Pathology residents and fellows tailor their training and job search strategies to an actively evolving specialty in the setting of scientific and technical advances and simultaneous changes in health care economics.
To assess the experience and outcome of the job search process of pathologists searching for their first non-fellowship position.
The College of American Pathologists (CAP) Graduate Medical Education Committee has during the past 5 years sent an annual job search survey each June to CAP junior members and fellows in practice 3 years or less who have actively searched for a non-fellowship position.
Job market indicators including job interviews, job offers, positions accepted, and job satisfaction have remained stable during the 5 years of the survey. Most survey respondents who had applied for at least 1 position had accepted a position at the time of the survey, and most applicants who had accepted a position were satisfied or very satisfied. However, most attested that finding a non-fellowship position was difficult. Despite a perceived push toward subspecialization in surgical pathology, the reported number of fellowships completed was stable. Respondent demographics were not associated with job search success with 1 significant exception: international medical school graduate respondents reported greater perceived difficulty in finding a position, and indeed, fewer reported having accepted a position.
Pathology residents and fellows seeking their first position have faced a relatively stable job market during the last 5 years, with most accepting positions with which they were satisfied.
The search for an initial job following training is an active and ongoing concern to pathology residents and recent graduates, to medical students considering pathology residency, to pathology educators including program directors and chairs, and to prospective employers. However, there is a lack of longitudinal data regarding the job search experience of recent pathology training program graduates. A model of anticipated workforce needs predicts a shortfall of pathologists starting in 2015 owing to a combination of anticipated retirements and an aging population with increasing health care needs.1 Despite this predicted shortfall, there is a perception among pathology residents and fellows that the job market is tight, with few job openings. Furthermore, following a decrease in the length of anatomic pathology/clinical pathology (AP/CP) board eligibility training requirements from 5 to 4 years in 2005, there was significant growth in pathology residents completing fellowship training. It is unclear whether this growth is a reaction to the reduction in training by residents responding to a perceived need for more experience,2 or whether it is a result of postponement of the job search due to a lack of available positions. There is a need for longitudinal data to monitor trends in the job seeking experience of pathology residents and how this correlates with demographic, educational, geographic, and training characteristics.
The College of American Pathologists (CAP) Graduate Medical Education Committee (GMEC) is charged to “assess preparation of residents and recent graduates for existing and evolving practice settings.” The GMEC is positioned to assess longitudinal changes in the pipeline of newly graduated or soon-to-graduate pathology residents and fellows through surveys distributed to targeted subsets of CAP members. In 2011 the GMEC constructed a survey that measures job search experience and outcome of residents/fellows and recent graduates of training who have been actively searching for a job. With slight modifications, the survey was distributed annually from 2012 to 2016, allowing robust longitudinal comparisons over a 5-year period.
METHODS
The CAP GMEC conducts an annual survey of recent graduates in early June of each year. To reach the target population, an email invitation to complete a voluntary, anonymous, online survey is sent to CAP junior members and fellows in practice 3 years or less. Two initial screening questions ask if the respondent actively searched for a non-fellowship position during the prior year, and if the position sought was the first position since completing residency/fellowship training. Respondents selecting “no” on either question are screened out of the survey. To make year-to-year job market comparisons possible, the GMEC strived to maintain consistency in survey content across administrations. Since 2012, the survey instrument has remained stable with regard to the core survey questions reported in year-to-year comparisons. In 2015–2016, questions were added on age and gender to aid in demographic analysis. Supplemental digital content, containing 3 supplementary documents: (1) Survey Questions, (2) Demographics (Tables 1 through 6), and (3) Job Search Success (Tables 1 and 2) is available at www.archivesofpathology.org in the April 2018 table of contents. The text of the survey questions used to collect the data reported here is available as Supplementary Document 1. Statistical analyses were performed by using IBM SPSS Statistics for Windows, version 23.0. (IBM Corp, Armonk, New York) as indicated in the text.
RESULTS
Survey and Respondent Characteristics
The survey was directed to CAP junior members and fellows, and as with any voluntary survey a self-selected subset chose to complete it. While the survey is not a population-based sample, the pool of respondents is stable over time and we can therefore make inferences about year-to-year changes. Overall survey respondent demographics are provided in Table 1 and are in line with those reported for incoming pathology residents in the 2015–2016 Accreditation Council for Graduate Medical Education (ACGME) Data Resource Book.3 United States medical school graduates (USMGs) include graduates of US MD and DO programs; international medical graduates (IMGs) are graduates of international medical schools, including US citizen graduates of Caribbean medical schools. Year-to-year demographic characteristics were stable and are provided in full in Supplementary Document 2 as follows: Table 1, Residency Training; Table 2, Medical School; Table 3, Gender; Table 4, Age; Table 5, Gender and Job Search Success; Table 6, Age and Job Search Success. The demographics of survey respondents therefore approximate those of pathology residents overall, with relative underrepresentation of IMGs among survey respondents.
Of note, this survey is a snapshot in time that captures respondents at different stages of their job search, and it does not capture the end results of the job search process for many respondents. In the last 2 years of the survey (2015–2016) we included a question regarding how long the respondent had been actively searching for a job (Table 2). Forty-five percent (n = 33 of 74) of respondents who had not accepted a position at the time of the survey reported that they had been searching for a job for 6 months or less. Among those who had accepted a position, most (61%, n = 119 of 196) reported that they had been searching for 6 months or less, with another 32% (n = 62 of 196) reporting accepting a position 6 to 12 months into their job search. We cannot exclude that survey respondents may report their job search length differently based on whether or not they had accepted a position at the time of the survey.
Survey Responses Indicate Stable Job Market Indicators For 5 Years
Most respondents across years, 86% (n = 646 of 752), had been invited to at least 1 job interview (Table 3). Despite some year-to-year variability, the percentage of survey respondents accepting a position has remained stable. At the point in time when they took the survey, 69% (n = 517 of 752) of survey respondents had accepted a position. Just under a third (31%, n = 232 of 752) of respondents across years received multiple job offers, whereas 42% (n = 314 of 752) received a single job offer. Eighty-nine percent (n = 448 of 505) of respondents who had accepted a position reported that they were satisfied or very satisfied with the position accepted (the degree of satisfaction remained stable over time; Table 4). While most respondents accepted positions in a nonacademic setting, 38% (n = 187 of 498) accepted an academic position. A 2-way χ2 test confirms the proportion of respondents accepting academic versus nonacademic positions has remained stable across years: χ24 = 3.333, P = .50.
Despite their success in being invited to job interviews and in receiving and accepting job offers that proved satisfactory, 80% (n = 409 of 509) attested that finding a non-fellowship position was moderately difficult, difficult, or very difficult. The degree of perceived difficulty was quite stable during the 5 years of the survey with the exception of a small statistically significant decrease in perceived difficulty in finding a non-fellowship position in 2016 as compared to 2014 (Table 5). Survey participants who reported difficulty finding a position were provided with a checklist of reasons for the perceived difficulty (Supplementary Document 3, Table 1: Reason for Perceived Difficulty Finding a Position). The overwhelming majority (93%, n = 216 of 233) stated that too few jobs were available; 35% (n = 81 of 233) also attested to being too limited in their geographic preference, whereas 16% to 17% attested to a mismatch between training and job requirements (n = 38 of 233), or inadequate training/experience (n = 40 of 233).
Fellowship Training and Specialization
Among survey respondents, 91% (n = 700 of 767) had completed at least 1 fellowship (Table 6) and this percentage was stable over the years that the survey was administered. Also stable was the significant minority of respondents, 25% (n = 195 of 767) on average, who had completed 2 or more fellowships. Among survey respondents who responded to a checklist question regarding the reason for completing 1 or more fellowships (n = 252), most reported doing so “To deepen my knowledge/skill in a subspecialty area of personal interest” (81%, n = 203 of 252) and/or “To enhance my CV/marketability prior to looking for a non-fellowship position” (71%, n = 180 of 252). By contrast, very few reported a direct relationship between fellowship enrollment and job offers (“Received a job offer that was contingent upon my completing a fellowship”: 8% [n = 21 of 252]; “I could not find a non-fellowship position”: 8% [n = 19 of 252]).
Six-hundred ninety-nine survey respondents reported completing a total of 934 fellowships. Sixty-one percent (n = 570 of 934) of fellowships completed were in subspecialty areas in which the American Board of Pathology (ABP) issues subspecialty certificates, or ABP-certifiable fellowships (Table 7). Cytopathology (reported by 25% of respondents, n = 172 of 699) and hematopathology (reported by 23% of respondents, n = 160 of 699) were the most common ABP-certifiable fellowships reported by survey respondents, which is in line with ACGME data.3 Among non–ABP-certifiable fellowships, most were surgical pathology related; 51% (n = 187 of 364) were in non–organ-specific surgical pathology (general or surgical/oncology); and 39% (n = 142 of 364) in organ-specific surgical pathology (breast, cardiovascular, gastrointestinal, genitourinary, gynecologic, pulmonary, renal, soft tissue), with the most frequent being gastrointestinal pathology (reported by 8% of respondents, n = 56 of 699). There was no trend toward an increase or decrease in organ-specific or non–organ-specific surgical pathology fellowships over the course of the survey (data not shown).
Job Seeker Characteristics and the Job Search Experience
We next sought to assess for any association between job seeker characteristics and the job search experience; except as indicated, comparisons are performed by using the 2-way χ2 test. There was no difference in the likelihood of having accepted a position according to gender or age (gender: χ21 = 0.3, P = .57; age [25–34 versus 35–44 versus 45–54 years]: χ22 = 2.8, P = .25). Both survey respondents who had and had not accepted a job reported flexibility in terms of geographic areas, with 60% (n = 433 of 735) endorsing looking for positions in at least 2 geographic areas (Supplementary Document 3: Table 2, Geographic Flexibility and Job Search Success). Those who had not accepted a position reported interest in slightly more geographic regions than those who had already accepted a position (mean, 3.5 versus 2.7; t(375.7) = −3.8, P < .001, independent samples t test).
Eighty-three percent (n = 642 of 769) of respondents had completed combined AP/CP training. There was no significant difference in job search outcome according to the type of residency training; 70% (55 of 79) of AP only–trained, 70% (26 of 37) of CP only–trained, and 69% (422 of 610) of AP/CP-trained respondents reported that they had accepted positions. Respondents who had not completed a fellowship had a significantly lower rate of accepting a position (43%, n = 21 of 49) than those completing 1 fellowship (70%, n = 341 of 487) or 2 or more fellowships (74%, n= 141 of 191). There is a statistically significant relationship between fellowship completion status and job search success: χ22 = 18, P < .001. However this effect is limited to the small group of respondents (9%, n = 67 of 767) who had not completed a fellowship, and completing 2 or more fellowships was not associated with a substantially higher rate of job search success than completing 1 fellowship. Respondents' subjective impression of difficulty in finding a job likewise did not differ by number of fellowships completed (1-way analysis of variance, F2,500 = .95, P = .40). There was a statistically significant association between specific fellowships and practice setting. Respondents who accepted positions in nonacademic settings were notably more likely to have completed a cytopathology fellowship (29%, n = 86 of 296 versus 16%, n = 29 of 180; χ21 = 10, P < .001). Respondents who accepted positions in academic settings were notably more likely to have completed a molecular genetic pathology fellowship (14%, n = 26 of 180, versus 2%, n = 6 of 296; χ21 = 27, P < .001).
The job search experience differed significantly for graduates of USMGs as compared to IMGs (Table 8). International medical graduates make up a significant proportion of pathology residents, ∼40% according to ACGME data,3 and also make up a substantial minority of survey respondents (27%, n = 204 of 760). Most IMG survey respondents did report accepting a position, but the percentage was significantly lower than among USMGs (63%, n = 119 of 190, versus 73%, n = 376 of 516; χ21 = 6.9, P = .008). While similar percentages of IMGs as USMGs had accepted an academic position at the time of the survey, (29%, n = 55 of 190, versus 25%, n = 128 of 516; χ21 = 1.0, P = .30), significantly fewer had accepted a nonacademic position (33%, n = 62 of 190, versus 47%, n = 241 of 516; χ21 = 11.2, P = .001). Our survey did not distinguish US citizen or permanent resident graduates of international medical schools from noncitizen, non–permanent resident IMGs; ACGME data indicate that about three-quarters of pathology resident IMGs are in the latter category.3
Job Search Outcome of International Medical Graduates (IMGs) and US Medical Graduates (USMGs)a
We next assessed for differences between IMGs and USMGs in quantified characteristics of the job search; a 2-sided independent samples t test was applied. International medical graduates reported applying to significantly more advertised positions than did USMGs (mean = 6.2 versus 4.9, t(701) = −3.996, P < .001), and similar numbers of unadvertised positions (mean = 2.5 versus 2.6, t(697) = .439, P = .70). International medical graduates also selected significantly more regions when asked in which geographic regions they were most interested in finding a position than did USMGs (mean = 3.3 versus 2.8, t(272.938) = −2.209, P = .03). International medical graduates attested to more difficulty in finding a position, with 32% (n = 38 of 119) of IMGs rating the process difficult as compared to 21% of USMGs (n = 78 of 376); mean ratings were equal to 3.8 and 3.4, respectively (t(493) = −3.302, P = .001). Among USMGs, applicants who had accepted a position reported applying to slightly more positions overall (mean = 6.5 versus 5.6, P = .046) within a more geographically restricted area of preference (mean = 2.6 versus 3.5 regions, P = .002) than USMGs who had not accepted a position. Among IMGs, there was by contrast no difference among successful and unsuccessful applicants in the number of total and advertised positions applied for or number of geographic areas of preference.
DISCUSSION
The CAP GMEC has for 5 years followed the job search experience of first-time non-fellowship job seekers through surveys administered near the end of the academic year. This yearly snapshot of the initial job search process reveals remarkable stability in job market indicators. Most respondents had been searching for a job for less than a year; nevertheless most respondents had accepted a position at the time of the survey, and of those, most were satisfied or very satisfied with their position. Employer data regarding recent hiring and hiring plans mirror these positive findings. The 2016 CAP Practice Leader Survey reported that more than 40% of the practices surveyed were seeking to hire 1 or more pathologists in the prior year, 63% of practices planned to hire in the next 3 years, and practices were filling some or all of their advertised positions in most cases.4 Practices were also optimistic looking forward; a separate CAP survey found that 85% of employers planned to hire at least 1 new-in-practice pathologist in the next 5 years.5
Of note, our survey of first-time job seekers represents a snapshot in time; many of our survey respondents were less than 6 months into their job search. Longer-term data on job search success of pathology residency program graduates are available from the Association of Pathology Chairs (APC). In 2013, the APC polled members regarding the number of graduates of their residency program in the years 2008–2012 who were known to have ever begun a “real” position in pathology, based on personal knowledge, receipt of credential verifications, or other sources. In May of the first, second, and third year after residency graduation, 32%, 73%, and 92% (n = 124 of 388, 262 of 357, and 329 of 359), respectively, of residency graduates had begun their first pathology job (R.D.H., S.Z.P., Wesley Y. Naritoku, MD, PhD, Stephen W. Black-Schaffer, MD, unpublished data from PRODS Workforce Survey 2013; written communication, January 31, 2017). Most pathology residents complete at least 1 fellowship after residency graduation and a significant subset of residents complete 2 fellowships. These data therefore indicate that by May of the year after their terminal training year most pathology residents will likely have accepted a non-fellowship position. The proportion of survey respondents accepting academic positions likely exceeds the percentage of practicing pathologists in academic practice. While this may reflect selection bias in our pool of survey respondents, it may also reflect a trend among pathologists to first accept an academic position and then enter the nonacademic job market when they have gained some experience.
In our survey of first-time non-fellowship pathology job seekers, the likelihood of having accepted a position did not differ according to demographic characteristics (age and gender), but did differ significantly according to whether the respondent was a graduate of an international as opposed to a US medical school. A significantly lower percentage of IMGs reported having accepted a position at the time of the survey. This difference was accounted for by fewer nonacademic positions accepted by IMGs. International medical graduates did not apply to fewer positions or restrict their search more geographically. Not addressed in this article are other potential factors related to job search success, including but not limited to size and location of residency program, access to and involvement in informal networking opportunities, whether expectations match between employer and job applicant regarding scope of practice, degree of subspecialization, workload, and full-time versus part-time status, characteristics of residency programs attended such as geographic location or program size, and lack of data on visa status and the possible effect of need for a J-1 visa waiver on job search difficulty and outcome.
Pathology is a relatively small specialty, accounting for 2% of all US first-year residents in 2015/2016.3 Given the smaller number of positions in any given area at a particular time, a certain degree of geographic flexibility is a necessity for many in looking for a non-fellowship position. This was reflected in our survey in that most survey respondents did indeed indicate that they were interested in positions in at least 2 broad geographic areas. Nephrology is also a smaller specialty with a high proportion of IMGs, 65%, as compared to 40% for pathology. A recent survey of nephrology fellows revealed a similar discrepancy in the job search experience of nephrology IMGs and USMGs, with nephrology IMGs reporting greater difficulty finding a satisfactory position than did nephrology USMGs (73% versus 43%).6 Geographic constraints also surfaced—54% of nephrology fellows said there were few or no local practice opportunities. Internal medicine as a specialty is relatively large; nevertheless in a recent survey reflecting the experience of final year medicine residents,7 most respondents reported starting a serious job search 1 year or more before completion of training, which is comparable to or longer than the job searches reported here based on pathology resident and fellow responses.
Most respondents to our survey continue to complete at least 1 fellowship, and those who had not completed a fellowship were significantly less likely to have accepted a position. By contrast, completion of 2 or more fellowships was not associated with a higher proportion of positions accepted or less perceived difficulty in finding a position. The number of fellowships reported by survey respondents was stable during the 5 years of the survey, in line with stable ACGME-accredited fellowship enrollment over the same period.3 Similar data are not available on non–ACGME-accredited fellowships. Cytopathology and hematopathology each represent approximately a third of ABP-certifiable fellowships reported, which matches a corresponding demand for hematopathology and cytology expertise in the CAP Practice Leader Survey.4 Non–ABP-certifiable fellowships, particularly surgical pathology and its subspecialties, are also very common, making up about a third of the completed fellowships reported in the survey; this matches a demand among potential employers for expertise in general pathology. These so-called selective pathology fellowships allow subspecialty training in an environment where such focused and flexible training is in demand by pathology employers and clinicians.8 The most common organ system–specific subspecialty was gastrointestinal pathology, which is in line with the stated need for gastrointestinal pathology expertise reported in the CAP Practice Leader Survey.4
Our survey indicates that the job search experience of pathology residency/fellowship training graduates has remained remarkably stable during the last 5 years. The process is subjectively difficult for many, but most do find a position that they find satisfying. Anecdotally, concerns regarding a tight job market contribute to anxiety among current pathology residents and might be contributing to the recent decrement in the medical student applicant pool to pathology residency programs.3 The data reported here are reassuring regarding stability in the job market; an anticipated upturn in pathologist retirements1 may soon increase the number of open positions. An early start, geographic flexibility, and networking for word-of-mouth nonacademic practice opportunities are likely important factors in job search success.
We thank former members of the College of American Pathologists Graduate Medical Education Committee for their many contributions to the success of this survey: Christopher Alley, MD, Sterling Bennett, MD, Noel Brownlee, MD, Jimmy Clark, MD, Ronald Lepoff, MD, Christopher Otis, MD, Patricia Raciti, MD, and Romualdo Talento, MD.