The participation of physician-scientists in the biomedical research enterprise helps bridge the gap between basic science and clinical practice to address the health care needs of the public. In recent years, however, the number of physicians engaged in research careers has declined by over half, as have the number of research psychiatrists and research fellowships for psychiatrists (1—5). Physicians with in-depth experience and formal training in research design, development, implementation, statistical analysis, and interpretation of scientific information are becoming a "rare species" (6). Indeed, a recent study investigating the success rates of M.D.s, as compared to Ph.D.s, in obtaining federal funding showed that physicians have lower rates of success with first-time R01 applications and are less likely to obtain subsequent R01 grant funding (7). Combined M.D./Ph.D. programs routinely graduate M.D.s with formal research training. However, the vast majority of medical school graduates do not have Ph.D.s and the number of physicians pursuing research careers is on the decline.

Barriers to Research Training for Physicians

A number of factors are likely associated with the decline in research training during or in close proximity to psychiatric residency. Although some research training options for premedical and medical students exist, they are often too brief to allow the trainee to make informed career decisions (8). Aside from well-established combined M.D./Ph.D. programs, research training experiences that are offered during residency are typically brief and fragmented across years of training (5). Other barriers are limits on faculty time and funding for mentoring (4, 9). The presence of at least one mentor during training is associated with the development of successful investigators and is especially important in acquiring skills in grant writing, project implementation, and manuscript preparation (10—12). This has become a critical issue as academic medical centers are experiencing unprecedented financial pressures, and faculty members are directed to spend increasing amounts of time in revenue-generating activities. While NIH mentored "K-series" mechanisms have been important in increasing the number of physicians choosing research careers, they do not provide direct financial support for mentoring or infrastructure support, and they have a relatively low indirect cost recovery rate of 8%. Finally, since research training traditionally extends the period and cost of medical education, students with excessive debt favor higher paying clinical salaries, especially residents in psychiatry since it is among the lowest paid medical specialty (3, 5, 9, 13—15).

Research Training During Residency

In response to the shortage of physician-scientists, the Institute of Medicine convened a panel to study research training in the careers of psychiatrists. The panel noted that residency-based research training might have the benefit of solidifying research career interests for greater numbers of psychiatrists. One of the recommendations was that "departments of psychiatry should organize optional research experiences and mandatory research didactics in residency as early steps in research career development" (3).

Based on the Institute of Medicine's recommendation, the Medical University of South Carolina (MUSC) launched an NIH-funded research track (R25 mechanism) in 2006 to address the institutional, financial, and regulatory barriers to residency research training (http://www.musc.edu/psychiatry/research/cns/DART/dart.htm). The primary aim was to incorporate a research track within a 4-year psychiatric residency program for physicians committed to pursuing clinical research careers. The research track can be integrated within any of the three main psychiatric residency programs offered at MUSC (i.e., general psychiatry, combined general psychiatry and internal medicine, and combined general psychiatry and general neurology). A secondary goal was to extend recruitment into earlier phases of medical training by offering summer research fellowships to medical students and undergraduate trainees.

The research track at MUSC, known as the Drug Abuse Research Training (DART) program, focuses on addictions research; however, residents with associated research interests (e.g., depression, bipolar disorder) also participate. The DART program is unique in that it establishes the infrastructure and organizational support necessary to train physician-scientists, and it addresses deficiencies in the mentoring of residents by providing compensation to research faculty mentors. In 2009, the DART program received the American College of Psychiatrists Award for Creativity in Psychiatric Education for its use of innovative strategies to overcome existing barriers to physician research training, namely (a) access to research mentors; (b) support for research training time during residency; and (c) financial support, such as funding for pilot projects, travel to present research at conferences, and eligibility to apply for the NIH Loan Repayment Program. This article describes the DART program including mechanisms of training, recruitment, and outcomes to date. The training components described herein should be exportable to other psychiatric residency training programs and potentially other specialties of medicine.

General Program Description

Drug Abuse Research Training (DART) is a 2-year research track, encompassing the final 2 years of psychiatric residency training (PGY-3 and PGY-4). Table 1 illustrates how the program is incorporated into the parent residency program. During PGY-1, potential DART residents are encouraged to meet with possible research mentors and consider areas of research that interest them. During PGY-2, interested residents formally apply to the DART program. Applications are submitted in the early spring of PGY-2 and include a curriculum vitae; a one to two page statement of research interests; and two letters of reference, one of which is from the applicant's research mentor who will be supervising his or her work during the DART program. Approximately three to four new residents are accepted into the program each year. Selection is based on a number of criteria, including the candidate's research potential, his or her research mentor, and the quality of the statement of research. While research experience and publication history are taken into consideration when reviewing candidates, prior research experience is not essential. This is in contrast to other physician-scientist training programs (14) that accept only residents with substantial prior research experience (e.g., M.D./Ph.D. degrees). Although we would not preclude dual degree M.D./doctoral-level students from participating, none has applied to the program thus far. One DART alumnus has an M.D. and a Ph.D., but he entered medical school after completing the Ph.D. and was not enrolled in a combined M.D./Ph.D. program.

Selected residents spend approximately 50% of their time during PGY-3 and PGY-4 in a well-coordinated curriculum that includes core research training activities described below that are designed to facilitate the attainment of the research skills listed in Table 2. These activities have been consistently evaluated by trainees as beneficial and have been ranked by trainees with regard to importance in the following order:

Core Research Training Activities

Each trainee is required to attend at least 80% of didactic seminars, submit at least one manuscript and one conference abstract, complete two courses (one online and one at MUSC) in the responsible conduct of research, engage in a pilot project, meet with their research mentor on a regular basis, and attend a grant writing workshop. In addition to these core requirements, an individualized training plan is generated for each resident. At the beginning of the program (PGY-3), an initial assessment of current research knowledge and skills is conducted and is used to guide the development of the resident's training plan (Needs Assessment and Annual Goals forms are available from the authors upon request). This assessment helps to ensure that deficiencies in training that would serve as an impediment to a successful independent research career are targeted. These forms are then reviewed during the biannual evaluations.

Recruitment of Residents

Three main forms of "marketing" were created to inform potential applicants about the DART program, including a program website (http://www.musc.edu/psychiatry/research/cns/DART/dart.htm), a brochure, and a newsletter. The website, in particular, includes information about the program and its requirements, how to apply, seminar series topics and dates, trainees and research mentor profiles, and trainee research dissemination activities. DART recruitment essentially begins during the residency application process (fourth year of medical school), although undergraduates and first-year medical students also have exposure to DART through the Summer Research Fellowship (described below). DART representatives participate in the psychiatric residency selection process to ensure that adequate information about the research track is available to the applicants and selection committee. During the PGY-1 and PGY-2 years, the DART program is advertised at lectures and meetings, and the program newsletter and brochure are distributed to PGY-1 and PGY-2 residents.

Summer Research Fellowship

In order to target students earlier in their career development, a summer research initiative was created. The Summer Research Fellowship provides the opportunity for undergraduate and medical students to spend 12 weeks working closely with a research mentor at MUSC. During this time, students assist with research projects and research-related tasks such as data collection and entry and literature searches. At the end of the summer, students give a formal presentation to their peers and program mentors describing their research experiences. Students receive ∼$3,000 for their efforts. Since its inception in 2008, the Summer Research Fellowship has involved 22 students, including medical students from MUSC as well as undergraduate students from other universities. Examples of summer research projects include the study of neuroimaging in nicotine dependence, stress reactivity and prescription opioids, women's postpartum mental health, and cocaine-induced neuroplacticity. The complete list can be found on the DART program website.

Trainee Accomplishments

Since its inception in 2006, 15 residents have participated in the DART program. The following activities are assessed annually: (a) the number and quality of manuscripts submitted and/or published, (b) the number and quality of abstracts submitted and/or accepted for presentation at research conferences, and (c) attainment of research or career development awards. The residents who have participated in the program thus far have generated 36 abstracts submitted to local and national meetings, 28 manuscripts submitted for publication, and 14 institutional review board applications. In addition, they have received numerous research-related and teaching awards (e.g., Career Development Institute Awards, travel awards to national conferences, the APIRE/Janssen Pharmaceutica Resident Research Award). Of the 10 residents who have graduated from the program, four have chosen to complete a fellowship in addiction psychiatry, four have chosen child psychiatry residencies, and two accepted junior faculty positions (Assistant Professor). Alumni of the DART program will be followed for 10 years in order to assess research-related activities and productivity (e.g., manuscripts, abstracts, funding proposals submitted and obtained, percent time spent conducting research, positions secured).

Funding Considerations

Some NIH funds were used to cover the costs of program administration, pilot projects, and research assistance (i.e., statistical consultation) for trainees. In addition, each primary mentor received 5% salary support. We consider this to be important to the success of the program because it allowed us have clear expectations for the time, effort, and role of the mentors. Because good mentorship is such a critical element in research training, this level of accountability of the mentors helped to insure the success of the trainees. The NIH funding provided the impetus to begin the program. The salary and administrative support was particularly helpful during the early stages of program development, which can be very time consuming. However, if the program continues beyond the period of NIH funding, the department will need to provide support for ongoing activities. In the case where a department is unable to obtain NIH funds for program development, an investment would need to be made to initiate a research track within the residency. In order to keep costs down, mentors could be asked to provide the necessary resources for trainee pilot projects. Nevertheless, the financial status of many psychiatry departments in the current health care financing climate may be an impediment to initiating such a program.

Conclusions and Future Directions

The Drug Abuse Research Training (DART) program provides a model to effectively integrate research training during residency without increasing the number of years of residency training. The program recruits and mentors physicians to help advance them toward becoming independent investigators, provides a consistent curriculum of research training that begins early on during residency, and provides summer research training for undergraduate and medical school students so they may be exposed to research early in their career trajectories.

Although the DART program helps overcome many obstacles to resident research, several important challenges exist. First, there is an insufficient "bridge" between completing residency and obtaining a faculty appointment. Most of the residents in the DART program are not ready to submit a Career Development Award during the PGY-4 year. At MUSC, we have several fellowships that residents may enter into immediately following residency, which also gives them more time to prepare the grant application, but many universities are unable to offer fellowships. Furthermore, if junior researchers join the faculty without protected research time, the success of their long-term research career goals may be jeopardized. Second, financial challenges remain an obstacle. The Loan Repayment Program has been extremely beneficial in helping relieve resident debt load; however, it has become increasingly more competitive to obtain. In addition, the financial considerations mentioned earlier may limit the exportability of this program.

Finally, more women and underrepresented minorities need to be integrated into residency research training programs (16, 17). Thus far, 38.5% of DART residents have been women and 7.1% underrepresented minorities. These numbers are likely a direct reflection of the make-up of MUSC's residency training program (41% women, 9% underrepresented minorities). The women in the DART program have done as well as their male counterparts in all arenas (e.g., research productivity, grant submissions, obtaining post-DART fellowship and faculty positions). The DART program's success at attracting women physicians to clinical research may be attributed to the mentoring process, in which the quality of the mentoring relationship is emphasized and mentors are compensated for their efforts. Sakamoto and Dipple (18) comment that one of the most important factors contributing to their success as women physician-scientists is supportive mentorship. Outreach efforts to underrepresented minorities and women coupled with careful attention to the mentoring process will be important in increasing diversity in the physician-scientist community.

In conclusion, the DART program has been successful in providing a clinical research experience for psychiatric residents, medical students, and undergraduates. Most of the DART graduates have embarked on a career in academic medicine in which they are continuing their research activities. If financial barriers can be overcome, programs like this one may be of value in providing a much needed increase in the number of physicians involved in clinical research.

Funding was provided by National Institute on Drug Abuse grants R25 DA020537 (KTB) and K23 DA021228 (SEB). The authors wish to thank Ms. Crissy Rowell for her invaluable assistance with the DART program.

At the time of submission, the authors reported no competing interests.