In response to several potent external forces, many departments of psychiatry have started to grapple with the problem of downsizing their residency programs. First, with the decrease in numbers of both U.S. and total applicants entering psychiatry, in 1996 fewer than half of the available postgraduate year (PGY)-1 positions in psychiatry were filled in the National Resident Matching Program by graduates in the United States or Canada. In recent years, some very prestigious academic programs have for the first time not successfully filled in the match. Second, health care policymakers have also questioned whether the nation is training too many specialists. Third, in many places local economic conditions, shaped heavily by managed care, reduced state and county support, and anticipated cuts in Medicare funding for postgraduate education, have already resulted in substantially reduced departmental and hospital training budgets. In response to these forces, a number of residencies have already downsized their programs. Other specialties are also confronting similar issues (1—3). A recent survey of residency programs conducted by the Division of Graduate Medical Education of the American Medical Association revealed that many medical specialties are experiencing or anticipating that graduates will have increasing difficulty in finding positions over the next few years, and many specialties are anticipating cutting residency positions over the next few years (3). Of the 107 psychiatric residency programs responding to the survey, 15.9% anticipated employment problems for the 1994—1995 cohort of graduates, and 19.6% reported the likelihood of their reducing resident-physician positions during the next 3 years. For perspective, 10.9% of internal medicine, 16.3% of diagnostic radiology, 18.6% of ophthalmology, 19.3% of pathology, 36.4% of gastroenterology, 43.5% of cardiology, and 60.6% of anesthesiology residency programs reported the likelihood of reducing resident- physician positions during the next 3 years.
The authors have worked with two programs that have already undergone downsizing. In one case, downsizing was a natural reaction to the economic needs of the home institution. In the other case, downsizing was elected to avoid compromising standards of quality to fill all residency positions and also to prepare for a perceived decrease in marketplace demands for psychiatrists. To explore alternative approaches to reducing the number of residency slots, the authors designed a questionnaire that was administered via telephone interviews to five training directors whose programs were known to be in the midst of planning for downsizing (Group 1). The results of this survey were presented at the 1993 American Association of Directors of Psychiatric Residency Training (AADPRT) meeting in a workshop attended by training directors (Group 2), 12 of whom completed our questionnaire. Of these, four programs had already made cuts, two were planning cuts, and six were "thinking about it." Based on our discussions with Group 1 program directors, the discussions in the AADPRT workshop, and the completed questionnaires, these 17 programs constituted the sample of convenience from which our information has been gathered. (Convenience Sample of Programs Concerned with Downsizing, 1993. Group I: Cornell, Einstein; UCLA-Neuropsychiatric Institute, U. Texas Southwestern; and Yale. Group II: Case Western, Duke, Hillside, Jefferson Medical College, Wisconsin, SUNY Downstate, UCLA-San Fernando Valley, U. Colorado, U. Miami, U. Minnesota, U. Texas San Antonio, and one anonymous program. A copy of the questionnaire is available on request from JY.) Recent discussions at several national meetings have made it clear that many more programs are now seriously considering reducing the number of residency positions.
The purpose of this paper is not to advocate downsizing, but rather to provide information to academic psychiatrists considering such changes so that they may become aware of the various issues, processes, and pitfalls experienced by programs that have already taken this action. Each program must, of course, make its own decisions based on its perception of service needs, funding issues, and standards of trainee quality.
Since the extent of completeness of questionnaire responses was highly variable, with many blanks, particularly from those programs that were just starting to think about downsizing, the information we present must be considered as extremely impressionistic and preliminary. We present quantitative data only where we had enough responses to do so. The remaining statements are based on our overall impressions and synthesis of the informal and qualitative information we derived from others, and on the experiences in our own programs.
Reasons given for downsizing appear in T1 (combining responses from programs that were already downsizing with those still contemplating it). About half the programs felt that they were obliged to downsize and half were doing so, or thinking about it, electively. Forced downsizings were almost all due to funding cuts by hospitals, universities, local or state governments, and/or by reductions in patient care income. Elective reductions were motivated by a wide variety of factors, most commonly actual or anticipated decreases in recruitment and the decision to not accept less qualified applicants. Other reductions resulted from more complicated structural departmental reorganizations, which were variably related to economic pressures. Reorganizations included integrating two formerly independent training programs, reorganizing the relationship of an academic department to its affiliates, shifting resources from inpatient services to outpatient sites, and/or closing hospital beds as part of a long-term hospital reconstruction plan. Economic forces are clearly widespread and are reflected in a variety of idiosyncratic reorganizations in different departments.
EXTENT AND SCHEDULING OF THE DOWNSIZING
The Group 1 programs we surveyed were generally very large ones. None had entering class sizes of less than 12. Prior to downsizing, the average class size entering these programs was 19.2, and after downsizing was 13.8. Most cuts were in the range of 10%—15%, but two programs downsized by almost 40%. In general, the larger the program, the greater the decrement. Among the Group 2 programs, moderate-to-large programs predominated (8—15 class sizes), although a few smaller programs were also represented. Most Group 2 programs were planning to reduce by 1—3 residents per year, proportionately the same as the very large programs represented in Group 1 (range 8%—20%).
Almost all programs tried to make their decisions at least 1 year in advance of implementation. This gave services time to prepare and enabled the training director to inform all applicants during recruitment of the upcoming changes and their impact. One program reached its final decisions in the midst of the recruitment season and phoned all applicants to tell them of the changes.
Most programs were planning a "rolling cut," initiating the reduced class size first in PGY-1 and then continuing at that level through the next 3 or 4 years, allowing the reduced size to gradually work its way through all years of the program. A small number planned to make large cuts in the PGY-1 year but then recruit additional PGY-2 residents. Those programs that still controlled the money that previously funded the larger number of residency positions were considering reinvesting those funds in special research residency tracks, in various fellowship positions, or in hiring various kinds of physician extenders.
Final decisions about the timing and size of the changes were almost invariably made intradepartmentally, but in a few cases the deans and medical center directors strongly influenced the process. Usually departmental executive committees made the decisions, but in some instances the decisions were made unilaterally by the chairs. The impetus for considering a reduction in residency positions often came from the chair or important service chiefs when economic issues were paramount. For example, when one hospital realized that it faced a crisis of shrinking revenue and unfilled beds, the hospital director and chair together calculated the number of house staff to be cut, along with cuts in many other staff positions. Where issues of maintaining the quality of residents was a main consideration, the impetus came from the training directors. In the latter case, training directors needed to do extensive lobbying and education to overcome resistance at all levels. Most faculty were unaware of the drop in medical student recruitment and, even when aware of it, tended to think that their own program should somehow be immune to its effects. Typically, they believed that if the program would just tone up its recruitment efforts they would not have to face the downsizing issue. One training director was struggling to make headway with his faculty when his chairman attended one of the AADPRT's annual meetings and became convinced of the necessity for downsizing.
Once the decision to downsize was made, details were usually left for the training committee to work out. These included curriculum adjustments, changes in call coverage schedules, allocation of house staff reductions among affiliates and services, etc. Some programs formed elaborate task forces that developed alternative plans for the training committee to consider. At all programs, residents were invariably well represented in the process. The training committee's recommendations were always presented to the departmental executive committee (or some equivalent) for further discussion and final authorization. Sometimes this was pro forma, as when a chair fully authorized those delegated with the responsibility for "reengineering" the program, but more often this review was very active and stimulated considerable conflict. Several training directors reported encountering at least one particular service chief with exceptionally strong concerns about the impact of trainee reductions on his or her service. Although several training directors feared that such strong concerns might result in attempted "end runs" by faculty members to the chair or "grandstanding" to subvert the process, in fact this was not reported. However, lesser grades of grumbling and disquiet among affected faculty were common.
Decision making around the specific implementation of downsizing specifics generally involved only members of the department, but in several instances decisions by hospital administrators above the departmental level were critical elements in the planning. In two large East Coast programs, for example, regulations and funding related to the New York State Bell Commission's report on house staff work hours resulted in increased available funding for paid on-duty call in large county and Veterans Affairs hospitals, thus helping to solve problems related to call frequency in the face of downsized residency classes.
Within each department, the training director, associate training directors, and chair were most critically involved in making and finalizing these decisions, usually with much input from service chiefs and house staff. The training directors we surveyed uniformly described that painful decisions were always made after participatory discussions with all, or at least the large majority of, service chiefs and house staff. Nevertheless, several programs reported substantial degrees of unhappiness on the part of junior faculty who perceived the loss of house staff from their training units as directly adversely affecting their careers and personal workloads. In some instances, loss of house staff directly translated into junior faculty having to assume additional primary patient-care responsibility or primary care responsibilities for the first time. Senior faculty often experienced the loss of trainees from their services as effectively terminating their work as teachers, a role they highly valued. In some instances, downsizing provided good excuses and opportunities to weed out weaker rotations and faculty assignments. In one program, the cuts involved closing units that were not making money, and the house staff reductions corresponded to these closures. However, other services were also affected by these cuts. For example, cuts in the inpatient units at one university hospital reverberated through its affiliated VA and county hospitals by necessitating corresponding changes in inpatient assignments at these institutions to compensate.
With few exceptions, cuts were allocated among the institutions based on relative funding contributions. Often a desire to maintain comprehensive and diversified training experiences led to creative negotiations between the department and hospitals. In one case, despite a reduction of residency positions, an exchange of three resident stipends per year for one faculty salary prevented any reduction in service delivery.
Once reductions were allocated among the hospitals, decisions had to be made as to which services within each hospital would be cut. These decisions involved careful balancing acts by the training committee. In each instance, proposals of "the" optimal clinical curriculum were weighed against other institutional issues, such as service needs, personalities, and the opinions of those with the most authority and power.
The impact of the reductions on individual faculty, services, and hospitals was highly variable and often profound. Ample preparation for the changes was critical to maintaining a comprehensive and balanced training program. Nevertheless, major economic pressures, often unforeseen and sudden, tended to create environments in which precipitous decisions were sometimes made, sometimes by fiat—by hospital directors or chairs or key service chiefs. However, even when long preparation and discussion with much input from all involved were possible, the negative impact on morale was frequently significant. Often, faculty and staff were narcissistically wounded even by the mere suggestion that the number of trainees on a service might have to be reduced—their pride, and sometimes their sense of entitlement, being attacked by no longer being assigned residents, or by having their house staff allocations cut relative to other services. Some faculty were said to have taken these reductions very personally, feeling that they were being intentionally slighted or mistreated by the training program. In several programs, even after plans had been precirculated for months some faculty reacted with surprise, outrage, and bitterness when they saw the final house staff schedules, as if they knew nothing of what was about to happen.
Several strategies for ameliorating the impact of downsizing were described. These included extensive discussion and consultation; deciding to close some services rather than thinly staff them all; assigning residents for only some months during the year to some previously year-round teaching services; transforming some house staff—free clinical services into teaching sites for medical students only; hiring physician extenders to assist with the service load; and disaffiliating with hospitals now regarded more as liabilities than assets.
Dealing with call issues is a crucial part of downsizing, since fewer residents are available to cover the same number of night, weekend, and holiday calls. Programs evolved different strategies. Some increased the frequency of call, but, at least in the larger programs, rarely did this exceed every eighth or ninth night. Most programs clearly intended to hold the line on night-call frequency to about once per week. Others restructured call entirely, creating night-float systems, setting up call teams headed by more senior residents, allowing home call after certain hours, or finding funds to establish paid call systems to handle the added responsibilities. One program developed a system whereby the emergency room was covered by paid PGY-4 residents rather than unpaid PGY-2 residents. Although restructuring tended to be more popular among the residents, some conflicts were described involving residents who started under one call system who finished under another.
Specific clinical services found a wide variety of solutions. In one instance, a consultation-liaison service that previously relied exclusively on residents shifted its coverage entirely to staff psychiatrists. In another, the nature of service delivery was restructured entirely, transforming from a house staff—based consultation model to one in which faculty liaisons became more important. In all programs, at least some of the inpatient training units found that they now had fewer or even no house staff. Outpatient services often hired staff psychiatrists to maintain necessary levels of service. Other mechanisms included adding medical students, physicians' assistants, research assistants, social workers, or nurse clinicians to help staff units in place of residents.
Once the decision to downsize is made, the training director should determine an overall strategy for planning and implementation. Widespread initial information sharing, discussion, and requests for input are critical to ensure a broad sense of participation and buy-in with the plan from all affected constituents. A Deming-style, Total Quality Management approach (4) will help to ensure that those faculty and house staff most likely to foresee serious glitches, those closest to the problems, are involved in the planning at the very beginning, when they can anticipate problems and forewarn those whose proposals are naive. This approach involves "stakeholders" at all levels from the very beginning of planning to provide early warnings and to facilitate buy-ins, including all levels of house staff, all levels of faculty, and all levels of administration. Larger programs may require more elaborate organizational strategies such as the ones described next, whereas smaller programs may collapse some of the task force and committee structures into fewer elements. Based on our personal experiences and discussions with several training directors who have been through the process, suggested guidelines for decision making and implementation for downsizing include the following, or some close variations (an abridged list appears in T2).
1. Inform all those likely to be affected or to care, including staff and house staff at all affiliates and services, about the decision, the extent of the downsizing, and the expected timetable for implementation.
2. Convene one or more "town councils" to facilitate brainstorming among the faculty and house staff. Resist making firm decisions at the town council level, because strong social pressures generated by aggressively outspoken small groups or idiosyncratic personalities may unduly promote certain factional interests.
3. Decide what work groups are to be appointed. Carefully select chairs and give them clear charges and timetables for committee work and reports. To whatever extent possible, the training director and associate training directors should chair these work groups. Committees should be composed of faculty, staff, and house staff. House staff on work groups should include those likely to be most knowledgeable about the system as well as those most likely to be affected. Thus, house staff representatives should include both senior residents who have been in the program for 3—4 years as well as those junior residents who are going to be in the program after the changes are completed. It is essential to include faculty whose absence one would undoubtedly come to regret later (for reasons of politics, badmouthing, sabotaging, perspective, etc.). Initial town council and work group meetings should be scheduled to permit sufficiently long and detailed discussions (e.g., day-long or half-day retreats or evenings).
4. Clearly describe work groups as having only an advisory role to the training committee and the training committee as being advisory to the chair/executive committee. Work groups should not be led to expect automatic rubber stamping of their decisions.
5. Encourage and facilitate frequent ongoing crosstalk among work groups, training committee members, executive committee members, and chairs to ensure that no one is blind-sided. Constantly float and test "trial balloons" so that hidden difficulties that might render otherwise inviting proposals nonviable are discovered early and wasted time and effort are kept to a minimum.
6. After the larger constraints are established (e.g., how many house staff per year and how many house staff at each training site for which rotations), establish work groups to develop the details for each specific year. Membership should include major rotation clinical chiefs, institutional chiefs, administrators, and house staff.
These groups should generate one or more alternative plans for service rotations, to include detailed monthly and daily clinical and didactic schedules. The most feasible options should be written up, with their pros and cons, and circulated among involved faculty, affected service chiefs, etc., for specific comment and feedback. These opinions (often best delivered in person) should then be brought back to the work groups so that the proposals can be refined and, in the best of worlds, achieve group consensus.
7. Appoint core curriculum seminar work groups for each affected PGY class. Membership should include representative major current course coordinators and house staff. As with the clinical service curriculum, after alternative options are generated, written drafts/options should be circulated among a larger faculty and house staff group. It is always wise to assume "zero-based budgeting" with regard to time set aside for seminars; that is, future time is not allocated to previously scheduled seminars simply on the basis of history or tradition. The amount of time each seminar is to be allocated each year should be based on a fresh review of the salience, quality, and value of the seminar and instructors in the overall education of the residents.
8. Authorize a work group to examine and reassess call coverage. The group may be chaired by the chief resident. Representatives should include residents elected by their peers. Depending on local situations, emergency room service chiefs may be included as working members or consultants.
9. Funnel all work group reports, options, and recommendations to the training committee and then to the chair and department executive committee for final actions and approvals.
As an alternative to this complex planning process, one very bright and/or autocratic person, usually the training director or chair, may simply decide what the new program will be, dictate that program, and be prepared to take the credit or blame and the inevitable heat.
Our survey of 17 downsizing programs revealed as many models as there were programs. Although each had to deal with some unique issues, many experienced common problems. These problems included faculty disbelief that downsizing was necessary, anxiety by virtually all faculty and house staff about any impending changes, morale problems once final decisions were made, and, usually, a sense of general satisfaction (less frequently dissatisfaction) once the changes were implemented.
Some programs sustained losses of affiliates, faculty, and/or funding along the way. Others were able to maintain themselves without sustained hurt feelings or significant losses. In many ways, the processes were almost as important as the outcomes. With sufficient lead time and extensive consultation, programs can creatively use the process of downsizing to develop new, adjusted training experiences that more realistically reflect the changing needs and realities of our institutions and ever-reforming professional activities.