Subspecialty choice: Why did you become a rheumatologist?




To determine the reasons trainees choose rheumatology as a subspecialty and to review the literature on career choices among physicians, particularly regarding the choice of subspecialty.


A questionnaire was designed to identify and analyze factors that influence rheumatology fellows to join the field of rheumatology. The questionnaire was administered online and answers were collated through the American College of Rheumatology Training and Workforce Committee, Subcommittee on Medical Student and Resident Recruitment. We reviewed the medical literature, using Medline and PubMed to find references to career choice among medical trainees.


The majority of rheumatology fellows had their initial exposure to rheumatology as second-year and third-year medical students, and >75% solidified their decision during internship and residency. Clinical rotations in rheumatology and exposure to role models and mentors were the most influential factors. Approximately 40% of rheumatology fellows cite their intellectual interest in the field as the most important contributor to their decision.


Career decision-making occurs throughout medical training. Exposure to clinical experiences and mentors are particularly influential and may have an impact during medical school, as well as during internal medicine residency training. These findings suggest that there are a variety of opportunities throughout medical training to influence career decision-making and improve recruitment into rheumatology. Additional financial resources that support recruitment efforts may be required, and followup studies assessing the effects of these efforts should be done.


According to the recently published workforce study carried out under the guidance of the Committee on Workforce and Training Issues of the American College of Rheumatology (ACR), there will be a considerable shortfall in the number of practicing rheumatologists in the US in the coming years (1). Understanding what motivates physicians to become rheumatologists is critical to recruitment efforts within the subspecialty. However, very little is known about the decision-making processes used by physicians in choosing subspecialty careers. We present what we believe to be the first survey of rheumatology trainees regarding their motivations to become rheumatologists. In addition, we review the literature on factors influencing medical career choices in general. This report should help provide a framework for future education, recruitment, and workforce policies within our discipline.


A questionnaire was adapted from models available in the literature (2, 3) in order to probe the ways in which trainees decide to enter the field of rheumatology (Figure 1). The questionnaire was e-mailed to all 469 current adult and pediatric rheumatology fellows in the US. A total of 177 fellows responded, including 150 from adult programs, 19 from pediatric programs, and 8 who identified themselves as being in combined programs (adult and pediatric or allergy/immunology and rheumatology). The response rate was 38%.

Figure 1.

Fellows survey: making the choice to become a rheumatologist.


Twenty-three percent of fellows had their initial exposure to rheumatology during their second year of medical school, and 27% in their third year (Figure 2A). However >50% said they first became interested in rheumatology during their internship and residency (Figure 2B), and >75% solidified their decision during this time (Figure 2C). Close review of the data, however, showed that many fellows first considered and decided on rheumatology as a career quite early in their medical training. Indeed, there were individuals making their decision at each point of training specified in the questionnaire.

Figure 2.

Responses (%) to the questions: A, “When was your first exposure to rheumatology?”; B, “When did you first become interested in rheumatology?”; and C, “When did you first become certain you wanted to be a rheumatologist?”

Analysis of the specific factors contributing to a decision to become a rheumatologist revealed a wide range of influences (Figure 3). However, the most frequently cited influence, identified by ∼35% of respondents, was a clinical rotation in rheumatology. An additional 28% of fellows cited a clinical mentor as their most important influence, and 14% cited an interaction with a patient.

Figure 3.

Responses (%) to the question: “What or who influenced you most in choosing a career in rheumatology?”

When asked to choose from a menu of options that might be included among their most important influences, ∼40% of fellows reported that intellectual interest in the field influenced their decision (Figure 4). Intellectual interest was the most frequent answer to the question “What aspects of rheumatology attracted you most to the profession?” Approximately 33% of fellows cited the intellectual content of rheumatology as being critical to their choice. In written comments, >20% of respondents specifically mentioned the complexity, variety, and depth of the diseases treated by rheumatologists, particularly regarding the diseases' underlying pathophysiologic mechanisms. Words like interesting, intellectually stimulating, and challenging were used in >13% of respondents' descriptions. Only 12% of fellows specifically cited the introduction of biologic agents for the treatment of rheumatic diseases as being among the aspects of the profession that attracted them most, and ∼10% noted the opportunity to form long-term relationships with patients.

Figure 4.

Responses (no.) to the question: “Which of the following entered into your decision-making?”

Lifestyle considerations was the second most-frequently chosen factor that entered into trainees' decision-making, although it was only mentioned by 14% of the trainees. Similarly, only 13% cited quality of life as being among the aspects that specifically attracted them to the profession.

Virtually every subspecialty of medicine, as well as general internal medicine, pediatrics, and family practice, was among the fields the current rheumatology fellows considered while making their career choices. The most frequent first choice after rheumatology was hematology/oncology, cited by 16% of those surveyed. Approximately 10% considered cardiology and 9% considered nephrology to be their next most-likely fields of choice.

When asked to define a rheumatologist, fellows frequently commented on the cognitive characteristics of the profession. Over 45% used terms like intellectual, intelligent, and curious to describe the characteristics of a rheumatologist. Fellows emphasized thoroughness, the detective work of fitting things together, and the ability to deal with uncertainty and complexity. Over 22% mentioned compassion, caring, and the ability to establish rapport with patients, and 11% specifically commented on the patience of rheumatologists. Approximately 10% emphasized that they consider rheumatologists to be superb internists and adept clinicians.

Over 70% of survey participants stated that enhancing exposure to the field would be the most important mechanism in attracting physicians to become rheumatologists. Respondents specifically stated that this exposure should occur during both medical school and residency, and ∼12% noted the importance of the clinical rotation in this regard. Another 10% of respondents noted the important influence of role models and mentors in attracting recruits to the field.


To our knowledge, no prior survey has specifically addressed what factors trainees consider when they choose to subspecialize in rheumatology. However, our review of the literature on career choices among medical students has revealed several factors that enter into medical career decision-making in general. These include exposure to an experience during medical school, exposure to a role model or mentor during medical school, exposure to programs or incentives designed to enhance recruitment, exposure to programs designed to enhance rheumatology education, the personality characteristics of the students themselves, and lifestyle considerations.

The educational literature suggests that exposure to a variety of experiences during medical school can have significant effects on ultimate career choice. In a questionnaire-based survey of 8,494 doctors graduating from all 23 UK medical schools from 1999 and 2000 (4), 45% of students cited “experience of a chosen subject as a student” and 27% cited “exposure to a particular teacher or department” as influencing their career choice “a great deal.” Only 14% of students' inclinations before medical school influenced their ultimate career choices. These observations are consistent with the findings of a survey of medical students that evaluated who among them chose surgery as a career (5). In a retrospective analysis of the surgery rotation case logs of 146 University of Wisconsin medical students, it was noted that students who chose general surgery saw more abdominal and general surgical cases than those who chose surgical subspecialities or nonsurgical fields (5).

Another example of an exposure commonly available to all medical students is the choice of an elective (6). In a survey of 2,883 students who took a sophomore radiology elective, 16% of respondents said that the elective influenced their career choice. Remarkably, of the respondents who went on to become radiologists, 77% were influenced in their career choice by the elective.

An analysis of the Association of American Medical Colleges' Graduate Student Questionnaires, and of career data from 555 medical students who graduated from a single medical school over a 5-year period (7), found that lack of exposure, such as not having participated in a research project, was a predictor of choosing a primary care career.

Based on number of studies, exposure to a role model or mentor during medical school can have a powerful influence on career choice. One study evaluated results of a questionnaire administered in 2002 to a random sample of 2,250 fourth-year medical students at all 121 US medical schools (8). The authors of this survey demonstrated that the only strong predictor of choosing general surgery was satisfaction with the quality of the attending teaching during the surgical rotation (odds ratio [OR] 2.14). Satisfaction was higher if the student had participated in hands-on activities in the operating room; had spent more time in conferences, in clinics, and in teaching rounds led by the attending; had been able to initially examine a new patient with an attending; and had a longer basic surgery clerkship. Satisfaction was lower if students spent more time retracting in the operating room, spent more time only observing in the operating room, and only took patients' medical history and performed physical examinations after the attending.

The importance of a mentor in choosing a generalist or specialist career was also cited in a review of influences on career choice in the Annals of Internal Medicine in 2005 (9). The students' perception of residents' and attendings' satisfaction with their career choices, peer support for their career choice, and exposure to course work and clerkships all influenced the students' career choices.

Similarly, choice of a specialist career has been noted to be closely linked to the experience a student has with a specialist during medical school (2). A study of medical students in their third-year clerkship concluded that exposure to highly-rated attendings and house officers was an independent predictor of the choice of an internal medicine residency. This may, in part, be linked to the fact that a role model may help a student appreciate the congruence between their approach to medicine and that of the specialty they chose (10).

Declining interest in generalist medical careers in the 1980s led to the creation of several programs designed specifically to enhance recruitment to general medicine. The Robert Wood Johnson Generalist Physician Initiative was one of the earliest of these programs (11). This 15-school initiative sponsored a sustained longitudinal experience in primary care in the preclinical year, and found that there was an increase in generalist output by ∼33%. Subsequently, the Health Resources and Services Administration sponsored the collaboration of 18 medical schools with more than 50 organizations to devise the Interdisciplinary Generalist Curriculum Project (12). Participation in the program was associated with an increased choice of a generalist career, and exposure to high-quality one-on-one preceptorships was one of the most important factors influencing this increase. Additionally, a statewide preceptorship program in Texas aimed at increasing the number of students choosing a family practice residency was reported in 2004 (13). Eight medical schools and >10,000 medical students participated in this program, which led to an increase in students entering family practice residencies, whether the students were exposed at the preclinical or clinical level (OR 1.62, 95% confidence interval [95% CI] 1.41–1.87; OR 2.31, 95% CI 1.99–2.68, respectively).

On a larger scale, the American College of Obstetricians and Gynecologists and the Association of Professors of Gynecology and Obstetrics collaborated to study the issue of the declining popularity of obstetrics and gynecology as a career choice (14). The investigators found that the following factors made their specialty less attractive: inadequate involvement of faculty in teaching; a lack of clear objectives for the clerkship; a lack of adequate feedback following the clerkship; alienation of prospective male applicants by attempts to attract women to the specialty; lifestyle and liability issues; and exposure to the clerkship only late in the curriculum, so that obstetrician/gynecologists were less likely to serve as role models and career mentors. To address these recruitment barriers, the sponsoring organizations developed strategies including improvement of faculty teaching skills, revision of medical student educational objectives, and prioritization of early contact with students. The results of these very comprehensive efforts have not yet been reported.

Financial incentives alone appear to be ineffective, at least in recruitment to certain locations (such as underserved areas), if not to certain fields (15). However, recruitment of students to rural areas is enhanced by exposure to rural practice during medical school and even high school, as well as by exposure to role models in rural practice (16).

A number of medical schools have reported development of musculoskeletal courses (17–20) aimed at developing physical examination skills among medical students. However, none of these curricula were intended to aid in recruitment into the field of rheumatology, and none of the published reports discuss outcome measures apart from the acquisition of physical examination skills. Therefore, the effect of these well-defined curricular innovations on recruitment is not currently known.

A joint injection clinic has also been developed (21) for second-year and third-year internal medicine residents, and has documented improved knowledge and confidence in both joint examination and joint injection skills. Any impact on resident interest in rheumatologic careers, however, was not assessed.

The personality characteristics of students also come into effect during their decision-making process. Making a choice to enter a subspecialty of internal medicine was investigated in a longitudinal, cohort study of 229 University of Western Australia medical students entering medical school either in 1984 or 1989 (3). This study showed that certain personality characteristics correlated with the choice to subspecialize. Those who pursued subspecialty training were more likely to have “more abstract thinking and mental capacity to discern relationships in terms of recognizing analogies and similarities and being able to form typologies” (i.e., they were less likely to be concrete thinkers). They were also more likely to be “people who stick to the rules, are conscientious and emotionally disciplined and somewhat staid” (i.e., less likely to be expedient). To our knowledge, no survey has examined the personality characteristics of rheumatologists.

Recent medical school graduates place considerable importance on lifestyle factors in making their career choices. This was boldly underscored in a widely cited 2003 article in the Journal of the American Medical Association (22), which found that the perception of the ability to maintain a controllable lifestyle was the most powerful predictor of specialty choice. The study used data from the National Resident Match Program, the San Francisco Match, and the American Urological Association from 1996–2000, as well as the American Medical Association's Physician Socioeconomic Statistics for 2000–2002. The authors defined a controllable lifestyle as one that permitted personal time free of practice requirements and control of weekly hours spent on professional activities. The authors found that the perception of a controllable lifestyle associated with a particular field accounted for 55% of the variability in recent patterns of specialty choice of graduating US medical students from 1996–2000. In contrast, the authors found that income accounted for only 9% of the variability, work hours 2%, and required years of training 4%. A survey of Australian medical students published in 2004 similarly suggested that the desire to balance work, family, and lifestyle was influential to the career choices made by both male and female students (23).

Our current report details the results of the first survey to ask rheumatology fellows about the reasons why they chose to become rheumatologists and reviews the literature regarding physician specialty choices in general. Trainees' career choices are made at many stages during medical training, offering educators multiple opportunities to influence those choices. A key decision-making point occurs during clinical rheumatology rotations, which is an experience many fellows cited as being important in solidifying their decision. These rotations may occur in medical school or during internship and residency training. Mentors and role models in clinical and research settings are also critical influences. Therefore, there is flexibility in the ways that rheumatology training programs can seek to enhance rheumatology exposure, allowing different programs to draw on their particular strengths in developing undergraduate medical curricula, clinical rotations, or other novel educational and research opportunities in rheumatology for trainees.

The educational literature provides added support to the results of our survey. Studies have demonstrated that specialty choices are often made on the basis of an elective or rotation during medical school, or after exposure to an inspiring mentor. This might occur in a longitudinal rheumatology clinic for interested students in their preclinical year(s), in a general medical clinic in which a rheumatologist acts as the supervising physician, or through participation in high-quality preceptorships. The ACR Research and Education Foundation, for instance, provides grants to trainees and educators in rheumatology who have the potential to support innovative educational programs, which might themselves serve as important recruitment tools.

Little is known about the personality characteristics of medical subspecialists, nor are there data to suggest that identifying a rheumatology personality type would be of value in recruitment. However, a controllable lifestyle is important to many trainees, and may be a relative strength for rheumatology as it is the second most-important factor in decision making, after intellectual interest, quoted in our survey results.

Although both our survey and the educational literature suggest that final career decisions are solidified during residency, most internal medicine residency programs are still predominantly hospital-based. Devoting energy to enhancing the clinical outpatient rheumatology rotation, and to attracting trainees to those rotations if they are not mandatory, is likely to yield high dividends in bringing trainees into the field.

In summary, our survey of rheumatology fellows has demonstrated that the consideration of a career in rheumatology begins early in the training years, and that early exposure to the field influences the decision-making process. Future recruitment strategies for medical schools and internal medicine residency programs should include the development of more one-on-one sessions with practicing rheumatologists for trainees and residents throughout their education. Rheumatologists should be encouraged to take a prominent role in teaching. Financial resources should be made available for those rheumatologists involved in teaching and training programs to attend seminars aimed at improving education and mentoring skills, and to provide protected time for them to mentor and to teach. Followup studies should be used to document the effect of such innovations on recruitment patterns in rheumatology, and to help refine current and future recruitment strategies.


Dr. Kolasinski had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Kolasinski, Bass, Kane-Wanger, Libman, Sandorfi, Utset.

Acquisition of data. Kolasinski.

Analysis and interpretation of data. Kolasinski, Bass, Kane-Wanger, Libman, Sandorfi, Utset.

Manuscript preparation. Kolasinski, Bass, Kane-Wanger, Libman, Sandorfi, Utset.


The authors wish to acknowledge the inspiration and leadership of Walter G. Barr, MD, Chair of the Committee on Training and Workforce Issues of the ACR, under whose guidance the work presented in this publication was completed. In addition, the authors thank LaTanya Batts and Amy Miller of the ACR for their tireless efforts in support of our work and on behalf of trainees in rheumatology.