To the Editor: There is probably not much disagreement that the increasing size of the geriatric population in the United States, with a concomitant expanding prevalence of psychiatric problems, will place an ever-greater burden on the health care system (+1). No health care profession, including geriatric psychiatry, is projected to have the minimum number of trained personnel to accommodate the needs of the future geriatric population (+2). Meeting the prospective needs of geriatrics and, in particular, geropsychiatry rests in part on the present and future foundation of teaching and research. Medina-Walpole et al. (+3) reported that geriatric fellows completing 2 or more years of training had greater commitment to research, teaching, multidisciplinary teams, and professional societies. However, studies suggest that the number of geriatric fellows who choose education and research career paths continues to decline with successive years of training. In addition, the National Academy of Science considers the development of new clinical investigators to be a critical National Institute of Health (NIH) problem (+4). These signs of inadequate training programs, not only for qualified geriatric clinicians, but also for clinician-educators and clinician-researchers who will train future geriatric specialists, suggest that creative solutions may be needed.

The number of geropsychiatry fellows and the percent of geropsychiatry fellowships with options for more than 1 year of training have fluctuated over the years. In 1981—1982, 12 geropsychiatry fellows graduated from fellowships in which they had an average of 1½ years of training (+2). In the late 1980s, one study revealed that of 18 geropsychiatry fellowships, 11 (61%) had second-year options (+5), while no third-year options were offered in 29 geropsychiatry fellowship programs surveyed in another (+2, +6). This limited pool of geriatric fellows, especially among those choosing 2 or more years of specialized training, was most noted among United States medical graduates (USMGs) in the early 1990s (+2). One vigorously debated innovative solution to this dilemma was the introduction of the 1-year fellowships in 1995. One-year fellowships gradually increased the number of residents entering geriatric medicine fellowships from 144 in 1996 to a peak of 269 in 1999 (+7). However, the total second-year and beyond positions that are filled has decreased from 100 in 1997—1998 to 79 in 2001—2002 (+2). Additionally, although the total number of first-year geriatric medicine positions has increased, the fill rate has fluctuated from a high of 91.2% in 1998—1999 to 69.4%, the lowest level since the start of the 1-year fellowship options (+7). There has been a corresponding decrease in both the number and percent of second-year and beyond geriatric medicine fellows from 100 (32.8%) in 1997—1998 to 79 (23.4%) in 2001—2002 (+7).

In geropsychiatry, the number of first-year fellowships has steadily increased from 82 in 1996—1997 to 137 in 2002—2003. The total number of geropsychiatry fellows has not kept pace with the number of fellowship slots, increasing only from 82 in 1996—1997 to 94 in 2001—2002. In contrast to geriatric medicine, the number of fellows in geropsychiatry beyond the first year has marginally increased from five in 1996—1997 to 13 in 2001—2002, although the numbers are much smaller than those of geriatric medicine (+7).

The indices of how geropsychiatry is preparing for the future are equivocal. Since the end of the inclusion of practice pathway geropsychiatry candidates in 1996 and the start of 1-year geropsychiatry fellows sitting for the geriatric certificates of added qualifications (CAQ), the number of issued CAQs has increased from 65 in 1998 to 87 in 2002, which reflects the increase in 1-year fellowships (+2). To USMGs, however, geropsychiatry fellowships seem to be less attractive than geriatric medicine fellowships. While the percentage of USMGs entering geriatric medicine fellowships has increased since the introduction of the 1-year fellowship option (40% in 1996 to 45% in 2002), a reverse trend has occurred in geropsychiatry (47% in 1996 to 43% in 2002). With a median of two fellows per geropsychiatry program, 62% of the fellowships had one or no USMGs (+2). In a recent survey of geropsychiatry programs in the United States, 36 of 46 responders (78%) offered 1-year fellowships with the remaining 10 (22%) offering additional training. Two of these 10 programs offered a third-year fellowship, and one offered a fourth-year fellowship (+7). Compared to 61% of geropsychiatry fellowships offering second-year options in the 1980s (+5), this 40% decrease of training opportunities beyond the first year in 2001—2002 may represent a dubious shift in priorities.

There are multiple barriers to recruiting geropsychiatry fellows and to encouraging them to pursue the prolonged training needed to become geropsychiatry educators and researchers (+4). There has been a decrease in both the number of psychiatry residents since 1993 and in the primary care initiative of the late 1990s (+2, +8). Additionally, and never to be underestimated, is the disincentive to pursue extended training for a nonprocedure oriented geropsychiatry career, which results in mounting debts for fellows. The average USMG debt is reported to be $103,855, with greater than 20% of USMGs owing more than $150,000 (+2). Especially for those, like myself, who have families, the barrier of extending training through residency, followed by a prolonged fellowship, can provoke continual anxiety and intrafamilial stresses.

Many of the barriers that influence how the geropsychiatry training programs evolve are based on federal policy and funding (+4). For example, the Veterans Health Administration (VHA) that funded 60% of the first-year geriatric medicine and psychiatry positions for academic year (AY) 2000—2001 has decreased its spending for second-year fellowship academic and research positions and reallocated funding to support the increased number of 1-year positions (+2).

As a newcomer to the field of geropsychiatry, I am still struggling to understand and navigate the complexities of the possible clinician-researcher option. For many fellows, this includes the conflict of substantial debt and family responsibilities versus the desire to pursue research and academic careers. While we do not have a crystal ball regarding the availability of finances for extended geropsychiatry fellowship programs, a combination of existing programs and future creative solutions may help expand the number of future academics.

Presently, the National Institute of Health (NIH) has a generous loan repayment program for up to $35,000 per year in return for doing not less than 20 hours of qualified clinical research for 2 consecutive years (+9). Additionally, the VHA has instituted a "special fellowships program" in an attempt to compensate for the reduction in their second-year fellowships. In this option, there are two second-year slots at each of seven geriatric research education clinical center (GRECC) sites for fellows that have completed one year in an American Council of Graduate Medical Education (ACGME) program. During this second year, activities in research, education, and career development will constitute 75% of the program (+2).

Perhaps increasing fellowship stipends and having a call-in center for managing debt, fellowship, and grant financing problems, possibly from a GRECC, would be useful for fellows who are not at GRECC centers. Moreover, while improving fellowship training in grant writing, time management skills, and interdisciplinary collaboration and consultation might improve the transition from junior to independent investigator (+10), these skills may also improve the transition from residency to educational and research fellowships.

Some suggested options for present psychiatry residency and geropsychiatry fellowship programs may be useful. Perhaps a first step would be the development of a national grant/financial clearing house and an introductory manual on how to enter career paths in research and academics and succeed in doing so. This may serve not only residents considering this career choice, but also undergraduate, graduate, medical school, and residency faculty members who would like to include teaching seminars on this topic but do not have global understanding of the myriad details. Additionally, geropsychiatry fellowships could begin in the fourth residency year and include 2 years of specialized training as is presently done for pediatric psychiatry (+11). For those fellows who wish to remain at their first-year site, an extension for an optional second or third year could be considered. This might include spending variable lengths of time with specific mentors and researchers at designated GRECCs and other qualified academic sites. The fellow would become a junior faculty member at the home site, engaging in clinical teaching and research activities. For this option, when the 1-year fellowship program would agree to support extension of the fellowship to a second or third year, there could be a fund-matching program involving private, state, and federal funding.

Bragg and Warshaw note that despite a variety of sources of financial support for geriatric medicine and geropsychiatry, interest beyond the 1-year fellowship in academic and research careers "remains weak" (+2). It is possible that the future number of VHA and NIH fellows may not be sufficient to refuel the pipeline of future independent investigators, senior leaders, and mentors needed "to nourish the field" of geriatrics (+2, +4). Additional creative efforts and funding to increase the number of clinician-educators and clinician-researchers would clearly be a valuable investment in the future.