Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 91
Research Training in Psychiatry Residency: Strategies for Reform 4 Institutional Factors The last chapter focused on regulatory issues that influence psychiatric research training during residency, noting that national oversight of the residency accreditation and board certification processes has considerable influence on the goals of residency training, research literacy, and research training. This chapter looks more directly at the programs themselves, considering the obstacles and strategies of institutions and departments regarding research training during residency. As financial constraints are central to this subject, the chapter begins with a brief discussion of how residency education is typically funded in the United States. It then addresses two key institutional factors that influence research training during residency: leadership and mentoring, and program and curriculum structure. The chapter ends with conclusions and recommendations that include a theoretical framework for evaluating institutions that aim to offer research education to psychiatry trainees. FUNDING ISSUES IN GRADUATE MEDICAL EDUCATION The General Funding Stream Graduate medical education (GME) funding for all residents comes primarily from the following sources: Medicare, Medicaid, the Veterans Administration, the Department of Defense, the National Institutes of Health (NIH), and the private sector (see Table 4-1) (Anderson et al.,
OCR for page 92
Research Training in Psychiatry Residency: Strategies for Reform 2001). As the numbers in Table 4-1 indicate, Medicare is the largest single source of GME funding. Medicare is also currently the most reliable source of GME funding because federal law requires annual entitlement payments to institutions that serve Medicare patients to subsidize both “direct” and “indirect” costs associated with training new physicians. Direct medical education (DME) payments subsidize resident stipends and benefits, faculty teaching time, and educational infrastructure. Indirect medical education (IME) payments, which are nearly twice as large as DME, are designed to subsidize the less visible costs associated with GME, including the fact that trainees tend to deliver less efficient care than do more experienced physicians (e.g., overprescribing tests), and that teaching hospitals typically treat the most severely ill patients. In an effort to minimize short-term operating costs, nongovernmental third-party payers are inclined to avoid GME costs that do not relate directly to patient care (e.g., certain IME costs or stipends for residents doing research training) (Knapp, 2002). This inclination has placed general financial pressure on the educational mission of institutions that train residents. It has also led to the introduction of proposed federal legislation aimed at ensuring that all users of medical care contribute equally to GME funding—legislation that was originally introduced by the late Senator Moynihan (D-NY) in 1999 and that has the strong support of the Association of American Medical Colleges (AAMC, 2001). TABLE 4-1 Sources of Graduate Medical Education Funding Source Amount (billions of dollars) Medicare 7.8a (2.7 direct, 5.1 indirect) Medicaid 2.3b VA/DOD/NIHc 2.0d Private-Sector Payers 6.0 NOTES: aYear: 2000. bYear: 1998. cVeterans Administration/Department of Defense/National Institutes of Health. dIn 2001, NIH training grants and fellowships accounted for $300 million of this amount. As these training and fellowship grants include Ph.D.’s and medical residents, the NIH contribution to GME is well below that $300 million dollar amount. SOURCE: Anderson et al. (2001).
OCR for page 93
Research Training in Psychiatry Residency: Strategies for Reform Funding Issues in Pediatric Graduate Medical Education Institutions that treat predominantly pediatric populations may receive lower amounts of GME funding than other institutions because they are less likely to treat Medicare’s primary beneficiary—the elderly. As a result, they are dependent upon non-Medicare sources, which are not entitlements and are subject to annual local or federal appropriations processes (Henderson, 2000).20 Although child and adolescent psychiatry residents at Medicare-funded institutions are considered 100 percent full-time equivalents (FTEs) for reimbursement calculation purposes, the reimbursement is based on an institution’s Medicare utilization, an index that reduces the reimbursement rate for institutions with a high pediatric caseload. Additionally, in child and adolescent psychiatry, as with all Accreditation Council for Graduate Medical Education (ACGME)-designated subspecialties, GME coverage drops to 50 percent for each FTE in postgraduate year 5 (PGY5) (ACGME, 2000a) because the Medicare law offers full reimbursement only for what the ACGME defines as “general” training (American Academy of Child and Adolescent Psychiatry [AACAP], 2002b).21 It is notable that child and adolescent psychiatry is considered a subspecialty of adult (general) psychiatry, even though pediatrics is not considered a subspecialty of internal medicine. Supporting Research Activity Through Graduate Medical Education Funds Research training is peripheral to immediate clinical care. Consequently, there are some limits on the use of Medicare GME funding to cover residents engaged in research training activities. A review of federal regulations pertaining to GME reimbursement from Medicare indicates that neither DME nor IME reimbursements are intended to cover activities outside of patient care.21, 22, 23 One regulation explicitly states: “The time spent by a resident in research that is not associated with the treatment or diagnosis of a particular patient is not countable.”22 This regulation clearly excludes “basic research” on nonhumans, although it 20 Healthcare Research and Quality Act of 1999. Pub. L. No. 106-129 (1999); Children’s Health Act of 2000. Pub. L. No. 106-310 (2000). 21 Direct Graduate Medical Education Payments. 42 C.F.R §413.86 (2001). 22 Special treatment: Hospitals that incur indirect costs for graduate medical education programs. 42 C.F.R. §412.105 (1999). 23 HHS (Health and Human Services). Counting Research Time as Direct and Indirect GME Costs. F.R.66(148): 39896. 2001.
OCR for page 94
Research Training in Psychiatry Residency: Strategies for Reform does not necessarily limit patient-oriented research, especially if such activity involves regular patient contact. Since the federal regulations additionally defer to the ACGME for determining the legitimacy of reimbursable resident activities, there is room for educational activities not exclusively linked to billable clinical productivity. Nevertheless, the concern that funding may be reduced if residents engage in research training adds uncertainty to an already tenuous stream of federal support for residency training in general. This concern prompted officials at the University of Michigan to obtain a grant from the National Institute of Mental Health (NIMH) to cover their psychiatry residency-based research training program (McCullum-Smith, 2002). Concern has also been heightened in the New England area as that region’s Medicare intermediary has asked institutions to refund GME money that supported residents engaged in “bench” research. Although the action targeted surgery residency programs that permitted a full “year-out” for residents to conduct basic laboratory work, it has had a discouraging effect on residency-based research training initiatives more broadly, especially those that involve basic research training components (personal communication, S. Benjamin, University of Massachusetts Medical School, July 22, 2002). The committee validated the above described Medicare restrictions on research activity by interviewing GME directors at institutions in Arizona, Arkansas, Georgia, and Washington State. These GME directors verified that research, and especially basic research activity by residents, typically is not reimbursable by Medicare. They also indicated that increased scrutiny by Medicare intermediaries is part of a more general effort among third-party payers to control their costs. At the same time, these GME directors were all aware that clinical research activities that encompass the diagnosis and treatment of patients are reimbursable under Medicare, although they acknowledged that the regulations are sometimes confusing to those engaged in the accounting process. A further and important consideration is the institutional flow of GME dollars. These funds usually are directed to hospitals rather than to departmental residency programs, and federal legislation dating back to 1986 prevents expanding the numbers of medical residents funded by Medicare (Knapp, 2002).24 Given the variability in institutional and departmental needs, GME funding for psychiatry training programs may or may not be proportional to the size of those programs. Training slots may be reallocated to other departments, or IME dollars, which are tendered by an institution to support the general training environment, may not 24 Consolidated Omnibus Budget Reconciliation Act of 1985, Pub. L. No. 97-272 (1986).
OCR for page 95
Research Training in Psychiatry Residency: Strategies for Reform proportionally finance all departments responsible for training residents. Thus, the flow of funds to support residency and fellowship training may be more generous for one program than another, adding to the uncertainty faced by individual training programs regarding their operating budgets. It is unclear whether GME funding adequately covers the true cost of training the next generation of medical doctors. What is known, however, is that resident compensation is far below what entry-level physicians earn after graduating from residency—a reality that has prompted some analysts to argue that residents themselves bear considerable cost in the training endeavor (Newhouse and Wilensky, 2001), and that furthermore is tied to a recent lawsuit by 200,000 medical residents claiming that the GME matching system supports the economic exploitation of physician trainees (AAMC, 2003; Miller and Greaney, 2003). Additionally, cost-saving measures in recent years have eroded general GME funding streams to teaching hospitals, as well as direct and indirect streams of capital to research training. Moreover, residency funding for pediatric programs, including child and adolescent psychiatry programs, is currently even less secure than funding for programs involving a substantial Medicare patient load. The above are key financial realities faced by all U.S. residency training programs, including those that train psychiatrists. General Research Funding Layered over the GME funding constraints described above are the general financial challenges imposed by the emergence of managed care. Across all of medicine, clinical reimbursement rates have decreased, yielding lower per-hour incomes for individual physicians and for the departments in which they serve. As a result of lower clinical incomes, residents and faculty have less discretionary time for research and research mentoring because they need to increase clinical volume to compensate for the lower reimbursement rates (AAMC, 2002b; Beresin, 1997; Ludmerer, 1999; Mirin, 2002; Pardes, 2002). Additionally, lower clinical income reduces the surplus revenues traditionally used by institutions to cross-subsidize research and other activities not encompassed by the clinical mission (AAMC, 1999; Jones and Sanderson, 1996). A recent study by the Commonwealth Fund Taskforce on Academic Health Centers (1999) found that nearly 10 percent of research at academic health centers is supported by surplus from faculty practice plans. Perhaps even more important, Moy et al. (1997) found that managed care
OCR for page 96
Research Training in Psychiatry Residency: Strategies for Reform penetration not only decreases surpluses from clinical work, but also appears to discourage institutions from seeking federal research dollars. This finding is particularly important as federal grants account for nearly 70 percent of the research funding for academic health centers (Commonwealth Fund Taskforce on Academic Health Centers, 1999). For departments of psychiatry, shortages in clinical income may be even more acute. Despite the recent introduction of mental health parity laws that have required some insurers to cover mental disorders at levels similar to those for other diseases, full mental health care coverage remains patchy both geographically (i.e., state by state) and with regard to the extent of coverage (e.g., small employers have been exempt from federal provisions, and parity for addiction is often excluded) (Frank et al., 2001b). Recent estimates cited in the Surgeon General’s Report on Mental Health indicate that as of 1997, medium to large corporations were offering mental health benefits to their employees valued at 3 percent of the total medical benefits, down from 6 percent just 10 years earlier (DHHS, 1999; HayGroup, 1999). Given the increased awareness of mental disorders and treatment that occurred during this decade (see Chapter 1) and the Surgeon General’s estimate that mental disorders account for more than 15 percent of the disease burden in industrialized countries, a 3 percent insurance benefit appears disproportionately low. Another analysis reported by the Surgeon General determined that if a family experienced $35,000 in mental health expenses during a given year, that family would be responsible for $12,000 out of pocket, compared with only $1,500 in out-of-pocket expenses for equally costly medical/surgical care in the same year (Zuvekas et al., 1998). This lack of insurance/reimbursement equity adds to the financial pressure faced by psychiatrists and other mental health practitioners with regard to declining clinical revenues secondary to managed care, and diminishes the opportunity to partially fund clinical research from patient care revenues. These financial realities exist in an age when patient-oriented research costs are increasing as a result of the growing complexity of such investigative endeavors (AAMC, 1999). Psychiatric research, like other biomedical research, relies on a multidisciplinary team approach (Beresin, 1997; Institute of Medicine [IOM], 2000; Meador-Woodruff, 2002; Meyer and McLaughlin, 1998; Roberts and Bogenschutz, 2001). For example, a brain imaging study of psychiatric patients requires not only considerable material investment in scanning equipment and facilities, but also ongoing technical support from various experts, including psychiatrists, physicists, neuroscientists, computer programmers, psychologists, and biostatisticians. The administrative costs of research have also risen in recent years because of increasing institutional and govern-
OCR for page 97
Research Training in Psychiatry Residency: Strategies for Reform mental oversight aimed at protecting the rights and safety of research participants and at ensuring that research dollars are well utilized (Holmes et al., 2000; Miller, 2001; Shalala, 2000). The above discussion characterizes the challenges of obtaining funding for GME and research activity more generally. Most of these constraints are not unique to psychiatry, but affect other branches of medical practice and research as well. Constraints on these resource streams logically translate into a short supply of money to finance the research training activities of residents and the underlying infrastructure needed to support that training. A Business Case for Research Despite the above financial limitations, many programs incorporate research and research training into their broad departmental activities. They do so largely because new knowledge, especially as it relates to enhancing patient care, fits naturally into the philosophy of most clinical departments and institutions. Nevertheless, the ideal of research and research training can be at odds with the immediate needs of patients and the financial bottom line of departments. Accordingly, the committee believes a specific business case for research and research training should be aggressively pursued by psychiatry departments and should be formalized as part of a department’s financial plan. This business case should include metrics to measure both the direct and indirect benefits of research activity and research training within a department. Regarding the direct financial benefits of research, the committee was able to identify only one study, by Chin et al. (1985), that compares research-generated departmental income with income from clinical activity. This study found that research activities yielded far more departmental revenue per faculty FTE than the faculty clinical practice plan ($944,000/year versus $250,000/year).25 Chin et al’s. work is based on 1981 data from a large and relatively wealthy department (Stanford University’s Department of Internal Medicine), so it has limited contemporary applicability and does not necessarily support a business case for research in less resource-intensive settings. Furthermore, Chin et al. do not factor in the resources necessary to support faculty during periods when grants are not funded or clinical volumes are not achieved. Never- 25 Based on a sample of 52 FTEs who, on average, spent 27 percent of their time on federal or other extramurally funded research.
OCR for page 98
Research Training in Psychiatry Residency: Strategies for Reform theless, the study does indicate that departments can develop measures of the financial return related to research activity. Metrics or accounting systems have been developed to quantify the departmental income and relative value of research compared with other activities, including clinical and teaching responsibilities (Kastor et al., 1997; Scheid et al., 2002). Although the committee is not aware of any studies that have used metrics to demonstrate the financial benefits of a sustained research effort by medical departments, it is reasonable to hypothesize that such metrics could help individual departments determine the feasibility of using intramural resources to pursue the goal of building an extramurally funded research portfolio. Finally, analogous metrics could be used to assess less direct benefits of department-supported research, such as the prestige or faculty or patron satisfaction associated with research activity. Specifically, successful research programs are likely to attract the most ambitious faculty and trainees, as well as patients, third-party payers, and benefactors who are interested in having access to and supporting cutting-edge technologies (Pardes, 2002). Strategies for Funding Smaller Programs Building a research program or research training effort in less resource-intensive settings is more challenging than sustaining a large, existing program. The current reality is that some institutions receive considerable federal funding, whereas the rest receive little or none (Brainard, 2002). Psychiatry is no exception in this regard, as most psychiatric research funding is concentrated in the top 10 to 15 percent of psychiatry departments nationwide (Pincus, 2002). Specifically, in 2002 the top 10 NIH-funded departments obtained a combined total of nearly $365 million, while the next 75 departments received a total of $386 million (NIH, 2003b). For child and adolescent psychiatry, the concentration of research wealth is even greater, with fewer than 10 child and adolescent divisions having a substantial research effort (Beresin, 1997). In internal medicine departments, by comparison, the concentration of research resources is less severe, with the top 10 departments obtaining a total of $893 million, compared with $1.615 billion for the next 75 departments (NIH, 2003b). Thus the top 10 departments in psychiatry obtained nearly 49 percent of NIH funding for that discipline, whereas the top 10 in internal medicine received only 36 percent of the corresponding aggregate funding. Consequently, it may well be that many or most psychiatry departments lack the technological infrastructure and critical
OCR for page 99
Research Training in Psychiatry Residency: Strategies for Reform mass of researchers necessary to effectively support comprehensive research activity and training. The obvious way for these smaller departments to build a research and research training effort is to seek extramural support. However, the disparity between resource-rich and resource-poor programs makes it difficult for the latter to compete for extramural support because funding agencies, especially those that fund training or early career award grants, are interested in the resources and environment of the applicants’ institution, including the qualifications of mentors or senior investigators. The general challenge of obtaining extramural funding has become even greater because the NIH budget-doubling initiative is complete as of 2003, and because significant declines have occurred in the U.S. economy since early 2001. Nevertheless, numerous private and public extramural funding options exist for biomedical researchers. Appendix B lists several government, foundation, and industry grants that support research training or research infrastructure during or in close temporal proximity to residency. Some of the sources of extramural research support are also summarized below. Large grants from NIH. Building infrastructure is important to small programs that wish to compete with larger institutions, attract quality researchers, and sustain research efforts. Two infrastructure grants—the Centers of Biomedical Research Excellence (COBRE) and the Biomedical Research Infrastructure Network (BRIN)—target 23 states and Puerto Rico, as these localities have historically been low utilizers of NIH funding mechanisms. Accordingly, these mechanisms may be models for the establishment of research infrastructure at institutions with less resource-intensive departments of psychiatry. The BRIN and COBRE grants are designed to build local biomedical research infrastructure, including personnel recruitment and training efforts, in regions having the greatest need for resource expansion. Considering that more than 50 percent of all NIMH funding goes to four states (New York, California, Pennsylvania, and Massachusetts)26 (NIMH, 2001a), it appears reasonable for departments of psychiatry in other states to consider these infrastructure-building grant mechanisms or other funding opportunities that target underrepresented regions or small departments. NIMH might encourage applications for these or similar funding mechanisms by marketing them more aggressively to small or emerg- 26 Census data for 2000 indicate that the population in these four states is 27 percent of the entire U.S. population.
OCR for page 100
Research Training in Psychiatry Residency: Strategies for Reform ing programs, and by encouraging resource-rich programs to partner with smaller programs such that the latter can enhance their research efforts while the former work to expand their geographic perspective and access to patient populations. The first BRIN (3 years in duration) and COBRE (5 years in duration) grants were awarded in 2000 and 2001, respectively. The BRIN grant is aimed at fostering collaboration among different institutions within one state, whereas the COBRE grant is awarded to one institution that may or may not collaborate with others. The BRIN grant encourages the creation of a research infrastructure that will attract research scientists. The COBRE grant operates with one senior scientist who fosters the development of junior investigators; to receive a COBRE grant, an institution must establish three to five multidisciplinary research projects. The maximum amount given to a state for a BRIN grant is $2 million/year (each state may submit no more than two applications for potential funding) (NIH, 2000b). The maximum amount given to an institution for a COBRE grant is $1.5 million/year, with a limit of three simultaneous submissions (NIH, 2001b; 2002c). As of spring 2003, most research projects funded under these infrastructure mechanisms support basic research, although clinical research is permitted. None of the funded grants focus on training psychiatry residents, although approximately 40 percent have a neuroscience component. The projects funded thus far are reviewed individually on an annual basis. A systematic and broad review of their overall success in terms of research productivity is not anticipated until 2004 (personal communication, F. Taylor, National Center for Research Resources, April 10, 2003). The NIH General Clinical Research Center (GCRC) grant is another model that can be used by NIH and other institutions to build research programs at less research-intensive institutions. Departments should consider these centers for the development of fledging research projects and training opportunities. Approximately 80 GCRCs currently support inpatient and outpatient facilities, core laboratories, bioinformatics programs, biostatisticians, and administrative technical personnel, all of which can be utilized by subscribing investigators at relatively modest cost (AAMC, 1999). Several GCRCs across the country have behavioral assessment cores that can assist with psychiatric research efforts (NIH, 2003a). Although these centers are at large, well-established institutions and are intended to support established investigators with peer-reviewed research funding, NIH encourages GCRCs to expand their efforts by supporting new training and research grants. Specifically, this means GCRCs are encouraged to support fledgling investigators conducting pilot studies and ultimately aiming to submit training or other grant ap-
OCR for page 101
Research Training in Psychiatry Residency: Strategies for Reform plications themselves. GCRCs can support these new investigators or trainees by offering logistic (e.g., human subject informed consent), scientific (e.g., statistical), and infrastructure (e.g., inpatient and outpatient facilities) resources. Industry or philanthropic support. There are numerous other sources of research and research training funds in addition to the federal funding mechanisms described above and in Appendix B. In a presentation to the committee in June 2002, Herbert Pardes, chief executive officer (CEO), New York Presbyterian Hospital (former director of NIMH), suggested that surplus income from industry-sponsored trials could be earmarked for departmental research. Dependence on industry funding, however, has drawbacks, as the work can be tedious and also can involve conflicts of interest (IOM, 1994; 2002b; Pincus, 1995). Therefore, such arrangements with industry need to be carefully conceived. Dr. Pardes and others have also made the point that medical institutions need to work aggressively to raise money for research from private sources, such as foundations and individuals in their community (Jacobs et al., 1997; Pardes, 2002). This notion is supported by public opinion surveys indicating that 61 percent of the population is willing to pay higher taxes to support research funding (Research!America, 2002), as well as by focus group and survey work done by the AAMC revealing that biomedical research and patient care rank well ahead of clinical GME in the minds of most voting Americans (Knapp, 2002). These findings indicate that general departmental fund-raising efforts and those targeting residency or fellowship research training programs are of interest to many potential private donors. In fact, philanthropic support for GME might well benefit from reminding potential donors that today’s residents are tomorrow’s researchers. One example of successful philanthropic fund raising occurred at the University of Texas at Southwestern, where the psychiatry department raised funds to support nine endowed chairs, four additional faculty positions, and $18 million in research activities from 1977 to 1996 (Meyer and McLaughlin, 1998). Collaboration with other departments or institutions. An alternative way for small programs to tap available clinical research resources is to seek out opportunities for interdepartmental or interinstitutional collaboration. A recent AAMC task force made the following recommendation: To enhance clinical research programs and infrastructure development, medical schools and teaching hospitals
OCR for page 122
Research Training in Psychiatry Residency: Strategies for Reform have no well-formulated method for evaluating their training program in terms of knowledge gains or research productivity. Several respondents indicated that there are two main components to the cost of research training: the direct inputs into the training, such as teaching time and equipment/facilities, and the costs associated with losing coverage in the clinics that residents typically staff. Respondents who estimated the costs associated with providing 4 hours per week of research training calculated that doing so would require additional funding of at least $5,000–$10,000 per trainee per year, plus $100 per hour to offset the costs associated with individual mentoring. Divisions with substantial extramural research funding may allow residents to assist with an ongoing study offering direct, but supported, research experience. Even resource-rich divisions, however, may not have the funds to support the ancillary activities (e.g., statistical analysis) required to successfully append a question to an existing research study. Respondents from small programs especially noted the shortage of mentor time as a limiting factor in research training. Most respondents said that the majority of their trainees did not appear to be interested in research education, basing this conclusion on residents’ poor attendance at the research courses offered. Consistent with that evidence, respondents from all levels of programs agreed that research activity in residency should be elective, not mandatory. Furthermore, many respondents suggested that recruiting interested and talented trainees to such research electives was a key challenge that should be addressed with at least two principal strategies. One strategy is to entice junior residents to research training as early as possible in their career, perhaps by formulating an exciting, nationally applicable curriculum in the integrated neural and behavioral sciences. Another strategy is to educate smaller programs about the numerous research training opportunities that exist (e.g., federal and foundation grants, new technology) through seminars, Internet sites, and other outreach methods. As briefly mentioned in Chapter 3, the principle professional society for child psychiatrists, the American Academy of Child and Adolescent Psychiatry (AACAP) has already made progress in developing model curricula for a “traditional” (i.e., 5-year) residency in adult and child psychiatry and for a 6-year program aimed at “the development of outstanding candidates who are interested in pursuing a career in academic child and adolescent psychiatry” (AACAP, 2003b:2). Both curricula offer a weekly 1.5-hour research seminar beginning in PGY2, research electives of 2 months’ duration in PGY3, and research activity beginning in PGY4. For the traditional track, 1 day is set aside for research in PGY4 and 3 days in PGY5. In the 6-year track, 80 percent of time in
OCR for page 123
Research Training in Psychiatry Residency: Strategies for Reform PGY5 and PGY6 is dedicated to “mentored research.” The program is being developed in collaboration with a number of stakeholders in psychiatry (including the Psychiatry RRC), and it will soon be implemented at Yale University (personal communication, J. Leckman, Yale University, April 4, 2003). Finally, the AACAP curricula are intended to serve as models for programs nationally. Barriers to broad implementation of such curricula include resident stipends beyond PGY4 (see the earlier discussion of GME funding) and the local availability of research mentors and other patient-oriented research resources. Program Success in Training Researchers The committee had neither the resources nor the mandate to gather or generate outcome data on a large sample of residency training programs; however, several programs voluntarily provided some limited data indicating the numbers of researchers that have emerged from their training programs. The data are of limited utility because they were not collected in a systematic fashion. Specifically, they do not represent a random sample of programs, and the resulting success rates are not necessarily comparable across programs. Additionally, it should be noted that most programs do not aim to train psychiatrist-researchers, but instead focus on clinical training, so it is unreasonable to expect that a sizable proportion of their trainees will end up on research career paths. Nevertheless, these data are presented in Table 4-2 to offer a summary view of research training rates across core residency programs (i.e., not a specialized research track). Despite the imprecision of the data collected, the numbers in Table 4-2 demonstrate that most residency programs yield career researchers well under 10 percent of the time. The difficulty encountered in obtaining these data—many programs provide only estimates—underscores the fact that residency-based research training is not typically monitored. As expected, the proportion of residents who end up in research careers is well below the proportion of research fellows who do so (see the above descriptions of the Columbia University, WPIC, and University of Michigan programs), again indicating the relevance of postresidency training for psychiatrists truly interested in research.
OCR for page 124
Research Training in Psychiatry Residency: Strategies for Reform TABLE 4-2 Research Training Outcome Data from Several Residency Programs in Psychiatry Psychiatric Residency Program (adult and child together unless noted) Percentage of Residents Moving into Researcha Time Period Brown University 5.0 1997–2002 Columbia University (adult) 30.0 1985–1999 Duke University (child) 7.5 1992–2002 Emory University 10.0 1997–2002 Indiana University 36.0 1997–2002 Johns Hopkins University (child) 36.0 1997–2002 Medical College of Ohio (child) 0.0 1977–2002 North Dakota University (adult) 0.0 1980–2002 State University of New York– University of Buffalo (child) 0.0 1999–2002 University of Arkansas 12.5 1997–2002 University of Connecticut 10.0 1996–2002 University of Michigan (adult) 19.0 1982–2002 University of Minnesota 10.0 1997–2002 University of Nebraska 2.0 1997–2002 University of Texas at Southwestern 5.0 1997–2002 Neuropsychiatric Institute, University of California, Los Angeles (child) 13.0 1997–2002 Virginia Commonwealth University 5.0 1997–2002 Washington University (adult) 14.0 1998–2002 Washington University (child) 23.0 1992–2002 Western Psychiatric Institute and Clinic 21.0 1996–2001 NOTE: aA rough index of the proportion of residents in a given program who move on to research careers. In some cases, this may mean they have been in research careers for several years; in others, it may mean they have recently transitioned to a fellowship or junior research position. Because these values were not obtained systematically, they are intended only as approximations, not as values for comparison across programs. SOURCE: Data were derived from various sources, including correspondence with training directors, website and literature reviews, and eight focused interviews with department chairs. Appendix C offers additional details regarding the programs listed.
OCR for page 125
Research Training in Psychiatry Residency: Strategies for Reform A crude comparison of the responses in this table with other research involvement rates among psychiatrists confirm the accuracy of the numbers represented. For example, metafile data from the American Medical Association (AMA) indicate that 2 percent of all practicing psychiatrists in the United States consider research their dominant professional activity (Pasko and Seidman, 2002). Likewise, APA survey data reviewed by economist Douglas Schwalm (2002a; see Chapter 5) indicate that just under 20 percent of psychiatrists engage in any (i.e., greater than 1 percent effort) research activity. Accordingly, meaningful levels of research activity by U.S. psychiatrists likely fall somewhere between 2 and 20 percent, suggesting that the numbers in Table 4-2, which average out to 13 percent, are reasonable, but likely include those who dedicate well under 50 percent of their professional effort to the research endeavor. The data in Table 4-2 may further be used to support the hypothesis that the majority of new researchers in psychiatry are trained at a small number of programs as only 5 of the 20 programs represented claimed that 20 percent or more of their residents moved into research careers. CONCLUSIONS AND RECOMMENDATIONS This chapter has described institutional, departmental, and curricular factors that influence research training in residency. Funding, mentoring, and resident scheduling issues appear to be the chief constraints on research training in residency. Funding for residency training is heavily influenced by Medicare GME policies, and that funding stream is under increasing negative pressure. Research is not generally considered part of core residency training. As a result, funding for research activity needs to be justified independently and obtained either from extramural grants or from discretionary internal funds (e.g., endowments, profits from practice plans). Leaders of medical institutions have control over how Medicare and other funds are distributed. They additionally set expectations regarding trainee and faculty activities through organizational systems, such as those that determine promotional policies and general resource allocation. Accordingly, leaders (e.g., department chairs, deans, presidents) play a key role in assigning value to and maintaining the research mission, which includes research didactics and activity within training programs. Therefore, the committee believes the following recommendation is critical to research training in psychiatry: Recommendation 4.1. The broad psychiatry community should work more aggressively to encourage uni-
OCR for page 126
Research Training in Psychiatry Residency: Strategies for Reform versity presidents, deans and hospital chief executive officers to give greater priority to the advancement of mental health through investments in leadership, faculty, and infrastructure for research and research training in psychiatry departments. Although this recommendation applies equally to most branches of medical research, psychiatric research is arguably of particular importance in this regard. This is the case because current opportunities in brain and behavioral research are so great (see Chapter 1), and because mental illness is the object of considerable stigma that appears to have the dual effects of inhibiting efficient health care delivery (e.g., getting patients to the doctor), and impeding full reimbursement for rendered mental health services. The Surgeon General’s Report on Mental Health demonstrates as well as any document the relative importance of mental health; the ways in which the brain and behavioral sciences have advanced in recent years and the relevance of future advances to overall health; and the extent to which deeply engrained stigma works against equitable funding for mental health care—inequities that adversely affect research advances, which are partially subsidized by clinical revenues (DHHS, 1999). Accordingly, medical administrators should be aggressively encouraged to invest in expanding research training in psychiatry as a first step to at least bring psychiatrists on par with the research efforts of many other medical specialists (e.g., subspecialties of internal medicine, neurology). Department chairs and other leaders can promote psychiatric research by developing and financing a long-term business plan that considers the monetary, marketing, and societal benefits likely to result from mental health research. Institutional executives need to be encouraged to invest in these plans by utilizing reasonable portions of their general funds (e.g., IME, dean’s tax, endowments) and by frequently including psychiatric research agendas in fund-raising efforts. At the same time, these leaders (especially those in psychiatry) should educate medical students and residents regarding the extraordinary intellectual ventures that accompany research in psychiatry. To the extent that such education and promotion efforts are already occurring, it is the committee’s sense that they need to be expanded if any real gains are to be made in the number of psychiatry trainees tracking to research careers. One of the most intensive forms of leadership is mentoring. Mentoring is probably the ingredient cited most frequently as necessary for effective research training. The shortage of mentors is also a commonly noted barrier to effective research training. Accordingly, the committee believes that
OCR for page 127
Research Training in Psychiatry Residency: Strategies for Reform financial incentives may be important to encourage more senior researchers, particularly at small institutions, to enter the mentoring pool. Accordingly, the committee makes the following recommendation: Recommendation 4.2. Academic institutions and their psychiatry residency training programs should reward the involvement of patient-oriented research faculty in the residency training process. The National Institute of Mental Health should take the lead in identifying funding mechanisms to support such incentives. This recommendation targets in particular smaller institutions with limited resources to offer a broad range of research experiences to and mentors for their trainees. Trainees in well-established programs are more likely to “pay for themselves” by extending the productivity of the mentor, and supplements to existing grants can be used to cover some of the costs associated with the trainees’ work. At less resource-intensive institutions, however, prospective trainees will likely be less familiar with research methods so that mentoring will require a greater investment of time with a potentially lower return in terms of trainee productivity. In these contexts, the committee encourages mechanisms to finance mentoring, with the provision that grant renewal would depend on the research success of the mentor’s past trainees. As an alternative to on-site mentoring, a remote system of mentoring might be devised to give both faculty and trainees the opportunity to be matched with individuals having similar interests outside of their institution. Furthermore, such a network might be sustained by offering senior mentors consulting fees or other remunerative support (e.g., travel, equipment) for their expertise and time. In addition to issues related to institutional leadership and mentoring, this chapter has reviewed clinical research training programs generally and several psychiatry residency programs with regard to research training. The programs reviewed are highly variable. For example, nonpsychiatry training includes clinical research programs that range from 1-year certificates to multiyear programs culminating with a Ph.D. Although this range appears to be geared in part to the broad range of applicants, an AAMC task force concluded that program variability reflects imprecision regarding the formal constitution of clinical research training. Research training in psychiatric residency is also variable. Nevertheless, common best practices are apparent from reviewing existing programs and published descriptions. Most programs offer research training in the latter years of residency, and even the most research-intensive in-
OCR for page 128
Research Training in Psychiatry Residency: Strategies for Reform stitutions route their research-oriented graduates toward additional training, usually in the form of a fellowship. Hands-on activity in residency is encouraged when resources and mentoring are available. Key course subject matter includes epidemiology, grant and manuscript writing, design of clinical trials, and research ethics. Unfortunately, little has been done to integrate research training into all or even most of the residency years. Additionally, existing curricula are not typically validated by careful long-term follow-up studies to determine whether trainees actually were encouraged to move into patient-oriented research careers, or toward more evidence-based practice methods. Therefore, the committee makes the following recommendation: Recommendation 4.3. The National Institute of Mental Health, foundations, and other funding agencies should provide resources to support efforts to create competency-based curricula for research literacy and more comprehensive research training in psychiatry that are applicable across the spectrum of adult (general) and child and adolescent residency training programs. Supported curriculum development efforts should include plans for educating faculty to deliver each new curriculum, as well as plans for evaluating each curriculum’s success in training individuals to competency and in recruiting and training successful researchers. On the federal level, the K30 mechanism is an obvious means of supporting some curriculum development, although it does not have provisions for stipend support and is rarely utilized by medical residents. The AACAP research pathways are, to the committee’s knowledge, among the best models generated to date for creating and evaluating an exportable model for training psychiatrist-researchers, in this case targeting those in child and adolescent psychiatry. Such efforts should be extended to various other settings, including resource-poor departments and those that emphasize a given subspecialty of psychiatric practice (e.g., psychotherapy, addiction, pain management). These curricula should be aimed at sparking residents’ interest in a lifelong career in patient-oriented research without interfering with core clinical training. The principal aim of this recommendation, however, is to ensure that all residents are adequately introduced to the concepts of research and that research training is not merely an afterthought to residency education. Thus the recommendation is focused on ensuring a foundation in the
OCR for page 129
Research Training in Psychiatry Residency: Strategies for Reform residency curriculum for patient-oriented research efforts. Even residents who intend to become clinicians should be introduced to the concepts and findings of patient-oriented research as a necessary complement to their clinical training. Curricula should be developed using established educational principles; it is especially important to include evaluation phases to verify the utility of the curricula in the training of patient-oriented psychiatrist-researchers and evidence-based practitioners (Sheets and Anderson, 1991). Novel ways to integrate research training into the residency experience for future clinicians and the next generation of independent investigators should also be considered. For example, Duke University is currently experimenting with a program that introduces research activity in PGY1 rather than waiting until later in the residency (list serve communication,29 G. Thrall, Duke University, January 12, 2003). With regard to curriculum development, the committee believes that, since psychiatric training programs vary considerably in terms of size and local expertise, they should be viewed along a hierarchical research training continuum that ranges from those providing only research literacy to those training large numbers of patient-oriented psychiatrist-researchers. The committee proposes such a continuum in Table 4-3 (see page 131). An important feature of this continuum is the detail it provides about program components (e.g., longitudinal participation in research) and the corresponding department infrastructure (e.g., mentors and existing grants) necessary to achieve various levels of research training. The schema represented in Table 4-3 shows how individual programs can consider their current infrastructure and build on their clinical and research strengths to enhance research training. For example, the presence of a large substance abuse clinic could be used as the foundation for a grant application to establish a research or research training effort in substance abuse, thereby advancing the program along the research training continuum. The continuum additionally is intended as a tool that can be used to implement the following recommendation: Recommendation 4.4. The National Institute of Mental Health should support those departments that are poised to improve their residency-based research training to achieve measurable increases in patient-oriented research careers among their trainees. Support for such programs should include funds to: 29 The list serve is maintained by the AADPRT.
OCR for page 130
Research Training in Psychiatry Residency: Strategies for Reform Hire faculty and staff dedicated to research and research training efforts. Acquire equipment and enhance facilities for research training. Initiate pilot and/or short-term research activities for residents. Educate adult and child and adolescent residency training directors and other faculty in how to promote and guide research career planning. This recommendation aims to encourage NIMH to enhance the resources and environment of programs that can realistically advance their training efforts to the next level on the research continuum set forth in Table 4-3. A request for applications would best call for proposals from across the continuum, with the aim of funding a few programs at each of the three delineated levels (i.e., purely clinical, moderate research training, superior research training). Review committees for such grants would be instructed to rank applications on the basis of each program’s ability to demonstrate a plan for moving to and sustaining a higher lever of research training. At the bottom end of the continuum, programs would be expected to instill research literacy in their residents. Programs would also be expected to encourage their residents to transfer to other institutions (after 3-years of training) and aim for research fellowships to optimize their research training; for weaker programs, some altruism would be required if they did not have the local infrastructure to support a promising trainee. Regarding the details of this recommendation, the first three bullets listed are linked quite directly to developing a research infrastructure. NIH or other agency grants—similar to the General Clinical Research Center or Biomedical Research Infrastructure Network grants—might be useful to this end. The expired Research Infrastructure Support Program (RISP), which still exists to help minority-based programs develop a foundation (see Chapter 5), is clearly a direct model for what is implied by this recommendation. The RISP was “…designed to enable institutions with relatively small but viable research programs…to develop into significantly stronger…research settings” (NIMH, 1994:2). That mechanism included possible support for: salaries, research training for junior investigators, and research instruments/equipment. One current RISP program announcement calls for applications for the funding of mental health services research at primarily clinical facilities. An important component of that announcement is that small programs are encouraged
OCR for page 131
Research Training in Psychiatry Residency: Strategies for Reform Table 4-3 Continuum of Residency-Based Research Training
OCR for page 132
Research Training in Psychiatry Residency: Strategies for Reform to develop direct collaborations with research-intensive institutions (NIMH, 2000c). This and other similar programs should be developed to improve research education at psychiatry training facilities. With regard to bullet number 3 under recommendation 4.4, pilot or short-term funding could be utilized opportunistically by departments to facilitate the inclusion of more residents in research training. This is the case because residency is typically a career phase that permits limited and transient opportunities for the pursuit of nonclinical interests. A modest, but available pool of pilot funding might be used to support one or more training slots or other research-related resources to accommodate qualified and motivated residents. The final item listed under recommendation 4.4 addresses the need to provide training directors and faculty with adequate instruction in guiding and nurturing potential researchers. Models at NIMH already exist in the form of seminars for K awardees (Tuma et al., 1987). Similar “retreats” for residency training directors and/or vice chairs of research could facilitate the flow of information on research training grants and other relevant matters to those most responsible for training residents. This recommendation also encourages the expansion and utilization of other means of information dissemination. These mechanisms include web-based resources, such as the NIH K Kiosk, which allows one to search and review various mentored career awards (NIH, 2003e). They further include on-line tutorials, such as one that currently exists on protecting the rights of research subjects (NIH, 2003f).
Representative terms from entire chapter: