We chose to study orthopaedic surgeons holding full-time academic
positions at any time between 1959 and 1998 inclusively. A full-time academic
position meant that the surgeon's income was under some sort of
institutional control, such as straight salary, salary with incentive,
dean's tax, or income-limitation plan. Faculty based at outlying
hospitals, commonly Veterans Administration and children's hospitals,
were included as long as they fell under an institutional financial
constraint. This definition of full-time excludes teachers commonly
described as volunteer faculty, clinical faculty, or geographic
full-time faculty. The distinction is not intended to denigrate
the contributions of less-than-full-time academic orthopaedists
to the mission of residency training programs; however, the level
of contribution by such individuals is difficult to standardize
and quantify. Furthermore, the academic, administrative, and financial
pressures to leave a faculty position are different for these individuals
than for those with full-time academic appointments. For programs
that did not exist in 1959 and those that did not have full-time
faculties in that year, we tracked the full-time faculty's academic
longevity and attrition from the time of inception of the faculty
compensation plan. We excluded programs that closed before 1996
and those with plans that started later than 1989. We also excluded
programs based at military institutions, such as the Walter Reed
Army Medical Program; those based at community hospitals, such as
Orlando Regional Health Care; and those based at health-science
universities and freestanding medical colleges, such as Oregon Health
Sciences University and the Medical College of Georgia. These inclusion
and exclusion criteria were chosen so that we could evaluate academic
longevity and attrition at currently existing programs that have
had faculty compensation plans for at least ten years and that were
sponsored by institutions with at least somewhat similar administrative,
budgetary, and tenure considerations.
We collected data from all eighty-two American universities that
had orthopaedic surgery residency programs and that met the above
criteria. (The institutions that were studied are listed in the
Appendix.) These institutions constituted 52 percent of the 157
orthopaedic residency programs in the United States and accounted
for approximately 61 percent (approximately 344) of the approximately
560 orthopaedic surgery residency positions in 19986. (These numbers are approximate because
of ongoing mergers and the resizing of several programs represented
in the numerator or the denominator.)
We devised a questionnaire and faxed it to all members of the
Academic Orthopaedic Society and to members of the American Academy
of Orthopaedic Surgeons who had mailing addresses at university
medical centers. The questionnaire requested information on the
following: name, birth year, verification of full-time status, institution(s)
where the surgeon was (or had been) employed, subspecialty (if any),
year that the surgeon joined the faculty, year that the surgeon
left the faculty (if applicable), next activity (if applicable),
and names of orthopaedic faculty members at the medical school and
residency program attended. Next activity referred to death; retirement;
private practice; and taking another academic position, regardless
of whether the subsequent position was at an institution included
in our study.
Additional names were gleaned from orthopaedic departments that
listed faculty members on departmental letterhead or on their departmental
Web site. Our preliminary working list was then discussed either
face-to-face or by telephone with at least two senior orthopaedists
familiar with each institution, who supplied names that otherwise
would have been overlooked. When possible, we faxed questionnaires
to these additional identified individuals. When a fax number was
not available, we attempted to contact the individual by mail, e-mail,
telephone, or direct contact at a meeting. When an individual could
not be contacted or did not respond, we requested the desired information
from departmental and institutional records. Obituaries published
in The Journal of Bone and Joint Surgery provided
additional information on surgeons who had died prior to data collection.
The data were compiled with use of Microsoft Access (Microsoft,
Redmond, Washington).
When our database was as complete as possible, we again showed
the spreadsheet for each institution to past and present department
chairpersons and other senior faculty members to verify the accuracy
of our information. At the time of the analysis, our data were more
than 99 percent complete for the following categories: birth year,
year that the surgeon joined the faculty, year that the surgeon
left the position (if applicable), subspecialty (if applicable),
and next activity (if applicable). These data were analyzed with
use of Microsoft Access, and the survivorship data were analyzed
with use of Stata (Stata, College Station, Texas). For the purpose
of the survivorship analysis, we defined failure as departure from
a full-time faculty position.
Our study group consisted of 1777 orthopaedic surgeons who at
some point between 1959 and 1998 had held a full-time faculty position
at one or more of eighty-two universities in the United States.
Two hundred and nine of these surgeons had been on the faculties
of at least two institutions, thirty-two had served on at least
three faculties, and four had served on four faculties. Thus, a
total of 2022 positions were studied.
The age at which the surgeon started his or her first faculty
position averaged thirty-two to thirty-four years during the period
of the study, with no decade-by-decade differences. Women constituted
4 percent of the entire study group, with greater numbers and percentages
of women accepting faculty positions in more recent decades (Table I).
A breakdown of the 1777 orthopaedists by subspecialty is given
in Table II.
Foot and ankle surgery and orthopaedic oncology were the two smallest
subspecialties represented. There was a significant relationship between
gender and subspecialty (p < 0.001, chi-square test). Women
were two to three times more likely than men to subspecialize in
pediatric or foot and ankle surgery, and they were three to four
times less likely than men to subspecialize in spine, trauma, or
adult reconstructive surgery. When the data were broken down according
to the decade that the surgeon had begun to practice, there were
noteworthy differences among the subspecialties (p < 0.001,
chi-square test) (Table II). General orthopaedics saw a precipitous
decline, from 40 percent in the early portion of the study to only
7 percent in the most recent decade. Hand surgery saw a 2.5 percent
increase over two decades and then declined. Adult reconstructive
surgery saw an increase in the second decade of the study and then
declined. Spine surgery and trauma surgery each saw a twofold increase
over the entire study period, and sports medicine had more than
a threefold increase.
The attrition of academic orthopaedic surgeons is enumerated
in Table III.
Nearly one-half (835) of the 1777 surgeons were still at their first
academic position, slightly more than one-fourth had moved to private practice,
and almost one-fifth had moved to another academic position. Of
the 320 who had taken a second academic job, more than one-half were
still there, 15 percent had left for private practice, and nearly
one-fourth had taken a third academic position. Of the seventy-four
surgeons who had taken a third position, more than one-half had
remained in that position and one-fifth had moved to a fourth position.
Small percentages of surgeons had left their first, second, or third
academic position because of death or retirement. One percent had
left for private practice and later had returned to a full-time university
position, often at the same institution.
Survivorship analysis showed that 53 percent of the 1777 faculty
members remained at their first job for at least ten years, 35 percent
remained for at least twenty years, and 25 percent remained for
at least thirty years (Table IV). With regard to the total time
spent in these 2022 academic positions, 70 percent of the 1777 faculty
members stayed for at least ten years, 53 percent stayed for at
least twenty years, and 39 percent stayed for at least thirty years.
No differences were noted when academic longevity was analyzed with
regard to gender, although the small number of women in our study
precludes precise statistical analysis.
Overall, approximately one-half, one-third, and one-fourth of
the faculty members remained at their first academic position for
at least ten, twenty, and thirty years, respectively. Longevity varied
significantly according to subspecialty. Hand surgeons (p < 0.005),
spine surgeons (p < 0.011) and general orthopaedic surgeons
(p < 0.072) remained in academic positions for shorter periods
than did surgeons in the other subspecialty groups Table IV).
There were also decade-by-decade differences in longevity. The
greatest longevity was noted in the first decade that was studied
(1959 to 1968) (Fig. 1). A downward trend was noted for
the second decade (1969 to 1978), and an additional decrease was
observed for the third decade (1979 to 1988) (p < 0.005, log-rank
test). The academic longevity for the most recent decade (1989 to 1998)
increased compared with that for the previous time-period and was
similar to that seen in the first two decades of the study.
The growth of full-time orthopaedic academic faculties over the
period of this study is reflected in Table V. There was a rapid and steady increase
in the number of surgeons starting an academic career during the
first thirty-five years of the study, whereas the number accepting
academic posts dropped in the most recent five-year period. The number
of surgeons leaving academic positions increased steadily over the
duration of the study. For the first twenty-five years, at least
2.3 surgeons started a faculty position for every one who departed.
In the most recent five-year period, only 1.3 surgeons started for
every one who left.
Some programs showed more apparent turnover in faculty than others.
However, after taking into account program size, growth, and length
of time that each program had existed, we were unable to clearly
identify programs for which academic longevity was greater or smaller
than the averages presented above.
Our primary purpose was to collect and present the data. Although
our findings are open to various interpretations, they may serve
as a framework for the study of academic demographics in other specialties
and for comparison with future data for orthopaedic surgery. Several
generalizations may be drawn from our findings.
First, despite a growing number of women in academic orthopaedics,
this group remains predominantly male. The percentage of female medical
school graduates was less than 6 percent in 1960, 8 percent in 1970,
23 percent in 1980, and nearly 41 percent in 19961,8.
In contrast, women accounted for 1 percent of orthopaedic residents
in 1977 and for 7 percent in l9961.
Our study shows that women also accounted for approximately 1 percent
of full-time orthopaedic faculty members in the late 1970s and for 7
percent in the 1990s (Table I). Female academic orthopaedists
are more likely to specialize in foot and ankle surgery and in children's
orthopaedics, while trauma surgery, spine surgery, and adult reconstructive
surgery remain overwhelmingly male-dominated. Academic longevity
was not found to be gender-related.
Second, subspecialization has swept though orthopaedics in the
last thirty years13. There was
a great burst of interest in adult reconstructive surgery following
the release of methylmethacrylate for use in total joint replacement
almost thirty years ago, and that subspecialty showed corresponding
growth in the 1970s, with a subsequent decrease that perhaps was
related to restricted Medicare reimbursement. The rise in hand surgery
may likewise be attributed to the technological advances brought about
by microsurgery and limb replantation in the 1970s. Sports medicine
and spine surgery have shown dramatic and more recent growth. Whether
these changes are related to economics, population demographics,
advances in surgical techniques and instrumentation, or other factors is
speculative.
Third, we found that hand, spine, and general orthopaedic surgeons
remained on full-time orthopaedic faculties for significantly shorter periods
of time than average. For hand and spine surgeons, this finding
may reflect the nature of these subspecialties and how dependent
or independent they are on tertiary medical centers for interdisciplinary
care, expensive and sophisticated technology, and ease of access
to appropriate operating-room facilities. The attrition of general
orthopaedic surgeons from academic positions probably parallels
the rapid growth in the subspecialization of orthopaedic surgery over
the time-span of this study.
Fourth, the fact that 47 percent of the orthopaedists who took
a full-time faculty position during the forty-year period of this
study were still at their first position at the time of this writing speaks
to the rapid growth of the specialty and the relative attractiveness
of such positions. The attrition to private practice among orthopaedists in
their second and third academic positions diminished progressively.
Once an orthopaedist left academia for private practice, however, there
was only a 1 percent chance of return. The decade-by-decade differences
in longevity are noteworthy and may imply variation in the relative
attractiveness of an academic position compared with that of private
practice over time. We did not explore the reasons that full-time
faculty members left for private practice, although commonly discussed
reasons were related to such factors as autonomy, compensation,
and leadership. The relative importance of such factors is currently
being explored in a follow-up study of orthopaedists who recently
left an academic position for private practice.
From the data presented in Table V, it is apparent that the rapid
growth in orthopaedic faculty positions has leveled and the number
of orthopaedic surgeons starting full-time academic careers is reaching
an equilibrium. The effects that health-care-financing reform and
other pressures will have on academic orthopaedics are unknown.
Some residency programs are voluntarily downsizing in response to the
perceived abundance of specialists, and a corresponding downsizing
of full-time faculty may ensue. As government support for teaching hospitals
diminishes14, there may be more
pressure for full-time faculty to increase clinical productivity,
probably at the expense of research and teaching, thereby making
academic positions less attractive. On the other hand, decreases
in third-party reimbursement rates for orthopaedists in private practice
may reduce differences in compensation between academic positions
and private practice, thus enhancing the attractiveness of an academic
career. For these reasons, it may be useful to reevaluate these
data in the future.
Roy A. Meals, M.D.
Hugh L. Bassewitz, M.D.
Frederick J. Dorey, Ph.D.
100 UCLA Medical Plaza
Suite 305
Los Angeles, California 90024-6970
E-mail address for R. A. Meals: rmeals@ucla.edu
The Eighty-two Institutions That Were
Included in
the Study, According to Geographic Location
Alabama: University of Alabama, University of
South Alabama
Arizona: University of Arizona
Arkansas: University of Arkansas
California: Stanford University, University
of California at Davis, University of California at Irvine, University
of California at Los Angeles, University of California at San Diego,
University of Southern California
Colorado: University of Colorado
Connecticut: University of Connecticut, Yale-New
Haven Hospital
Florida: University of Florida, University of
Florida at Jacksonville, University of Miami
Georgia: Emory University
Hawaii: University of Hawaii
Illinois: Loyola University, Northwestern University, Southern
Illinois University, University of Chicago, University of Illinois
at Chicago
Indiana: Indiana University
Iowa: University of Iowa
Kansas: University of Kansas
Kentucky: University of Kentucky, University
of Louisville
Louisiana: Tulane University, Louisiana State
University
Maryland: Johns Hopkins University, University
of Maryland
Massachusetts: Tufts University, University
of Massachusetts
Michigan: Michigan State University, University
of Michigan, Wayne State University
Minnesota: University of Minnesota
Mississippi: University of Mississippi
Missouri: University of Missouri, University
of Missouri at Kansas City, Washington University, St. Louis University
Nebraska: University of Nebraska/Creighton University
New Hampshire: Dartmouth-Hitchcock Medical Center
New Mexico: University of New Mexico
New York: Albert Einstein College of Medicine
at Yeshiva University, Columbia University, State University of
New York at Brooklyn, State University of New York at Buffalo, State
University of New York at Stony Brook, State University of New York
at Syracuse, University of Rochester
North Carolina: Duke University, University
of North Carolina, Wake Forest University
Ohio: Case Western Reserve University, Ohio
State University, Wright State University, University of Cincinnati
Oklahoma: University of Oklahoma
Pennsylvania: Pennsylvania State University,
Temple University, University of Pennsylvania, University of Pittsburgh
Rhode Island: Brown University
South Carolina: University of South Carolina
Tennessee: Vanderbilt University
Texas: Texas Tech University, University of
Texas at Houston, University of Texas at San Antonio, University
of Texas Southwestern
Utah: University of Utah
Vermont: University of Vermont
Virginia: Medical College of Virginia/Virginia
Commonwealth University, University of Virginia
Washington: University of Washington
Washington, D.C.: George Washington University,
Georgetown University, Howard University
West Virginia: West Virginia University
Wisconsin: University of Wisconsin