Imagine that the 2020 US Census has been completed and the results show your state has lost 8% of its population and now has only 13 million residents. Your state loses several congressional representatives, federal funding for hospitals and school lunch programs is reduced, and new businesses begin to pass over your state in favor of states that are expanding. Several years later, an audit is performed, and the previous census numbers were found to be incorrect; instead of having a decrease in population, your state actually had a substantial gain in the number of residents. The state in fact now has 20 million residents, not 13 million. However, by this point, the state has already experienced the consequences of the error.
Although the above scenario may seem improbable, a similar situation was found in the assessment of the pathology workforce, as described in this issue by Robboy et al.1 The authors conducted an analysis of data that they obtained from national organizations and state medical board registries after a recent report2 indicating an apparent substantial decrease (8.3%) in the number of US pathologists between 2012 and 2017. The authors discovered a discrepancy between these data, which relied on readily available data from the Association of American Medical Colleges (AAMC) and data maintained by the American Medical Association (AMA), known as the AMA Physician Masterfile, which is generally available only through third-party (commercial) vendors. Although the published AAMC data are ultimately derived from the AMA Masterfile, it listed 12 839 pathologists in the workforce for 2017, whereas the AMA listed 19 923—50% more than the AAMC’s number.
Through examination of the data sources, the authors discovered that the AAMC data had excluded most pathologists with an associated subspecialty in the AMA Masterfile; for instance, 1618 pathologists were excluded because they were listed as practicing cytopathology and not categorized under a primary training category, such as anatomic pathology. The authors note that it has become commonplace for pathology residents to undergo subspecialized fellowship training, which likely magnified this phenomenon. Unfortunately, these AAMC data have been frequently used as the basis of several publications and studies, likely because of the greater availability to those interested in analyzing workforce data. Of importance, the authors note that a similar method may be used in reporting the workforce data for other specialties. For instance, they discovered a similar apparent decrease in the number of active general surgeons between 2012 and 2017 according to the AAMC data; is this a reflection of reality or an artifact of some subspecialty surgeons being excluded?
The authors found an additional discrepancy in workforce data reported by the Accreditation Council for Graduate Medical Education (ACGME) that concerns the classification of physicians based on subspecialty and disregards a physician’s primary specialty. For example, all dermatopathologists are classified as dermatologists in the ACGME data set regardless of whether their primary certification was from the American Board of Pathology (ABPath) or the American Board of Dermatology. However, as the authors note, data from ABPath indicate that 61% of the 1166 dermatopathologists trained since 2004 hold primary certification from ABPath; it would be more appropriate to classify these physicians as pathologists rather than as dermatologists if only a single option can be chosen. This same example suggests that the ACGME method affects other medical specialties as well.
The number of physicians in a workforce is a critical variable in workforce modeling, which can be used to determine whether there are enough physicians as the amount of work changes in future years. In a previous study, Robboy et al3 described a comprehensive model of the pathology workforce in which variables such as patient needs and population size can be altered to assess whether increases in the pathology workforce will keep up with future demand. However, the reliability of the model’s output depends on the reliability of pathology workforce data at times of known demand. Use of the AAMC data, which has undercounted the number of pathologists needed for a given amount of work, would lead to a much lower predicted number of pathologists needed for a given increase in population and/or complexity of work.
Perceptions of the pathology workforce are highly individualized. To a graduating trainee, the workforce may seem oversaturated if they cannot find work in their hometown, even if numerous jobs exist elsewhere in the same state. To those seeking to hire pathologists, the workforce may seem insufficient if they cannot find a pathologist qualified for specific job requirements, even if numerous pathologists have applied for the position. To a neurosurgeon, the pathology workforce may seem inadequate if a neuropathology-trained pathologist is unavailable to read frozen sections, even if the on-call pathologist is sufficiently trained and experienced to read such specimens. To a patient living in a rural area, the pathology workforce may seem understaffed if an onsite evaluation for adequacy is unavailable, requiring them to return for a subsequent biopsy. The quality, geographic distribution, and adaptability of the workforce are important yet complicated factors that could eventually be captured in advanced workforce models. The recent severe acute respiratory syndrome coronavirus 2 outbreak has provided an extreme example of the need for physician adaptability; for many pathologists, this has meant a rapid shift in focus to laboratory testing, molecular microbiology, immunology, and transfusion medicine.4
Accurate workforce modeling along with quantitative and objective data are critical for a proper workforce assessment and can help avoid a data vacuum in which anecdotal evidence drives perception.5 Furthermore, the publication of inaccurate data can be misleading to medical students who are determining their desired medical specialty. To be successful, an accurate assessment of a physician workforce requires a consistent, long-term commitment from physicians and physician organizations. Physician organizations need to determine accurate methods for assessment, execute assessments in a standardized manner, continually evaluate the accuracy of their assessments, and engage physicians and other stakeholders, some of whom may not be organizational members, to participate in the assessment process. The process should be conducted in an objective fashion and involve a diverse group of stakeholders to avoid bias. Physicians entering a specialty should be taught the importance of workforce assessment and be asked to commit to career-long participation. Ideally, the process would remain nonintrusive and streamlined to avoid further physician burnout and detachment. Although no census can be entirely accurate, quality improvement processes can and should be established to identify and subsequently correct inaccuracies large enough to have a meaningful effect.
Published: July 16, 2020. doi:10.1001/jamanetworkopen.2020.10908
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 VandenBussche CJ. JAMA Network Open.
Corresponding Author: Christopher J. VandenBussche, MD, PhD, Department of Pathology, The Johns Hopkins Hospital, 600 N Wolfe St, Pathology Bldg, Room 406, Baltimore, MD 21287 (firstname.lastname@example.org).
Conflict of Interest Disclosures: Dr VandenBussche reported receiving royalties from Springer Publishing and Wolters-Kluwer, receiving a research money award to The Johns Hopkins University School of Medicine as a subcontractor of a National Cancer Institute Small Business Innovation Research award to Thrive Earlier Detection, and receiving speaker fees from Ebix Continuing Medical Education.
Additional Contributions: Daniel C. Ehlke, PhD, SUNY Downstate Health Sciences University School of Public Health, Brooklyn, New York, provided valuable input. He was not compensated for his work.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
VandenBussche CJ. Discrepancies in the Reporting of Pathology Workforce Data—Is Reevaluation Also Needed for Other Medical Specialties? JAMA Netw Open. 2020;3(7):e2010908. doi:10.1001/jamanetworkopen.2020.10908
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: