VA Health Care: Changes in Medical Residency Slots Reflect Shift to
Primary Care (Letter Report, 04/12/2000, GAO/HEHS-00-62).

Pursuant to a congressional request, GAO reviewed the Department of
Veterans Affairs (VA) changes in medical residency slots, focusing on:
(1) VA's goals in realigning its residency program and the goals
accomplished so far; (2) the reasons for changes in the number of
graduate medical education residency slots; and (3) the views of VA
facility and medical school officials on the effect of the changes on
resident training and on the potential to train residents at VA
community-based outpatient clinics.

GAO noted that: (1) the changing health care environment has resulted in
less demand for specialty physicians and more demand for primary care
physicians; (2) this shift influenced the changes in residency slots at
VA; (3) over the last three academic years, VA realigned its graduate
medical education program and achieved its goal to train 48 percent of
its residents in primary care; (4) VA's strategy eliminated 251
residency slots in specialty care and converted 714 specialty residency
slots to primary care slots; (5) these changes reduced the number of
residency slots from 8,910 to 8,659; (6) the major reasons for the
changes in residency slots were: (a) VA's decision to increase primary
care residency slots and decrease specialty slots; and (b) medical
school decisions to restructure their programs to meet changing demands
for physicians or accreditation requirements; (7) VA initiated the
majority of changes in the residency slots at the six facilities GAO
visited, but these changes were consistent with the medical schools' own
initiatives to meet changing demands; (8) VA and medical school
officials characterized the changes as generally mutually beneficial
because they were consistent with current health care practices
nationally; (9) changes in residency slots have not been disruptive to
training, according to VA and medical school officials at the six
facilities GAO visited; (10) when VA reduced the number of residency
slots, for the most part those residency slots reappeared at other
hospitals affiliated with the medical schools; (11) in addition, VA and
medical school officials said that some training opportunities exist at
VA's community-based outpatient clinics; and (12) however, VA is not
pursuing establishment of such slots because: (a) sufficient
opportunities exist for primary care training in outpatient clinics
located at VA hospitals; (b) remote community-based clinics present a
commuting problem for the residents; and (c) the physicians who would be
required to supervise and train residents at remote clinics might not be
able to obtain faculty status at the medical schools.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  HEHS-00-62
     TITLE:  VA Health Care: Changes in Medical Residency Slots Reflect
	     Shift to Primary Care
      DATE:  04/12/2000
   SUBJECT:  Veterans hospitals
	     Medical education
	     Physicians
	     Medical schools
	     Training utilization
	     Community health services
IDENTIFIER:  VA Veterans Integrated Service Network

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************

GAO/HEHS-00-62

Appendix I: Objectives, Scope, and Methodology

14

Appendix II: Number of VA Medical Resident Primary Care
and Specialty Slots in Academic Year 1999-2000

16

Appendix III: Percentage of Residency Slots in Primary Care
Training, by Network

18

Appendix IV: Selected VA Facility and Affiliated Medical School
Residency Slots Before and After VA Residency Realignment

20

Appendix V: Summary of Community-Based Outpatient Clinic Data
for the Six VA Facilities Visited

21

Appendix VI: Comments From the Department of Veterans Affairs

22

Appendix VII: GAO Contact and Staff Acknowledgments

24

Table 1: Changes in the Number of Residency Slots Supported by VA 7

Table 2: Changes in Residency Slots at the Six Facilities Visited 8

CBOC community-based outpatient clinic

UCLA University of California at Los Angeles

VA Department of Veterans Affairs

VISN Veterans Integrated Service Network

Health, Education, and
Human Services Division

B-283989

April 12, 2000

The Honorable John D. Rockefeller IV
Ranking Minority Member
Committee on Veterans' Affairs
United States Senate

Dear Senator Rockefeller:

Many changes under way in the Department of Veterans Affairs (VA) have the
potential to affect the relationship between VA medical facilities and their
affiliated medical schools. For example, VA has been reorganizing its health
care delivery system by moving more care from inpatient to outpatient
settings. It is also integrating facilities and consolidating programs and
services at health care facilities to improve efficiency, which could
potentially affect its graduate medical education program. Also, over the
last 3 years, VA has been realigning its graduate medical education program,
which involves reducing specialty residency slots while increasing the
number of primary care slots--such as those in family practice and internal
medicine.

You expressed concern that limited information is available explaining the
rationale for changes in the numbers and types of medical residency slots
and describing the effect these changes have had on the resident training
environment.1 In this context, you asked us to address the following
questions: (1) What were VA's goals in realigning its residency program, and
did VA accomplish its goals? (2) Who initiated the changes in the number of
graduate medical education residency slots, and why? (3) What are the views
of VA facility and medical school officials on the effect of the changes on
resident training and on the potential to train residents at VA
community-based outpatient clinics? To assess residency realignment, we met
with officials from VA's Office of Academic Affiliations, reviewed VA's
residency realignment goals, and compared the goals with data from the last
3 academic years. For the other questions, we visited six VA facilities that
experienced among the largest net changes in their number of residency
slots. At each facility, we met with senior management officials, clinical
service chiefs, and representatives from the affiliated medical school(s) to
determine their rationale for changes in the number of residency slots and
to obtain views about the effect the changes have had on the training
environment and on the opportunities for training at VA community clinics.
We did our work from May 1999 through January 2000 in accordance with
generally accepted government auditing standards. Additional information on
our scope and methodology is included as appendix I.

The changing health care environment has resulted in less demand for
specialty physicians and more demand for primary care physicians. This
shift, in turn, influenced the changes in residency slots at VA. Over the
last three academic years,2 VA realigned its graduate medical education
program and achieved its goal to train 48 percent of its residents in
primary care. VA's strategy eliminated 251 residency slots in specialty care
and converted 714 specialty residency slots to primary care slots. These
changes reduced the number of residency slots from 8,910 to 8,659.

The major reasons for the changes in residency slots were (1) VA's decision
to increase primary care residency slots and decrease specialty slots and
(2) medical school decisions to restructure their programs to meet changing
demands for physicians or accreditation requirements. VA initiated the
majority of changes in the residency slots at the six facilities we visited,
but these changes were consistent with the medical schools' own initiatives
to meet changing demands. VA and medical school officials characterized the
changes as generally mutually beneficial because they were consistent with
current health care practices nationally.

Changes in residency slots have not been disruptive to training, according
to VA and medical school officials at the six facilities we visited. When VA
reduced the number of residency slots, for the most part those residency
slots reappeared at other hospitals affiliated with the medical schools. In
addition, VA and medical school officials said that some training
opportunities exist at VA's community-based outpatient clinics. However, VA
is not pursuing establishment of such slots because (1) sufficient
opportunities exist for primary care training in outpatient clinics located
at VA hospitals; (2) remote community-based clinics present a commuting
problem for the residents; and (3) the physicians who would be required to
supervise and train residents at the remote clinics might not be able to
obtain faculty status at the medical schools.

The U.S. health care system is undergoing profound changes. Rapid
technological advances, changing economic and demographic factors, and an
emphasis on cost containment have resulted in a dramatic decrease in
inpatient care and a corresponding increase in ambulatory services.
Recognizing the need to adapt to changes occurring in the health care
environment, VA is reorganizing its health care delivery system. VA replaced
its four health care regions with 22 Veterans Integrated Service Networks
(VISN). VA's reorganization has increased veterans' access to ambulatory
care, emphasized primary care, and decentralized decision-making. To attract
and retain veterans who do not live close to VA medical facilities, VA has
also established community-based outpatient clinics that emphasize primary
care.

Each of VA's 22 networks oversees between 4 and 11 medical facilities. Some
networks have integrated facilities, a critical piece of VA's overall
strategy to enhance the efficiency and effectiveness of health care delivery
to veterans. An integration involves restructuring clinical and
administrative services within two or more medical facilities into one
health care delivery system. Integrations are meant to allow VA to provide
the same or higher quality services to veterans at a reduced cost. As of
January 2000, networks had initiated 26 integrations involving 54
facilities.

Transforming VA's health care delivery system from an inpatient to an
outpatient focus, increasing reliance on primary care, and consolidating
services in two or more facilities could have direct implications for VA's
education mission--one of its four core missions.3 For example, VA's
initiatives will increasingly involve consolidating programs or possibly
closing facilities to eliminate duplication among nearby VA facilities.
Because VA is a major source of medical training opportunities, medical
schools clearly have a vested interest in VA's transformation.

Since 1946, VA facilities have had affiliation agreements with medical
schools. VA appoints medical school faculty as VA staff physicians so that
they can teach and supervise medical residents and students who are training
at VA facilities. Of VA's 172 medical centers, 130 are affiliated with 107
medical schools. In fiscal year 1999, VA spent about $356 million for
residents' salaries. VA is supporting 8,659 residency slots during the
1999-2000 academic year. Annually, 33,000 residents rotate through these
slots--almost one-third of all residents trained in the United States.

These affiliations are beneficial to both VA and medical schools. Residents
provide a portion of the care to veterans at VA facilities. According to VA,
when residency slots cannot be filled, the cost of replacement by
nonphysician practitioners or fully trained physicians is greater. Residents
are taught and supervised by VA physicians who hold academic faculty
appointments, provide direct patient care, and conduct research. This
combination of patient care, education, and research makes VA positions
attractive for the recruitment and retention of highly qualified physicians.
An additional advantage is that some VA-trained residents become career
employees.

For academic year 1999-2000, VA met its goal of increasing to 48 percent the
portion of its residents training in primary care. In 1997, VA began
reducing the number of specialty residents trained by eliminating some slots
and shifting some slots to primary care. This goal was established on the
basis of the national need to reduce the total number of resident training
slots and the number of residents trained in specialty care.

Recognizing the changing nature of health care, the Under Secretary for
Health appointed a committee to develop a strategy that would ensure that
VA's graduate medical education program was prepared to meet present and
future health care needs.4 He asked the committee to include any issues that
might improve VA's graduate medical education program, including the number
and types of training opportunities.

In May 1996, the Residency Realignment Review Committee recommended that VA
increase its proportion of primary care slots to 48 percent by eliminating
250 residency slots in disciplines other than primary care and shifting 750
slots from specialty care to primary care. The Committee defined primary
care slots as internal medicine, family practice, geriatric medicine,
preventive medicine, obstetrics, and gynecology. Specialty care was defined
as those slots that provide support to VA primary care and includes
psychiatry, general and other types of surgery, cardiology and other
subspecialties of internal medicine, pathology, and radiology (see app. II
for a list of primary care and other specialties). The Committee suggested
that its recommendations be implemented over a 3-year period ending with
academic year 1999-2000. VA accepted the Committee's recommendations.

Table 1 shows the changes VA made to attain its goal of having 48 percent of
its residency slots in primary care. This reduced the number of residency
slots to 8,659 systemwide. See appendix III for a list of changes by
network.

Table 1: Changes in the Number of Residency Slots Supported by VA

                    Baseline                                   Change from
                    academic                                   baseline
                    year 1995-96   Academic year               year

                                   1997-98  1998-99 1999-2000
 Total VA-supported
 residency slots    8,910          8,848    8,721   8,659      -251
 Total VA-supported
 residency slots in 3,442          3,655    3,813   4,156      +714
 primary care
 Total VA-supported
 residency slots in 5,468          5,193    4,908   4,503      -965
 specialty care
 Percentage of
 VA-supported
 residents in       39             41       44      48         +9
 primary care

Note: Fifty-eight VA facilities designated 281 medical subspecialty and
psychiatry residency slots as primary care because residents spent a portion
of their rotations providing primary care. This accounted for 3 percent of
the total 8,659 slots.

VA Residency Slots

The changing health care environment, such as the increased emphasis placed
on primary care, was the main reason for the shifts and reductions in the
207 residency slots at the six facilities we visited. VA's residency
realignment effort increased residency slots in primary care and reduced the
number of specialty slots, and in the meantime, the affiliated medical
schools decided to restructure their programs to meet changing demands for
physicians or accreditation requirements.

To understand the reasons for the changes in the number of residency slots,
we visited six VA facilities that had experienced among the largest net
changes in the number of slots--Albany, New York; Albuquerque, New Mexico;
Gainesville, Florida; Muskogee, Oklahoma; Philadelphia, Pennsylvania; and
West Los Angeles, California. During our facility visits, we obtained
information on the specific reasons for each change and whether VA or the
medical school initiated the change.

As shown in table 2, the six VA facilities initiated more changes that added
residency slots than the medical schools did because of VA's desire to
increase the number of primary care residents trained. On the other hand,
the medical schools initiated changes that reduced residency slots more
often than VA did.

Table 2: Changes in Residency Slots at the Six Facilities Visited

                                  Residency     Total
                      Residency   slots         residency     Percentage of
                      slots added               slots         total change
                                  decreased
                                                affected
 VA-initiated change  58          62            120           58%
 Medical
 school-initiated     14          73            87            42%
 change
 Total changes for
 six facilities       72          135           207           100%

Of the 120 changes in residency slots that VA initiated, 24 increased the
number of primary care resident slots and 49 decreased the number of
specialty residency slots. Various reasons accounted for the other 47 slots
affected. For example, the VA facility in Albuquerque obtained 12 residency
slots in general internal medicine from another VA facility within the same
network. This change was made because an affiliated medical school, the
Texas Tech University Medical School, decided to reduce the size of its
general internal medicine program at the VA facility in Amarillo, Texas. VA
changed the remaining 35 residency slots for various reasons, including
adding new residency programs, reducing existing residency programs, and the
changing patient workload demands.

When medical schools initiated changes in residency slots, the vast majority
of the changes resulted in decreases in the number of residency slots.
Seventy-three of the 87 changes in residency slots shown in table 2 resulted
in such reductions. Medical schools decided to reduce residency slots mainly
because of restructuring decisions or accreditation requirements. While
medical schools did not usually initiate large changes in the number of
residency slots, this was the case at two of the locations that we visited.

� In the first case, one of the two medical schools affiliated with the VA
facility in Philadelphia, MCP Hahnemann University School of Medicine,
discontinued its affiliation with VA after a merger with another school and
subsequent assessment of its patient workload and the number of residents it
could support. This resulted in a decrease of 24 residency slots at VA in
four different resident training programs.

� In the second case, the University of Oklahoma College of Medicine-Tulsa,
which is affiliated with the VA facility in Muskogee, was experiencing
accreditation problems with some of its residency programs. Because the
medical school was affiliated with four hospitals, it decided to consolidate
its residency programs so that it could deal with fewer hospitals. Since the
Muskogee facility was located 45 miles away from the school, the school
pulled back 17 residency slots--13 slots in general internal medicine and
family practice and 4 general surgery slots.

The remaining 46 changes in residency slots initiated by the medical schools
resulted for a variety of reasons. For example, the medical schools made
changes because of difficulty in utilizing and filling residency slots in
certain specialties or decisions to downsize certain specialties. Two
medical schools, for example, reduced 10 residency slots in anesthesiology
because of difficulty recruiting residents. Other reasons included changes
in patient workload and adding new resident training programs. For example,
a medical school relocated a training slot in rheumatology to another
affiliate because the workload at VA was not adequate to support the number
of residents assigned. Another medical school increased the number of slots
in cardiology to build its program in that specialty.

The shifting of residency slots from specialty to primary care and the
elimination of specialty slots have not been disruptive to training,
according to VA and medical school officials at the locations we visited.
When VA initiated changes in residency slots, it was generally done in
collaboration with the medical schools because the medical schools have been
under pressures similar to those VA has experienced to meet changing health
care demands.

Most VA and medical school officials we interviewed said that while the
elimination of some slots and the shifting of others had sometimes been
difficult to implement, the realignment did not adversely affect training of
the remaining residents or the accreditation of residency programs. In most
cases, when VA reduced slots, the medical schools relocated them to their
other affiliates. Therefore, the medical schools' residency programs
generally maintained their size (see app. IV for a comparison of the number
of medical schools' residency slots before and after VA's residency
realignment initiative).

Medical schools were under similar pressures to change residency programs
while VA was realigning residency slots. For example, some medical schools
reduced specialty residency programs because of state legislation to
increase the number of residents trained in primary care. According to the
National Conference of State Legislatures, 11 states passed legislation in
the early 1990s to have medical schools increase their training of primary
care physicians. Seven of those state laws required the schools to direct at
least 50 percent of their graduates into primary care.5 The University of
California at Los Angeles (UCLA) School of Medicine, which is affiliated
with the VA facility in Los Angeles, was affected by such legislation. In
1992, the California legislature passed a bill that required medical schools
to increase their primary care residency slots to 50 percent of the total
number or lose up to $8 million in state funding. Although this legislation
was subsequently vetoed, in a 1994 memorandum of understanding with the
state of California, UCLA agreed to increase its percentage of residents in
primary care to 60 percent over a 10-year period. Thus, when the VA facility
in Los Angeles initiated increases in primary care residency slots and
decreases in specialty care slots, it actually helped UCLA meet its goal.
Similar legislation affected schools elsewhere, such as the Albany Medical
College, which is affiliated with the Albany VA facility in New York.

According to medical school and VA facility officials, most residents have
remained satisfied with their education experience. Medical school officials
told us that on the basis of surveys and interviews regarding VA rotations,
there has been no change in residents' satisfaction levels. Residents like
their VA rotations and, in some cases, consider them their best experiences.
In addition, medical schools value the opportunities VA rotations afford
their residents. VA provides education opportunities that residents are not
always able to get at other affiliates, such as working with a larger
population of patients who are older and have multiple and advanced
diseases. Furthermore, VA officials told us that some residents prefer VA
rotations because they (1) have more independence and greater
responsibility, (2) believe the teaching atmosphere is better at VA
facilities because VA physicians spend more time with the residents, and (3)
spend more time with patients.

While most VA and medical school officials did not believe that the training
environment had been impaired by VA's residency realignment, some cited
examples of how the environment had been affected. For example, the decrease
in subspecialty residency programs has resulted in some loss of interaction
between specialists and primary care residents. In addition, VA and medical
school officials told us that reductions in slots will mean that some
residents will miss the opportunity to treat patients who are older and have
multiple and advanced diseases.

VA and medical school officials agreed that some training opportunities are
available at VA's community-based outpatient clinics (CBOC). CBOCs differ
from traditional freestanding VA outpatient facilities in that they provide
primary care to veterans and frequently use private providers who contract
with VA. The type of care veterans receive at these clinics is comparable to
that available at a private physician's general practice office. The CBOCs
associated with VA facilities we visited are located from 5 to 240 miles
away.

Residents were not being trained at any of the 32 CBOCs associated with the
six facilities we visited. (See app. V for specific information regarding
the CBOCs at these facilities.) Moreover, VA has no plans to train residents
at these locations. VA officials told us they are not using CBOCs for
resident training because (1) sufficient opportunities for primary care
training exist in outpatient clinics located at VA hospitals; (2) CBOCs
located far from the hospital present a commuting problem for the residents;
(3) physicians who would be required to supervise and train residents at the
remote clinics might not be able to obtain faculty status at the medical
schools; and (4) the variety of education experiences residents would
receive at CBOCs is limited, so residents would need to train at additional
locations.

In commenting on a draft of this report, VA concurred with its content. We
made minor revisions on the basis of VA technical comments, as appropriate.
VA's comments are included in appendix VI.

We are sending copies of this report to the Honorable Togo West, Secretary
of Veterans Affairs, and other congressional committees with an interest in
this issue. We will also make copies available to others on request.

If you have any questions about this report, please call me at (202)
512-7101 or Walter Gembacz at (202) 512-6982. Other major contributors to
this report are listed in appendix VII.

Sincerely yours,

Stephen P. Backhus
Director, Veterans' Affairs and
Military Health Care Issues

Objectives, Scope, and Methodology

The Ranking Minority Member of the Senate Committee on Veterans' Affairs
expressed concern that there was limited information available explaining
the rationale for changes in the numbers and types of residency slots or
describing the effect these changes have had on the resident training
environment. In this context, he asked us to address the following
questions: (1) What were VA's goals in realigning its residency program, and
did VA accomplish its goals? (2) Who initiated the changes in the number of
graduate medical education residency slots, and why? (3) What are the views
of VA facility and medical school officials on the effect of the changes on
resident training and on the potential to train residents at VA
community-based outpatient clinics?

To determine whether VA's realignment of its residency program accomplished
its goals, we met with officials from the Veterans Health Administration's
Office of Academic Affiliations, reviewed VA's residency realignment goals,
and compared the goals with the latest available data on residency slots. We
examined numbers of residency positions for academic years 1995-96 through
1999-00, the period during which VA implemented its residency realignment.

To address who initiated the changes and why, we visited six VA facilities
that experienced among the largest net changes in the number of residency
slots. In selecting the facilities, we considered the number of residency
slots supported by VA. We selected facilities from six different Veterans
Integrated Service Networks (VISN) and, among them, selected two integrated
facilities. The six facilities selected were located in Albany, New York
(VISN 2); Albuquerque, New Mexico (VISN 18); Gainesville, Florida (VISN 8);6
Muskogee, Oklahoma (VISN 16); Philadelphia, Pennsylvania (VISN 4); and West
Los Angeles, California (VISN 22).7

At each facility, we interviewed senior management officials and clinical
service chiefs at VA medical centers and representatives from the affiliated
medical schools to determine the reasons for increases or decreases in the
number of residents trained in a specific specialty. For facilities where
the residency positions were decreased, we determined whether the medical
schools relocated the slots to affiliates or eliminated the slots.

To obtain views about the effect of changes in residency slots on the
graduate medical education training, including opportunities for training
residents at VA community-based outpatient clinics, we interviewed senior
management officials and clinical service chiefs at VA facilities and
representatives from the affiliated medical school(s).

Number of VA Medical Resident Primary Care and Specialty Slots in Academic
Year 1999-2000

Continued

             Type of slot             Number of slots
 Primary care
 Family practice                      147
 Geriatric medicine                   170
 General internal medicine            3,476
 Gynecology                           21
 Obstetrics/gynecology                7
 Preventive medicine                  38
 Specialty care
 Addiction psychiatry                 1
 Allergy and immunology               9
 Anesthesiology                       187
 Colon and rectal surgery             1
 Cardiovascular disease               266
 Critical care                        7
 Dermatology                          140
 Diagnostic radiology                 262
 Emergency medicine                   6
 Endocrinology and metabolism         64
 Geriatric psychiatry                 28
 Gastroenterology                     180
 Hematology                           16
 Hematology/oncology                  107
 Infectious diseases                  81
 Nephrology                           89
 Neurology                            255
 Neurological surgery                 57
 Nuclear medicine                     29
 Occupational medicine                17
 Ophthalmology                        237
 Orthopedic surgery                   216
 Otolaryngology                       162
 Pathology                            198
 Pulmonary/critical care              159
 Plastic surgery                      47
 Physical medicine and rehabilitation 189
 Psychiatry                           778
 Radiation oncology                   24
 Rheumatology                         48
 General surgery                      706
 Thoracic surgery                     49
 Urology                              179
 Vascular surgery                     7
 Total                                8,660

Note: Because the numbers of residency slots for each specialty were
rounded, the number of slots totals to 8,660 instead of 8,659; VA actually
has a total of 8,659.2 slots.

Percentage of Residency Slots in Primary Care Training, by Network

P

          Academic year 1995-96
          Percentage of  Number of     Total number

 Network  primary care   primary care  of residency

          slots          slots         slots
 1        36             192.7         530.4
 2        35             91.4          261.9
 3        35             220.0         621.0
 4        48             157.5         330.7
 5        43             102.5         236.5
 6        42             153.0         368.3
 7        39             177.0         452.0
 8        34             191.0         558.5
 9        42             232.5         553.5
 10       37             97.0          259.0
 11       40             130.5         323.5
 12       38             256.0         679.5
 13       42             91.0          214.7
 14       41             82.7          200.0
 15       39             134.0         347.1
 16       35             242.0         696.3
 17       36             124.0         340.0
 18       49             148.4         303.3
 19       39             89.3          226.3
 20       45             118.3         261.3
 21       35             135.0         388.5
 22       36             276.0         757.6
 National 39             3,442.0       8,910.0

 Academic year 1999-2000                1995-96 versus 1999-2000

             Number of                                         Change in
                                                               total
 Percentage  primary     Total          Percentage   Change in
 of primary  care
 care slots  resident    number of      change in    number of number of
             slots       residency      primary care primary   residency
                         slots          slots        care slotsslots
 45          230.8       515.4          9            38.1      -15.0
 42          104.2       245.9          7            12.8      -16.0
 46          270.8       587.5          11           50.8      -33.5
 56          178.5       320.9          8            21.0      -9.8
 54          131.7       246.0          11           29.2      +9.5
 48          168.3       348.8          6            15.3      -19.5
 49          213.1       438.4          10           36.1      -13.6
 45          246.6       546.5          11           55.6      -12.0
 50          262.9       528.3          8            30.4      -25.2
 50          126.1       253.9          13           29.1      -5.1
 50          161.0       319.0          10           30.5      -4.5
 47          310.2       656.8          9            54.2      -22.7
 48          93.9        197.6          6            2.9       -17.1
 52          100.7       194.5          11           18.0      -5.5
 48          162.0       339.0          9            28.0      -8.1
 45          309.9       682.1          10           67.9      -14.2
 46          154.6       333.0          10           30.6      -7.0
 58          176.0       304.7          9            27.6      +1.4
 50          117.6       233.9          11           28.3      +7.6
 55          148.4       268.8          10           30.1      +7.5
 44          166.5       379.7          9            31.5      -8.8
 45          322.6       718.5          9            46.6      -39.1
 48          4,156.0     8,659.0        9            714.0     -251.0

Selected VA Facility and Affiliated Medical School Residency Slots Before
and After VA Residency Realignment

                       Percentage of                          Total number of
                       residents in        VA residency       residency slots
                       primary care        slots              associated with
                       slots                                  medical school
 VA facility/affiliated
 medical school        1995-96 1999-2000   1995-961999-2000   1995-96 1999-2000
 Albany/Albany Medical
 College               37      43          81.4   66.9        419     381
 Albuquerque/University
 of New Mexico         36      54          99.8   112.9       424     439
 Gainesville/University
 of Floridaa           27      37          122.0  110.0       591     547
 West Los Angelesb     35      46          328.6  322.0
 University of
 California LA                                                1,900   1,900
 University of Southern
 California                                                   1,000   1,000
 Muskogee/University of
 Oklahoma              81      100         21.0   4.0         159     144
 Philadelphia          42      43          123.0  92.2
 University of
 Pennsylvania                                                 783     938
 MCP Hahnemann
 University                                                   703     587

Notes: Although the number of residency slots at the Albany Medical College
went down from academic year 1995-96 to 1999-2000, VA did not initiate any
of the reductions.

Although the number of residency slots at the University of Florida went
down from academic year 1995-96 to 1999-2000, VA initiated seven of these
reductions and only one of these was due to VA's residency realignment
effort.

aIntegration of Gainesville facility with Lake City facility.

bIntegration of West Los Angeles facility with Sepulveda and Los Angeles
outpatient clinics.

Summary of Community-Based Outpatient Clinic Data for the Six VA Facilities
Visited

                                     Range of
 Location of VA Number of            distance         Are community-based
 facility       community-based      between          clinics utilized for
 visited        outpatient clinics   clinics and VA   training residents?
                                     facility (in
                                     miles)
 Albany         10                   5 to 205         No
 Albuquerque    5                    139 to 240       No
 Gainesvillea   5                    35 to 180        No
 West Los
 Angelesb       8                    5 to 192         No
 Muskogee       2                    50 to 65         No
 Philadelphia   2                    45 to 60         No

aIntegration of Gainesville facility with Lake City facility.

bIntegration of West Los Angeles facility with Sepulveda and Los Angeles
outpatient clinics.

Comments From the Department of Veterans Affairs

The following is GAO's comment on the Department of Veterans Affairs' letter
dated March 13, 2000.

1. No revision. Although we did not mention the Residency Realignment Review
Committee by name, we gave prominence to the realignment of the graduate
medical education program in the opening paragraph and in the results in
brief of this report.

GAO Contact and Staff Acknowledgments

Walter Gembacz, (202) 512-6982

The following staff made key contributions to this report: John Borrelli,
Marcia Mann, Sigrid McGinty, Maria P. Vargas, and Stefanie Weldon.

(406178)

Table 1: Changes in the Number of Residency Slots Supported by VA 7

Table 2: Changes in Residency Slots at the Six Facilities Visited 8
  

1. Residents have graduated from medical school and are undergoing clinical
training to prepare them to practice medicine independently.

2. The academic year runs from July to June.

3. VA's four core missions are patient care, education, research, and
medical backup to the Department of Defense in the event of a national
security emergency.

4. The Residency Realignment Review Committee was convened in the fall of
1995 by the Under Secretary for Health. The Committee was made up of
nationally acknowledged experts in graduate medical education representing
both VA and the private sector. The Committee was charged with making
recommendations for the possible realignment of the graduate medical
education program.

5. See Carol S. Weissert, Ph.D, and Susan Silberman, Holding Medical Schools
Accountable: A Study of State Legislative Action and Implementation , report
submitted to the Robert Wood Johnson Foundation (Feb. 28, 1998).

6. This facility, which is part of the VA North Florida/South Georgia Health
Care System and also includes the Lake City VA Medical Center, was
integrated in 1997.

7. This facility, which is part of the Greater Los Angeles Health Care
System and also includes the Sepulveda and Los Angeles Outpatient Clinics,
was integrated in 1998.
*** End of document. ***