	TP	FN
Normal	FP	TN

 

Using these terms, the following metrics can be defined. 

Sensitivity = TP/(TP+FN) 

Specificity = TN/(TN+FP) 

Positive predictive value (PPV) = TP/(TP+FP) 

Negative predictive value = TN/(TN+FN). 

 

What is not so clear from these definitions is the effect of disease prevalence on the predictive values of test results, particularly the more commonly-used positive predictive value. Although standard textbooks of clinical pathology make reference to the fact that disease prevalence influences the positive predictive value, this relationship is seldom explained. 

Effect of Overlapping Values 

Given the definition of PPV above, it is apparent that this metric is a function of the TP and FP values. If there were no false positives, the PPV would be TP/TP or 100%. This would occur if the test could reliably distinguish normal persons from persons with disease without any overlap of values. However, this is rarely the situation with laboratory results that fall on a continuous scale. Almost always there is some degree of overlap between results from normal and diseased subjects. For example, low levels of blood markers of myocardial infarction can be detected in healthy subjects. These may overlap the lower range of values found in patients with MI. When there is such overlap, some cut-off point must be established to distinguish a positive from a negative result. It is worth noting that even results which are reported as "postivie" or "negative" by analytical instruments are generated from continuous scales by using a cut-off point. An example of this is the reporting of HIV-1 antibody screening tests. In a popular assay system, these are detected by ELISA reactions which result in changes in absorbance as measured by a spectrophotometer. The lower end of absorbance readings in HIV-1 positive patients overlap those of normal subjects. The computer in the analytical instrument defines a cut-off point of absorbance readings and then reports the results as positive or negative based on the cut-off value. Given that most assays show some overlap of values between normal and diseased subjects, it is apparent that it is very unusual for assays to have no FP results. 

Prevalence of Disease Influences Positive Predictive Value 

There is another situation in which there are no false positive results, namely, if everyone in the population had the disease. In this situation, every positive result would be a true positive. There could be no false positive results, and the PPV would be 100%. Conversely, if no one in the population had the disease, every positive result would be a false positive. There could be no true positives, and the PPV would be 0%. This leads us to conclude that the disease prevalence influences the PPV by influencing the true positive and false positive rates. 

Bayesian Analysis to Determine Positive Predictive Value 

The effect of disease prevalence may be determined by performing a Bayesian analysis. Consider a disease with a prevalence of 10% in a given population. A diagnostic test for the disease has both a sensitivity and a specificity of 95%. What is the positive predictive value of the test? 

Using Bayesian analysis we can determine the likelihood that a person with a positive test result has the disease as follows. For those not familiar with Bayesian calculations, the steps are outlined here and shown in the following tables. 

The two mutually exclusive possibilities are that the person has disease or is normal (row A). The a priori probability of having disease is 10% (0.1) and of being normal is 90% (0.9) (row B). If a person has the disease, the probability of having a positive test result is 0.95. If a person is normal, the probability of having a positive test result is 0.05 (row C). These probabilities are obtained from the sensitivity and specificity values above. If the sensitivity is 95%, then FN is 5%. If the specificity is 95%, then FP is 5%. 

The joint probabilities are the products of multiplying values in rows B and C (row D). The final probability for each of the two mutually exclusive states is obtained by dividing each value in row D by the sum of the two values in row D. This is shown in row E. 

 

A		Disease	Normal
B	A priori probability	0.1	0.9
C	Probability of a positive test result	0.95	0.05
D	Joint probability	0.1 × 0.95 = a	0.9 × 0.05 = b
E	Final probability	a/(a+b) = 0.68	b/(a+b) = 0.32

 

In this case, with a disease prevalence of 10%, a person with a positive test result has a 68% probability of having disease, and a 32% probability of being normal. 

In a population in which the disease prevalence is 0.1% (0.001), the calculations are as follows: 

 

A		Disease	Normal
B	A priori probability	0.001	0.999
C	Probability of a positive test result	0.95	0.05
D	Joint probability	0.001 × 0.95 = a	0.999 × 0.05 = b
E	Final probability	a/(a+b) = 0.019	b/(a+b) = 0.981

 

In this case, with a disease prevalence of 0.1%, a person with a positive test result has a 1.9% probability of having disease and a 98.1% probability of being normal. 

The effect of disease prevalence on the positive predictive value for a test with sensitivity and specificity of 95% is shown in the following table. 

 

Prevalence	PPV
0.1%	1.9%
1%	16%
10%	68%
20%	83%
50%	95%

Conclusion

The prevalence of a disease in the population has an important influence on the positive predictive value of a diagnostic test. With increasing disease prevalence, the more likely it becomes that a person with a positive test result has the disease, and the less likely it becomes that a positive result is a false positive. The effect of disease prevalence can be easily determined by Bayesian analysis.

 

Copyright, © Anthony A. Killeen, 1999-2007. All rights reserved.

Molecular Testing for Cardiac Rejection

Believe it or not it's been 40 years since the first human to human heart transplant was performed by Christiaan Barnard in South Africa.  The success of this procedure is now attested to by the fact that over 2000 patients in the United States receive a heart transplant each year with an estimated 50% 10 year survival.  In 1967 the first patient to receive a heart transplant survived only 18 days.

 

One of the main medical concerns following heart transplantation is rejection of the organ by the recipient immune system.  A new test, AlloMap, is now available that can detect the presence of cardiac rejection noninvasively [Starling RC et al., J Heat Lung Transplant, 25:1389, 2006].  The test is based on the measurement of expression of certain genes in peripheral blood mononuclear cells.  A panel of 11 genes is reliably informative for the identification of patients who have a very low risk of organ rejection.  While the negative predictive value of the test is extremely high the positive predictive value is quite low, in part because of the fact that with modern immunosuppressive therapies, relatively few patients develop acute cellular rejection particularly after more than a year post-transplantation. It is possible that the new test will reduce the number of cardiac biopsies that are performed posttransplantation.  If so, this will be probably be welcomed by patients.

Presumably the AlloMap test will be applied to other organ transplants such as liver, kidney and lung.  At the time of the first human heart transplant the procedure seemed extraordinary.  In many ways, the ability to detect organ rejection by gene expression profiling in a blood sample seems equally extraordinary.