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 133
Research Training in Psychiatry Residency: Strategies for Reform 5 Personal Factors The last two chapters focused on external factors that impact on a resident’s decision to pursue research training. This chapter turns to more intrinsic or personal factors that influence such career choices. It briefly reviews innate characteristics that correlate with the decision to pursue a research career, and then personal financial issues that impact on research career initiation and development. Finally, it addresses gender, racial, and ethnic issues as factors relevant to training major subsets of psychiatric trainees. Included is a discussion of issues faced by foreign medical graduates who matriculate into psychiatric residency programs in the United States. The chapter ends with conclusions and recommendations. INNATE CHARACTERISTICS Several key personal characteristics correlate with the decision to pursue a research career. They include motivation and drive, and intellectual capacity and scientific orientation.
OCR for page 134
Research Training in Psychiatry Residency: Strategies for Reform Motivation and Drive Some of the personal factors that move one to pursue a research career transcend or precede formal medical educational experiences. Motivation and drive, or rationale and persistence, are certainly relevant in the pursuit of any complex goal. Formal education and mentoring may have some impact on these characteristics, but other, less tangible and less malleable factors are likely to be relevant, if not dominant. Personal experiences that motivate one toward a research career include direct or familial experience with mental illness. Such is the case for genetics researcher Edwin Cook who said, “I do [autism research] because…I always wanted to know what was wrong with my brother and to help him….” (National Public Radio, 2002). Alternatively, one may have extraordinary curiosity and skill that lead to a productive research career despite the absence of direct support and encouragement in the context of formal medical training. This was the case with Eric Kandel (1998), a psychiatrist and Nobel Prize winner for his neuroscience work, who recently wrote that his psychiatric residency involved very little scholarly activity and virtually no research training. Despite these omissions from his training, he and many of his peers went on to become successful basic and patient-oriented researchers, evidence that skilled researchers can emerge from residency programs with little or no research training. These examples are presented as a reminder of two principles. First, certain characteristics that correlate with research productivity are difficult if not impossible to shape in the context of formal medical training. Second, it is wise for any field to identify, support, and attempt to attract the brightest and most driven candidates, and to encourage them to pursue a career aimed at critiquing and expanding that discipline’s knowledge base. Intellectual Capacity and Scientific Orientation In addition to motivation and drive, intellectual capacity and scientific orientation are logically correlated with research productivity. Research requires the ability to master an existing and constantly expanding knowledge base, and to formulate and test new ideas in an effort to clarify, broaden, or even revise what is known about the subject under study. Additionally, research productivity is dependent upon regular and detailed oral and written communications with peers as a key means of
OCR for page 135
Research Training in Psychiatry Residency: Strategies for Reform validating the accuracy and relevance of an experiment or theory. In psychiatric research, it is reasonable to assume that such skills correlate, at least partially, with scores on standardized tests used to evaluate applicants to medical school (Medical College Admission Test [MCAT]) or to verify that medical students and graduates are prepared for independent practice (U.S. Medical Licensing Examination [USMLE]). To the extent that these proxies for research aptitude are valid, there is some indication that the discipline of psychiatry may not be attracting the brightest students, and from this one can infer that inadequate recruitment of the brightest students may negatively impact on psychiatry’s ability to expand its ranks of patient-oriented researchers. Sierles and Taylor (1995) reviewed literature on medical students’ interest in psychiatry and found that, as of the late 1980s, those favoring psychiatry tended to have relatively low science scores compared with their peers who were interested in other medical specialties. A more recent review of MCAT and USMLE scores by economist Sean Nicholson of the Wharton Business School further supports that conclusion (Arcidiacono and Nicholson, Unpublished; Nicholson, 2002). Nicholson’s findings are based on 1996–1998 survey and exam score data from the National Board of Medical Examiners for approximately 33,000 medical school students. Nicholson found that fourth-year medical students who chose psychiatry as their specialty had the lowest average scores on their preclinical USMLE (Step 1, a measure of basic medical science knowledge) compared with those selecting 15 other specialties (see Table 5-1). He also found that those aiming to pursue a career in psychiatry upon entering medical school ranked ninth on the MCAT. The higher ranking of these students on the MCAT may be related to the fact that this test includes more verbal/social science content than the USMLE, or that some of the more successful test takers favoring psychiatry in the first year of medical school changed their specialty selection 3 years later at the time of the USMLE (see Table 5-1). What is not clear from Nicholson’s analysis is whether the apparent differences in exam scores represent normally distributed samples of training physicians or the distributions are multimodal (e.g., those choosing psychiatry may be composed of one group with relatively high scores and another group with relatively low scores). Additionally, Nicholson’s analysis is limited because 30 percent of the universe of more than 47,000 medical graduates from 1996–1998 did not complete one of the surveys or otherwise had incomplete data records (Arcidiacono and Nicholson, Unpublished). Accordingly, the analysis is incomplete; nonetheless it represents the best comparison of early and later training examination scores that the committee was able to identify for this report.
OCR for page 136
Research Training in Psychiatry Residency: Strategies for Reform TABLE 5-1 Examination Scores by Specialty Choice upon Entry to Medical School (MCAT) and upon Entry to Residency (USMLE) Specialty Choicea Mean MCAT Score (rank) Mean USMLE (Step 1) Score (rank) Ear, Nose, and Throat (ENT) 28.4 (5) 224.4 (1) Dermatology 26.6 (17) 219.2 (2) Surgical Subspecialty 28.6 (3) 218.4 (3) Orthopedic Surgery 28.4 (5) 218.1 (4) Urology 27.5 (14) 215.7 (5) Radiology 28.7 (2) 213.8 (6) General Surgery 28.3 (7) 213.4 (7) Ophthalmology 28.0 (9) 213.4 (7) Pathology 27.6 (12) 213.3 (9) Internal Medicine 28.5 (4) 212.3 (10) Emergency Medicine 28.1 (8) 211.4 (11) Obstetrics-Gynecology (OB-GYN) 27.1 (15) 208.1 (12) Pediatrics 27.6 (12) 207.4 (13) Anesthesiology 27.1 (15) 206.8 (14) Family Practice 27.8 (11) 204.8 (15) Psychiatry 28.0 (9) 204.2 (16) Undecided 28.8 (1) N/A Total 28.2 210.5 NOTES: Data are from 33,110 medical students who graduated from medical school from 1996 to 1998. The sample represents a subset of more than 47,000 graduates during that time period. Individuals were excluded because of missing data or because they refused to consent to having their information used for research purposes. MCAT = Medical College Admission Test; USMLE = U.S. Medical Licensing Examination; N/A = not available. aChoice when polled at 1st year of medical school with respect to MCAT scores, and at 4th year of medical school with respect to USMLE scores. SOURCE: Arcidiacono and Nicholson (Unpublished), Nicholson (2002). Using 1992 Association of American Medical Colleges (AAMC) survey data, Nicholson also found that medical students who considered research an important factor in determining their choice of medical specialty had higher MCAT scores than those who did not consider research an important factor. In further support of the link between MCAT scores and research aspirations, albeit without any statistical validation, data
OCR for page 137
Research Training in Psychiatry Residency: Strategies for Reform from a 1994 questionnaire on over 10,000 medical school graduates showed that the more than 1,000 individuals with research interests had higher science scores (mean = 10.32) on their MCAT than their colleagues who were uninterested in research (mean = 9.22) (Kassebaum et al., 1995). These data fall well short of full confirmation that high standardized test scores on either the MCAT or USMLE correlate with successful patient-oriented research careers, but they do support the concern expressed by some in psychiatry that the discipline is currently not attracting the brightest medical students (Hyman, 2002b; Meyer, 2002). In addition to attracting applicants with somewhat lower scientific aptitudes than other branches of medicine, psychiatry has historically attracted those with a predisposition for the social sciences and humanities (Fishman and Zimet, 1972; Lee et al., 1995; Nemetz and Weiner, 1965; Paiva and Haley, 1971). This view was reinforced by John March, Director of Programs in Child and Adolescent Anxiety Disorders and Developmental Psychopharmacology, Duke University, who said in his presentation at the committee’s workshop: Psychiatry has a very strong humanistic culture…unlike the rest of medicine, which has gotten very technological…. So you find folks that are drawn to psychiatry who don’t have the kind of minds that tend to like scientific reductionistic reasoning (March, 2002). This apparent recruitment bias is, in certain ways, good for patient-oriented research as it likely enhances the doctor–patient relationship. However, there is also evidence that some medical students avoid psychiatry because they believe it is a discipline with a limited scientific basis (Feifel et al., 1999). This latter issue may be linked to psychiatry’s difficulty in attracting medical students with the scientific orientation necessary to function as successful biomedical researchers. Difficulties in recruiting and retaining individuals in the field may also be the result of the stigma society attaches to mental illness and persons with mental illness. It is reasonable to assume that medical students may harbor some of these same negative perceptions about people with mental illness and about the medical professionals who treat them. For example, a 2001 survey of second-year medical students at the University of Arkansas found that more than 30 percent believed electroconvulsive therapy (ECT) was used as a form of punishment, and 40 percent believed psychiatrists did not use ECT appropriately (Clothier et al., 2001). This rather dramatic misperception is in stark contrast to the realities of ECT (U. S. Department of Health and Human Services [DHHS], 1999).
OCR for page 138
Research Training in Psychiatry Residency: Strategies for Reform Furthermore, psychiatry has long been a profession for which cultural stereotypes abound, and most of those stereotypes are not altogether positive. As discussed in Chapter 4, psychiatrists engaged in patient-oriented research are underinvolved in medical student and resident education. Therefore, medical students’ exposure to psychiatrists frequently does not include exposure to psychiatrists doing exciting research that addresses intellectually challenging, interesting, and clinically relevant issues. The media frequently depict mental health professionals in general and psychiatrists in particular as inept or eccentric, rather than bright, well-trained medical professionals who are addressing extremely important health care issues (British Broadcasting Company, 2002). Clearly, one way to redress this negative stereotyping is to ensure that medical students are exposed to the best and brightest psychiatry has to offer, including successful patient-oriented researchers. PERSONAL FINANCIAL CONCERNS Quite apart from the innate characteristics discussed above, medical students and psychiatric residents have financial concerns that intensify as they progress through their training. This section describes those concerns as they pertain to entry into residency and possible pursuit of a research career. Student Debt Trainees incur substantial educational debt while in medical school. In 2002, the cost of attending a public medical school averaged nearly $28,000 per year, while that of attending a private medical school averaged $44,000 (AAMC, 2002c). Data from 2000 to 2002 indicate that over 80 percent of medical students take out loans to finance their medical education (AAMC, 2002c). In 2002, average debt among medical school graduates stood at $104,000 (AAMC, 2002b; 2002c; Sung et al., 2003). After adjusting for inflation, the median level of debt for medical school graduates doubled between 1985 and 1998 (Proctor, 2000; Zemlo et al., 2000). Further investigation supports the hypothesis that debt load has an impact on career decisions. Data from the early 1990s, for example, demonstrate that medical students with high debt favored high-paying specialties over primary care disciplines (e.g., pediatrics and family
OCR for page 139
Research Training in Psychiatry Residency: Strategies for Reform medicine) (Berg et al., 1993; Colquitt et al., 1996). Given that mean salaries for psychiatrists fall in a range comparable to that for the primary care specialties (AAMC, 2002b; Bureau of Labor Statistics, 2002; Nicholson, 2002), it is plausible that students with high debt will avoid psychiatry in favor of a more lucrative specialty so they can repay their educational debt more easily (see the discussion below). Residency and Fellowship Stipends Compensation for residents and fellows is low in comparison with entry-level salaries. Data for 2002 from New York State place entry-level salaries for new graduates of psychiatric residencies at $124,000 and $144,000 for adult and for child and adolescent psychiatrists, respectively (Nolan et al., 2003). In comparison, first-year postgraduate (PGY1) residents at Columbia University earn $43,000, and research fellows (PGY5 to PGY8) receive stipends of approximately $75,000 (personal communication, R. Rieder, Columbia University, March 31, 2003). Since New York State supplements Medicare graduate medical education funds with state funds, other regions where state support is not available will likely compensate residents at even lower levels. The shortest core residency training period accredited by the Accreditation Council for Graduate Medical Education (ACGME) is 3 years (e.g., general internal medicine, pediatrics). Additional residency or research fellowship training beyond that point is not necessary for core certification (e.g., to take the psychiatry boards), but rather requires extended training and only sometimes yields additional certification. Thus, a recent medical school graduate may weigh the value of additional training against the short-term loss in income that is represented by differences between training stipends and entry-level salaries, differences that can dissuade a new physician from pursuing psychiatry or research training. Salaries of Psychiatrists Compared with Other Specialties Psychiatrists, especially child and adolescent psychiatrists, earn less than many other medical specialists with similar post–medical school training periods (see Table 5-2), although it is notable that among 4-year residencies, pathologists start at lower salaries than those of adult psychiatrists. Among 5-year residencies, child and adolescent psychiatrists have lower starting salaries than urologists, general surgeons, radiologists, and orthopedic surgeons. Overall salary structure does not favor psychiatry ov-
OCR for page 140
Research Training in Psychiatry Residency: Strategies for Reform TABLE 5-2 Median Expected Starting Income for Graduates of New York State and California Residency Training Programs, 2002 Specialty Minimum Number of Years of Post-graduate Medical Education Median Expected Salary (2002) (thousands of dollars) New York California Anesthesiology 4 194 206 Dermatology 4 155 148 Ophthalmology 4 133 137 Neurology 4 126 138 Adult Psychiatry 4 124 120 Pathology 4 118 112 Orthopedics 5 225 208 Radiology 5 218 198 General Surgery 5 157 164 Urology 5 155 206 Child Psychiatry 5 144 140 SOURCE: ACGME (2002a), Nolan et al. (2003). er many other medical disciplines, a reality that appears to influence career decisions. To further understand the impact of income expectations and salary on an individual’s choice of specialty, the committee again turned to the work of Sean Nicholson at the Wharton School of Business. Using 1992 data from the National Residency Matching Program (NRMP) and the AAMC, Nicholson (2002b) considered how specialty selection is influenced by specialty income. He found a direct and significant correlation between specialty income and relative demand for a given specialty among medical students.30 Figure 5-1 summarizes this finding graphically. What is apparent from this graph and from the evidence presented 30 Relative demand for a given specialty is defined as the ratio of the number of graduating medical students selecting a specialty as their first choice, divided by the total number of national slots available in that specialty.
OCR for page 141
Research Training in Psychiatry Residency: Strategies for Reform above is that psychiatry is one of the lowest-paying specialties in medicine, and that this relatively low compensation may impact negatively on recruitment to the discipline’s residency. At the same time, income correlations alone likely oversimplify the importance of economics to medical students selecting a specialty. A 2003 study found that medical students were apt to consider specialties based on, from most to least important, expected annual work hours, length of residency training, weekly hours worked, and earnings (Thornton and Esposto, 2003). Thus it appears that medical students value time as well as money. Salary Differences Between Clinicians and Researchers In addition to high student debt, low stipends during residency and fellowship, and lower salaries in comparison with those of other medical specialties, psychiatry trainees interested in research may face the possi- FIGURE 5-1 Ratio of demand to supply of medical students who designate a given residency as their first career choice, versus the expected lifetime income of that residency. Based on 1992 National Residency Matching data for 14,030 U.S. medical graduates. Supply is the total number of national slots available in that specialty. SOURCE: Data from Nicholson (2002).
OCR for page 142
Research Training in Psychiatry Residency: Strategies for Reform bility that they will not receive the same compensation as their clinician counterparts. There appears to be no published literature on the salary differences between psychiatrists who are researchers and those who are clinicians. However, the committee was able to find published data comparing the salaries of researchers and clinicians in psychology. In a 2001 survey of members of the American Psychological Association, the mean salary for new31 doctoral-level psychologists who rated themselves as full-time researchers (both basic and clinical) and who were not faculty at academic institutions was $58,000 (n = 126).32 For new31 clinically-licensed psychologists who claimed direct mental health care as their primary activity and who also were not based at an academic institution (n = 175), the mean salary was $63,500.32 For faculty at academic institutions, both researchers and non-researchers, with the rank of lecturer, instructor, and assistant professor (n=599), the mean salary was $61,000 (Singleton et al., 2003). 32 Although this survey did not isolate clinical or other psychologists engaged in part-time research, nor does it permit full adjustments for years of professional experience, the data support the hypothesis that psychologists engaged predominantly in research, at least in their early professional years, earn lower base salaries than their clinician counterparts. It is plausible that this salary differential extends to other clinical behavioral disciplines, such as psychiatry. To consider psychiatrists directly, the committee commissioned the work of economist Douglas Schwalm of Louisiana State University, as his dissertation (2002b) focused on the impact of managed care and research activity on psychiatrists’ incomes. At the request of the committee, Schwalm (2002a) adjusted his analysis to consider specifically the impact of research involvement on those incomes. He examined data from the 1998 National Survey of Psychiatric Practice, conducted by the American Psychiatric Association (APA). Of the 1,500 APA member psychiatrists surveyed in that year, 1,076 responded, and 628 provided annual income information. Of the subsample providing annual income data, 112 (18 percent) reported some involvement in research activities; these included 26 female respondents.33 The basic statistical model regressed the log-income to a quadratic of years of experience, gender, research, and work setting.34 After removing the effect of work setting, 31 “New” is defined as having 0 to 9 years experience 32 Salary extrapolated to 12-month equivalent. 33 Psychiatrists were considered researchers if they self-reported themselves as such and if they spent any time in research (i.e., no minimal cut-off was used to define a researcher). 34 Such log-income regression models are based on the human capital model devised by Gary Becker, which has been used extensively to estimate the effects of several factors, including income, on career choice (Becker, 1993; Mincer, 1974).
OCR for page 143
Research Training in Psychiatry Residency: Strategies for Reform which is nonsignificant, and adjusting for the significant effects of years of experience and gender, Schwalm found that on average, psychiatrist-researchers earn approximately 20 percent less than their nonresearcher counterparts. Assuming a retirement age of 70, this loss of income translates to a reduction in lifetime earnings of $600,000 to $1 million (1998 dollars). For women, the annual salary difference between researchers and nonresearchers is negligible; however, the extrapolated hourly wage favors female clinicians over researchers (see Table 5-3). The modeling results further indicate that, while psychiatrists who spend any (i.e., greater than 1 percent) of their time engaged in research activities earn less than their full-time clinician counterparts, psychiatrists who devote more than 20 percent of their time to research earn more than those who spend a smaller proportion of their time engaged in research activities, although their salary never recovers to the level of full-time clinicians. There are two key limitations to the above analysis. First, researchers in this dataset are disproportionately at academic institutions (e.g., medical schools), and thus their salaries are likely lower in part because of TABLE 5-3 Selected Data from the American Psychiatric Association’s 1998 National Survey of Psychiatric Practice Full-Time Employed Psychiatrists (n = 628a) Males Females Nonresearcher Annual Salary $143,000c $111,000 Researcher Annual Salary $121,000 $114,000 Nonresearcher Hours Workedb 52 44 Researcher Hours Workedb 56 51 Nonresearcher-Implicit Waged $55/hour $50/hour Researcher-Implicit Waged $43/hour $45/hour aOf these respondents, 112 had spent some time as researchers, including 26 females and 31 non-Caucasians. bRespondents gave the number of hours they worked in the week prior to the time at which they completed the survey. cSignificantly different from male researchers (p ß0.01). dImplicit wage = annual salary/(hours of work × 50). SOURCE: Data adapted from Schwalm (2002a).
OCR for page 156
Research Training in Psychiatry Residency: Strategies for Reform then required to return to their country for 2 years before applying for permanent status in the United States (Baer et al., 1998) unless they agree to serve in areas with unmet health professional needs (Baer et al., 1999). Third, IMGs on temporary or student visas are ineligible for grants such as National Research Service Award fellowships and training grants and mentored career awards (NIH, 2000a; 2002b; 2003c). The result is that talented and ambitious IMGs may be prevented from enrolling in research training programs. This situation may contribute to a substantial loss in the number of patient-oriented researchers not only in this country, but also in other countries to which these individuals might potentially contribute should they decide to return home or emigrate elsewhere. The committee agrees with the utility of screening non-U.S. medical graduates for basic skills; once these requirements have been met, however, outstanding residents should be given full opportunities to engage in research training and to contribute to psychiatric research, as many foreign IMGs have made important contributions in the past (Balon et al., 1999). Akin to allowing IMGs to serve in geographic areas with a need for health professionals, Congress should consider designating specific areas of research training and patient-oriented research activities as federally underserved disciplines, thereby permitting individuals who have demonstrated the necessary aptitude and ambition to train as patient-oriented researchers. The committee is fully aware that the current political climate makes such a suggestion controversial, but believes it to be appropriate given the previous success experienced by psychiatry and other fields of medicine with IMG researchers and the need for more physicians to be engaged in patient-oriented research. Underrepresented Racial and Ethnic Minorities Unlike women and IMGs, certain racial and ethnic minorities, including African Americans, Hispanics, and Native Americans, represent a limited proportion of psychiatrists in general and investigators in particular. As noted in Table 5-4, in 1999 African Americans and Hispanics composed just over one-quarter of the U.S. population and just over 11 percent of psychiatric residents. In that same year, however, fewer than 6 percent of all applicants for NIMH funding and fewer than 4 percent of all applicants receiving such funding came from those minorities (National Advisory Mental Health Council’s Workgroup on Racial/Ethnic Diversity in Research Training and Health Disparities Research, 2001).
OCR for page 157
Research Training in Psychiatry Residency: Strategies for Reform Minorities may not engage in research activity for at least two reasons. First, minorities, and particularly African Americans, tend to be suspicious of the mental health system, as they are at increased risk of being misdiagnosed, committed or incarcerated, or inappropriately medicated (DHHS, 2001a; IOM, 2003; Lawson, 1996; 2000). Second, as with women, there is also a shortage of ethnic and racial minorities who can serve as research role models for medical students. Data indicate that African Americans, Hispanics, and Native Americans are underrepresented as full-time faculty in health science, natural science, and social science departments (National Advisory Mental Health Council’s Workgroup on Racial/Ethnic Diversity in Research Training and Health Disparities Research, 2001). Data from 2001 indicate that fewer than 5 percent of full-time medical faculty with an M.D. are underrepresented minorities (The Robert Wood Johnson Foundation, 2003). During most of the 1990s, there was only one African American psychiatrist with a career development award or an R01 grant, and that number has since increased only marginally.36 Additionally, African American students are disproportionately trained at historically black colleges and universities, which before 2001 received no federal funding for psychiatrists (National Advisory Mental Health Council’s Workgroup on Racial/Ethnic Diversity in Research Training and Health Disparities Research, 2001). Given that almost 29 percent of African American medical school graduates attended such colleges and universities from 1950 to 1998 (IOM, 2001b), it is imperative that these institutions develop and promote their research infrastructure to ensure an adequate supply of minority mental health researchers who can serve as role models. As noted during a recent workshop on minority training hosted by NIMH: With a lack of networking/mentoring comes a lack of advice, of acknowledgment, considerations for positions, publications and funding. Additionally, appropriate mentors are needed who understand and can help students make the language ‘shift’ from their heritage and culture to the ‘mainstream.’ (NIMH, 1999:7) Although racial and ethnic minorities face a number of obstacles in becoming researchers, there are also numerous opportunities for members of these groups to receive research training support. However, these opportunities appear to be insufficient to attract large numbers of appli- 36 Based on the committee’s review of the Computer Retrieval of Information on Scientific Projects (CRISP) database.
OCR for page 158
Research Training in Psychiatry Residency: Strategies for Reform cants. Research training grants and fellowships often target minorities. Additionally, the National Institutes of Health (NIH) funds minority supplement grants to encourage senior investigators to mentor minority trainees in the context of their ongoing work (National Advisory Mental Health Council’s Workgroup on Racial/Ethnic Diversity in Research Training and Health Disparities Research, 2001). Success rates among applicants for such grants are very high: of 51 applications for NIMH minority supplement grants submitted in 1998, 100 percent were funded, for a combined total of almost $5.5 million (NIMH, 2000a). Several professional societies in psychiatry also offer research training opportunities for minorities. Other minority funding opportunities include an NIMH-supported T32 grant first received by the APA in 1989 to develop a program aimed at minority research training in psychiatry (see Chapter 2). The American Academy of Child and Adolescent Psychiatry also has funding programs targeting scholarships and fellowships for minorities (see Appendix B for additional information on funding opportunities). Recognizing the need for mentor support, a recent NIMH report recommends the creation of a national mentoring network of senior minority and nonminority investigators to develop extended relationships with minority trainees and investigators (National Advisory Mental Health Council’s Workgroup on Racial/Ethnic Diversity in Research Training and Health Disparities Research, 2001). Mentoring for minorities has already been addressed directly at some universities. For example, a 1998 review of the mentoring program of the University of Pennsylvania found that minority physicians were unsuccessful at finding research mentors, and those minority physicians who were mentors were overburdened with the additional responsibilities of serving on committees and task forces to ensure minority representation (Johnson et al., 1998). To remedy the situation, the university aggressively recruited minority faculty such that between 1993 and 1997, their numbers increased by 32 percent. This hiring effort was coupled with general counseling, research development, training in research methods, and grant writing (Johnson et al., 1998). Similarly, to increase the number of minority faculty, Harvard University created the Minority Faculty Development Program in 1991 to encourage medical students to consider careers in academia and to promote career development of junior faculty (Curry, 1997; Potts, 1992). Yet another important example of strategies to build a critical mass of minority researchers is NIMH’s use of the R24 mechanism to award Minority Research Infrastructure Support Program (M-RISP) grants to a number of institutions to support mental health research and individual investigator/faculty development. The M-RISP is awarded to institutions with a significant number (greater than 30 percent) of racial and ethnic
OCR for page 159
Research Training in Psychiatry Residency: Strategies for Reform minority students or to a Native American tribe that applies in collaboration with a mental health training institution. Awards are made for 3 to 5 years, depending on whether the grant application is new or renewable, at a maximum direct cost of $400,000 (NIH, 2000c). Thus it is apparent that there is considerable funding for minority researchers. However, many of those targeted by these opportunities either are not aware of the funding or lack the core resources that would allow them to prepare an application. Outreach efforts to educate minority psychiatrists about these grants would likely increase those applications. Furthermore, it is important that fledgling researchers receive guidance and mentoring from senior psychiatrist-researchers to ensure adequate recruitment and retention. CONCLUSIONS AND RECOMMENDATIONS Individuals considering patient-oriented research in psychiatry are likely influenced by a number of personal factors. Some of these factors are so intrinsic that it is difficult to imagine the formal educational process, especially in adulthood, having much impact on them. For example, perhaps the best any discipline can do with regard to candidates having exceptional drive or talent is to encourage them toward that specialty. There is some evidence that psychiatry may not be competitive in attracting the top medical students, despite unparalleled opportunities in the clinical brain and behavioral sciences. Part of the problem may be the compensation of psychiatrists as compared with that for other disciplines. The anticipated salaries for psychiatrists, whether academic or practicing, are near the bottom of the physician pay scale. Accordingly, concerns about loan repayment and overall financial well-being may discourage potential patient-oriented researchers from extended research training that would further delay them from achieving their full earning potential. Accordingly, one obvious recommendation the committee makes is as follows: Recommendation 5.1. The National Institute of Mental Health and other funding agencies should seek mechanisms to offer increased financial incentives, such as loan repayment, to trainees who commit to research training and research involvement beyond core psychiatry residency.
OCR for page 160
Research Training in Psychiatry Residency: Strategies for Reform The committee encourages these agencies to be creative in designing ways to increase or otherwise supplement training stipends for promising psychiatry trainees. Maintaining and expanding loan repayment programs will likely be an important part of this strategy. As early as 1989, NIH offered debt-repayment grants to physician-researchers. NIH intramural and extramural debt-relief grants have included those targeting certain diseases (e.g., HIV, infertility), patient populations (e.g., children, minorities), and clinical research specifically (IOM, 1994; Ley and Rosenberg, 2002; Nathan, 2002; NIH, 2002a). In 2002, NIMH budgeted $2.5 million for its loan repayment program to support 53 trainees and junior faculty. Recipients receive a maximum of $35,000 each year for up to 2 years, during which time they commit to at least 20 research hours each week. The budget was doubled for 2003, and eligibility was expanded to include those who do not have an NIH grant. In 2002, 82 percent of all applicants to the NIMH program were funded, this high rate likely being due to the fact that the program was new and thus not broadly known among potential applicants (NIMH, 2002d). What remains to be seen is whether these grants will entice more individuals to become researchers or simply reward those already on the research track (Pardes, 2002). Despite the absence of direct evidence, it is axiomatic that debt relief will attract some young psychiatrists to research, and survey data support this hypothesis. For example, a recent web-based survey of 86 neurology residents found that loan repayment incentives would encourage 76 percent of those residents to consider extended postresidency research training that included “the expectation of future practice in academic neurology” (Doherty et al., 2002). Given that 63 percent of this sample intended to become academics, it is most important that 15 of 17 individuals who expressed uncertainty about an academic career and 8 of 15 who had previously ruled out such a path responded favorably to the idea of research training if it were supported with loan repayment assistance. Loan repayment programs are likely to be attractive to trainees with considerable debt, particularly members of underrepresented minorities, who typically graduate with higher levels of debt than students overall (NIMH, 1999; National Research Council [NRC], 2000; Spar et al., 1993). Furthermore, the Surgeon General’s report on mental health notes that African Americans have only one-tenth the net worth of Caucasians (DHHS, 2001a). A complementary strategy to loan repayment is for funding agencies to supplement research training stipends as they are well below the income levels one could expect from direct entry into clinical practice. In 2003, for example, the annual stipend for someone in their fifth year of
OCR for page 161
Research Training in Psychiatry Residency: Strategies for Reform training after medical school was just over $44,000 (NIH, 2003d), well below the entry level salary of a junior psychiatrists which are double or triple that amount (see Table 5-2). Accordingly, the committee believes that increasing compensation represents an important and obvious way to encourage more residents to undertake additional training before entering their first full-paying position. Stipend amounts for training are determined largely under the authority of the Secretary of Health and Human Services. The remuneration is designed principally to defray living expenses incurred during training, and thus federal legislation dating back to 1974 requires M.D. and Ph.D. fellows to receive the same stipend levels (NRC, 2000). This structuring means that stipend increases for psychiatrists or even physician-investigators in isolation are not likely under the current system. Provisions for supplementing income, via moonlighting or non-federal sources, however is permitted by NIH regulations (NIH, 2001a). Columbia University, for example, has several fellowship programs, and the state provides funding to increase training stipends (Rieder, 2001; 2003). The department of neurology at the University of Rochester also permits research fellows to negotiate their compensation arrangements based on their productivity in applying for extramural funding (Griggs, 2002). Alternatively, some institutions allow research training to overlap with junior faculty status as an incentive to potential researchers. Foundations, philanthropists, and other third-party supporters also represent important potential sources of funding for financial incentives to trainees committed to patient-oriented research careers. In the case of residents, whose time is limited by the demands of clinical training, department funds might be used to increase residency compensation by creating “research moonlighting” opportunities, a practice currently in place at Columbia University (personal communication, R. Rieder, Columbia University, April 2003). At Columbia, research moonlighting typically engages one resident per year and involves chart review or some other research activity that can be carried out on a part-time basis. To the extent that such arrangements lead to a meaningful product and give residents hands-on experience, they may serve the dual purpose of training and of supplementing personal income. Ultimately, recommendation 5.1 is about investing in the trainee specifically—not to make training compensation equal to the pay one would receive as an independent practitioner or investigator, but to increase the probability that a few more young psychiatrists will prolong their full-time educational experience to pursue the goal of developing into an independent patient-oriented researcher.
OCR for page 162
Research Training in Psychiatry Residency: Strategies for Reform Although financial expectations play a role in their career decisions, medical students and residents are equally if not more concerned about other issues, such as lifestyle and the intellectual content of their selected profession. Furthermore, new physicians are understandably anxious about the challenges involved in securing long-term research funding. The committee observed that in recent years, several well-respected editorial writers have written extensively about the problems associated with clinical research without promoting the endeavor as an exciting option for new physicians (Lieberman, 2001; Schrier, 1997; Shine, 1998). The preface to a recent opinion piece by an established psychiatrist-researcher, for example, notes that research careers are not glamorous, may not be intellectually exciting, are tedious, and typically involve considerable delayed gratification (Lieberman, 2001). While such characterizations are partially true of most hard-earned achievements, the committee is concerned that they may inappropriately overshadow the positive aspects of a research career. Thus we make the following recommendation: Recommendation 5.2. Individuals and institutions involved in the education and mentoring of medical students, residents and fellows should strongly convey to these trainees the benefits (professional and societal) associated with patient-oriented research in psychiatry. Promotion strategies might include support for student interest groups; brochures, websites, and other media; and summer research training opportunities. This recommendation is based on two principles: first, research offers a number of nonsalary benefits that can at least partially offset the demands and financial concerns associated with a research career (see Box 5-1 above); second, stakeholders should promote these benefits at all stages of training (i.e., during undergraduate education, medical school, residency, and fellowship). As discussed in Chapter 2, early exposure to research opportunities is one of the few correlates to a sustained research career. The above recommendation is made to encourage recruitment strategies that emphasize the growing scientific evidence base underlying the practice of modern psychiatry (Charney et al., 2001; Goldman, 2002). As obvious as this recommendation may appear, the committee nevertheless believes such promotion strategies are logically and plainly tied to recruiting students and residents to research careers. The committee furthermore believes that the typical dialogue regarding
OCR for page 163
Research Training in Psychiatry Residency: Strategies for Reform research in general and clinical research in particular focuses so much on problems that it is no wonder that many medical students shy away from the endeavor. Implementation of this recommendation should not involve a concealment of the challenges to patient-oriented research (e.g., standardization of complex treatments and symptoms, struggles for funding), but instead should strive to present the positives along side the negatives. Finally, the demographics of the psychiatry workforce suggest that special measures are needed to ensure that talented women and IMGs are encouraged to pursue careers in patient-oriented research. Additionally, as is the case for all branches of medicine, greater involvement of underrepresented minorities is imperative if psychiatry is to offer the most responsive care to a diverse U.S. population. Accordingly, the committee makes the three recommendations presented below. Recommendation 5.3. Departments of psychiatry, supported by the National Institute of Mental Health and other psychiatric organizations, should provide leadership in recruiting and retaining more women for psychiatry research careers. Such efforts should include: Increasing part-time training and job sharing opportunities. Developing a critical mass of female role models and mentors. Working with institutions to improve child day care programs. Addressing institutional promotion and tenure issues, such as the tenure clock, that may be perceived as barriers to female trainees. Educating women about the time flexibility of research careers. Considerable progress has already been made in this area, so there are numerous programmatic examples from which to draw. A 1996 report published by the AAMC outlines some programs that provide a supportive infrastructure for female trainees and faculty members across medicine. Some of these programs offer mentoring support (e.g., Loma Linda University), provide child care (e.g., Beth Israel Hospital in Bos-
OCR for page 164
Research Training in Psychiatry Residency: Strategies for Reform ton), and support part-time opportunities and flexible scheduling (e.g. University of Virginia), while others track faculty development and review promotion and salary policies to ensure that the needs of women faculty and trainees are considered (e.g., University of California at Davis, Stanford University, Johns Hopkins University, Harvard University) (AAMC, 1996). Recommendation 5.4. Psychiatry training programs, academic medical centers, psychiatry organizations, and the federal government should work together to facilitate research training for international medical graduates who have the potential to make outstanding research contributions to psychiatry. Retention of the most productive of these international graduates in U.S. academic psychiatry programs should also be a joint effort. The committee is not aware of any institutions or psychiatry residency training programs that focus on accommodating the unique needs of IMGs. However, certain professional organizations have addressed some of the issues faced by these individuals. For example, the American Association of Directors of Psychiatric Residency Training has established a mentoring program for IMG residents (AADPRT, 2003). The APA, with funding from the Pfizer Company, publishes The International Psychiatrist Newsletter, which offers support and information to IMGs in psychiatry (APA, 1997b). Finally, a possible model for training foreign IMGs may come from the Epidemic Intelligence Service, which offers classroom- and field-based training to 60 to 80 health professionals, including 8 to 10 foreign trainees, each year (Centers for Disease Control and Prevention, 2002; White et al., 2001). Similar programs that increase international cooperation in the area of mental health research might enhance not only psychiatric research efforts in the United States, but also mental health efforts internationally. The committee believes recommendation 5.4 is especially important given the new visa restrictions faced by non-U.S. IMGs in the wake of the 2001 terrorist attacks. Recommendation 5.5. Psychiatry research training programs should increase the numbers of underrepresented minority researchers by employing the following strategies:
OCR for page 165
Research Training in Psychiatry Residency: Strategies for Reform Recruit minority faculty in multiple disciplines to serve as role models and mentors. Pursue funding from NIMH and other funding agencies that support minority trainees and faculty. Inform more minority psychiatrists about research training and other funding opportunities. Data indicate that racial and ethnic minorities apply in relatively low numbers for NIMH funds (National Advisory Mental Health Council’s Workgroup on Racial/Ethnic Diversity in Research Training and Health Disparities Research, 2001). One option for reversing this trend is to make minority medical students and psychiatrists more aware of the opportunities and needs that exist in patient-oriented research (Curry, 1997). Low application rates may, in no small part, be related to the fact that minority medical students often are from families with limited accumulated wealth, and thus they become highly dependent on loans or are otherwise constrained from career paths that delay full financial remuneration. Therefore, increased stipends and loan repayment programs (per recommendation 5.1) may be of particular relevance to minorities. Structured programs to match minorities to role models with similar experiences and to build a minority presence in departments have been engineered at some institutions (e.g., University of Pennsylvania), and these programs represent examples for others to emulate.
OCR for page 166
Research Training in Psychiatry Residency: Strategies for Reform This page intentionally left blank.
Representative terms from entire chapter: