Participants
Our primary study group was composed of twenty-one orthopaedic surgery
residents from a large university training program. The second study group
consisted of twenty-five full-time orthopaedic surgery faculty members from
two institutions; sixteen were at the same large university training program,
and nine were at a second large university training program. Upon review and
approval of our project by the institutional review boards of our
participating institutions, each study group was surveyed in a voluntary,
anonymous fashion. Each participant received a random code number, records of
which are confidentially maintained by the study administrator.
Survey Instrument
Participants completed a 102-question survey consisting of six
sections:
Background and demographic information.Job satisfaction, assessed with use of the Maslach Burnout Inventory
(twenty-two questions), a validated and accepted
instrument9. It
consists of three subscales: nine items measuring emotional exhaustion, i.e.,
a drained, depleted feeling arising because of excessive psychological and
emotional demands; five items assessing depersonalization, i.e., the tendency
to view others in an excessively detached, impersonal, even cynical manner;
and eight items appraising personal achievement, i.e., a sense of competence
and accomplishment. Maslach Burnout Inventory norms for medical workers were
developed from the responses of a sample of 1104 American doctors and
nurses9. Scores are
compiled for each subscale and are categorized by thirds in accordance with
the normative
distribution9.
Emotional exhaustion and depersonalization correlate with burnout, while
personal achievement is inversely proportional to
burnout9.Psychiatric morbidity, assessed with use of the General Health
Questionnaire-12, a validated, widely used mental disorder screening
instrument that has been translated into eleven languages and used in more
than fifteen
countries16. It has
been found to be a highly reliable indicator of depression, social
dysfunction, anxiety, and somatic
symptoms12. The
twelve items are symptoms of psychiatric morbidity that receive a score of 0
or 1 on the basis of the frequency with which the subject has experienced the
symptom in the recent past, yielding a maximum score of
1217. Scores of 4
or higher have been shown to be indicative of substantial psychiatric
morbidity17.Stressful aspects of life and work, assessed with use of twenty-three items
specifically designed for this instrument. Included were questions identifying
specific stressors (e.g., sleep deprivation, financial concerns, and
relationships with staff) as well as questions regarding perceptions of
harassment, discrimination, and overall life and career satisfaction.Responses to stress, with use of eighteen novel questions related to
self-care and stress management strategies.Relationship issues encountered by married participants and those in
ongoing committed relationships, assessed with use of the Revised Dyadic
Adjustment Scale (fourteen questions) and self-reported descriptions of the
balance between work and home life (fifteen questions). The Revised Dyadic
Adjustment Scale is a validated instrument assessing marital adjustment on a
69-point scale18.
In the Revised Dyadic Adjustment Scale format, "adjustment" is
considered a neutral term that refers to the quality of the relationship as
determined by the
individual18. In
the normative data for the scale, the mean score (and standard deviation) for
couples who scored in the not-distressed range was 52.3 ± 6.6, while
the mean score for those in the distressed range was 41.6 ± 8.2; a
score of =46 is considered to reflect a distressed
relationship18. In
addition, fifteen questions pertaining to the balance between work and home
life were adapted from the work of Geurts et al., who evaluated work-home
interference as a "critical mediating pathway in the relationship
between work and home characteristics and work-related and general
psychological health
indicators."6
Background and demographic information.
Job satisfaction, assessed with use of the Maslach Burnout Inventory
(twenty-two questions), a validated and accepted
instrument9. It
consists of three subscales: nine items measuring emotional exhaustion, i.e.,
a drained, depleted feeling arising because of excessive psychological and
emotional demands; five items assessing depersonalization, i.e., the tendency
to view others in an excessively detached, impersonal, even cynical manner;
and eight items appraising personal achievement, i.e., a sense of competence
and accomplishment. Maslach Burnout Inventory norms for medical workers were
developed from the responses of a sample of 1104 American doctors and
nurses9. Scores are
compiled for each subscale and are categorized by thirds in accordance with
the normative
distribution9.
Emotional exhaustion and depersonalization correlate with burnout, while
personal achievement is inversely proportional to
burnout9.
Psychiatric morbidity, assessed with use of the General Health
Questionnaire-12, a validated, widely used mental disorder screening
instrument that has been translated into eleven languages and used in more
than fifteen
countries16. It has
been found to be a highly reliable indicator of depression, social
dysfunction, anxiety, and somatic
symptoms12. The
twelve items are symptoms of psychiatric morbidity that receive a score of 0
or 1 on the basis of the frequency with which the subject has experienced the
symptom in the recent past, yielding a maximum score of
1217. Scores of 4
or higher have been shown to be indicative of substantial psychiatric
morbidity17.
Stressful aspects of life and work, assessed with use of twenty-three items
specifically designed for this instrument. Included were questions identifying
specific stressors (e.g., sleep deprivation, financial concerns, and
relationships with staff) as well as questions regarding perceptions of
harassment, discrimination, and overall life and career satisfaction.
Responses to stress, with use of eighteen novel questions related to
self-care and stress management strategies.
Relationship issues encountered by married participants and those in
ongoing committed relationships, assessed with use of the Revised Dyadic
Adjustment Scale (fourteen questions) and self-reported descriptions of the
balance between work and home life (fifteen questions). The Revised Dyadic
Adjustment Scale is a validated instrument assessing marital adjustment on a
69-point scale18.
In the Revised Dyadic Adjustment Scale format, "adjustment" is
considered a neutral term that refers to the quality of the relationship as
determined by the
individual18. In
the normative data for the scale, the mean score (and standard deviation) for
couples who scored in the not-distressed range was 52.3 ± 6.6, while
the mean score for those in the distressed range was 41.6 ± 8.2; a
score of =46 is considered to reflect a distressed
relationship18. In
addition, fifteen questions pertaining to the balance between work and home
life were adapted from the work of Geurts et al., who evaluated work-home
interference as a "critical mediating pathway in the relationship
between work and home characteristics and work-related and general
psychological health
indicators."6
Statistical Analysis
Statistical analysis was performed on all completed questionnaires.
Descriptive statistics and pairwise correlations were calculated. The simple
correlation coefficient was used to estimate the strength and to test the
significance of bivariate relationships. The Pearson and non-parametric
Spearman correlations were calculated, and similar results were obtained with
use of the two methods. Pearson correlation coefficients are presented for
consistency. Simple t tests were used to compare mean responses on
standardized scales. Differences generating p values of <0.05 were
considered significant, while those generating p values between 0.05 and 0.1
were considered to suggest a trend.
The first five sections of forty-six questionnaires (twenty-one from
residents and twenty-five from faculty) were completed and analyzed. In
addition, thirty-seven subjects (sixteen residents and twenty-one faculty) who
reported that they were married or involved in an ongoing committed
relationship completed section six of our survey, and these results were also
analyzed for data pertaining to relationship issues. The decision to include
the relationship data from nonmarried subjects who reported ongoing committed
relationships as well as from married persons is supported by the fact that
only one-quarter of American households consist of what most people think of
as a traditional family: a married couple and their
children19.
Furthermore, cohabitors are at least as likely as individuals in their first
marriage to remain together after five
years20,21.
Background and demographic results are reported in
Table II.
Job Satisfaction: The Maslach Burnout Inventory
In comparison with the norms for American health-care
workers7, the
resident group showed high levels of burnout, scoring in the upper third for
emotional exhaustion and depersonalization and in the middle third for
personal achievement. In contrast, the faculty group showed low levels of
burnout, scoring in the lower third for emotional exhaustion, the middle third
for depersonalization, and the upper third for personal achievement
(Table III).
Psychiatric Morbidity
Seven residents (33%) and two faculty members (8%) had a score of =4
(range, 4 to 7) on the General Health Questionnaire-12, indicating significant
psychiatric morbidity (p <
0.01)3.
Life and Work Stress
The average overall quality of life, as rated on a scale that ranged from 0
to 4 points, fell within the "moderately satisfying" range (mean,
2.5 points) for the resident group. By contrast, the faculty group rated their
overall quality of life in the "very satisfying" to
"extremely satisfying" range (mean, 3.6 points). Both groups rated
the overall stress level of their work as "moderately stressful"
(mean, 2.6 points for residents and 2.2 points for faculty); however, both
groups unanimously reported that they would again choose to pursue a career in
orthopaedic surgery.
Responses to Stress
All residents reported having faculty mentors; however, with use of a 0 to
3-point scale, the residents reported that their mentors were of "little
help" or "no help" (mean, 0.9 point). Nine (36%) of
twenty-five faculty participants reported having mentors and that talking with
those mentors helped "a little" (mean, 1.6 points). Residents
discussed concerns with their colleagues "quite a bit," whereas
faculty did so "a little" (mean, 2.1 and 1.2 points,
respectively). Both groups stated that they talked about their concerns with
family, friends, and/or partners "quite a bit" (mean, 2.6 points
for residents and 2.0 points for faculty).
With regard to self-care, neither group reported difficulty unplugging from
work. Neither group reported that they used relaxation techniques or sought
formal counseling to help to cope with stress. Both groups claimed that they
drew on religion or faith in God "a little" (mean score, 1.2
points for residents and 1.7 points for faculty on a 0 to 3-point scale).
Residents reported that they exercised an average of two and one-half times
each month, while faculty reported exercising approximately four times each
month. Both groups denied cigarette use and reported "a little"
alcohol use (mean, 1.2 points for residents and 1.0 point for faculty).
Relationship Issues
Sixteen residents (76%) and twenty-one faculty members (84%) reported being
married or in an ongoing committed relationship. Responses to relationship
inquiries are scored on a 0 to 5-point scale with a maximum score of 70 points
and with a score of <46 points indicating relationship distress. The
majority of subjects from each group scored above the distressed range on the
Revised Dyadic Adjustment Scale with regard to marital adjustment and
satisfaction; the mean score (and standard deviation) was 49.7 ± 7.3
points (range, 38.0 to 62.0 points) for residents and 51.7 ± 6.6 points
(range, 39.8 to 66.0 points) for faculty. However, three residents and six
faculty members scored within the distressed range.
Residents reported, on the average, being "fairly satisfied"
(mean, 5.3 points on a 1 to 6-point scale) with the overall quality of their
marriage or relationship, whereas faculty reported being "extremely
satisfied" (mean, 5.8 points). Both groups reported that work conflicted
with family life "occasionally" (mean, 2.7 points for residents
and 2.4 points for faculty on a 0 to 4-point scale).
The residents with a committed relationship reported that their spouses
worked an average of 31.8 hours (range, zero to sixty hours) each week outside
the home. The faculty members with a committed relationship reported that
their spouses worked an average of twenty hours (range, zero to fifty hours)
per week outside the home. Both groups reported spending an average of
forty-five to ninety minutes alone and awake with their spouse on a typical
workday. Neither group viewed the work schedules or commitments of their
spouses as a substantial source of family stress. Additionally, both groups
perceived that their spouses had more frequently made career sacrifices. Both
groups perceived that their families paid attention to their feelings (mean,
1.9 points for each group on a 0 to 3-point scale) and appreciated the way
that the subject handled his or her work "quite a bit" (mean, 1.6
points for residents and 1.8 points for faculty members).
Regarding support outside their family or relationship, both groups
reported that their colleagues and their colleagues' families were "a
little" supportive (mean, 1.3 points for each group on a 0 to 3-point
scale). However, the department was perceived by both groups as being
"not at all" helpful in facilitating their spouses' adjustment
(mean, 0.5 point for residents and 0.6 point for faculty). In fact, potential
efforts by the department to assist the spouse in understanding the work
stress of the subjects were predicted by the residents to be only "a
little" helpful (mean, 0.8 point, on a scale of 0 to 1 point), while the
faculty predicted that these efforts would help "not at all"
(mean, 0.4 point).
Finally, the resident group rated the overall balance between work and home
life as "somewhat satisfying" (mean, 4.1 on a scale of 1 to 6
points), while the faculty rated the balance as "fairly
satisfying" (mean, 4.6 points).
Positive Correlations
The Pearson and nonparametric Spearman correlations were calculated, and
similar results were obtained with use of the two methods. The Pearson
correlation coefficients are presented for consistency.
Burnout
Emotional Exhaustion
Among residents, high levels of emotional exhaustion were correlated with
anxiety about clinical competence (p < 0.02; correlation coefficient,
0.50), increased conflict between work and home life (p < 0.001;
correlation coefficient, 0.66), stress in relationships with faculty (p <
0.01; correlation coefficient, 0.51) and senior residents (p < 0.07;
correlation coefficient, 0.41), and increased perceptions of work as stressful
(p < 0.002; correlation coefficient, 0.62).
Among the faculty members, increased emotional exhaustion correlated with
anxiety regarding clinical competence (p < 0.002; correlation coefficient,
0.57), worry about the future because of the number of orthopaedic surgeons in
the field (p < 0.02; correlation coefficient, 0.23), stress in
relationships with other faculty (p < 0.003; correlation coefficient,
0.63), financial concerns (p < 0.02; correlation coefficient, 0.45),
increased perception of work as stressful (p < 0.006; correlation
coefficient, 0.53), and increased conflict between work and home life (p <
0.08; correlation coefficient, 0.35).
Emotional exhaustion tended to be reduced by increased time alone with
their spouse (p < 0.09; correlation coefficient, —0.41) in the
resident group and by increased perception of support from colleagues and
colleagues' families (p < 0.06; correlation coefficient, —0.39) in
the faculty group.
Depersonalization
For residents, as their reported number of work hours increased, so did
their score on the burnout subscale for depersonalization (p < 0.01;
correlation coefficient, 0.55). The opposite pattern was observed with
faculty. As their work hours increased, faculty members had decreased levels
of depersonalization (p < 0.02; correlation coefficient, —0.44).
Stress in relationships with nursing staff correlated with increased
depersonalization in residents (p < 0.039; correlation coefficient, 0.46)
and faculty (p < 0.03; correlation coefficient, 0.43). For residents,
increases in anticipated debt load at the completion of training (p <
0.001; correlation coefficient, 0.66) increased the level of
depersonalization. For faculty, higher scores were correlated with increased
alcohol use (p < 0.02; correlation coefficient, 0.47) and greater levels of
concern regarding alcohol and drug abuse (p < 0.007; correlation
coefficient, 0.53).
For residents, having a father who is a physician correlated with lower
ratings on emotional exhaustion (p < 0.05; correlation coefficient,
—0.42) and depersonalization (p < 0.04; correlation coefficient,
—0.46). Among faculty, the levels of depersonalization decreased as the
number of children they had increased (p < 0.07; correlation coefficient
—0.38). Depersonalization scores also were lower among faculty members
who reported a better quality relationship with their mother (p < 0.07;
correlation coefficient, —0.37).
Personal Achievement
For residents, being a parent was found to correlate with increased scores
on personal achievement (p < 0.05; correlation coefficient, 0.47). The
scores on personal achievement also increased as satisfaction from talking
with colleagues informally increased (p < 0.05; correlation coefficient,
0.43) or as satisfaction from talking with friends and family about concerns
increased (p < 0.007; correlation coefficient, 0.58).
The levels of personal achievement among faculty members correlated with
their ratings of the overall quality of their marriage (p < 0.04;
correlation coefficient, 0.41) and overall work-family balance (p < 0.05;
correlation coefficient, 0.39). Surprisingly, for faculty, the personal
achievement level increased as the number of hours that his or her spouse
worked outside the home each week increased (p < 0.002; correlation
coefficient, 0.58).
Psychiatric Morbidity
Among residents, increased stress in relationships with senior residents
correlated with increased scores on the General Health Questionnaire-12 (p
< 0.04; correlation coefficient, 0.45) indicating psychiatric morbidity,
whereas satisfaction from speaking with a mentor was associated with decreased
scores (p < 0.02; correlation coefficient, —0.48). For faculty,
increased General Health Questionnaire-12 scores were found to correlate with
increased levels of worry about the number of orthopaedic surgeons in the
field (p < 0.0004; correlation coefficient, 0.67).
Relationship Issues and Conflict Between Work and Home Life
In the resident group, increased conflict between work and home life was
found to correlate with increased levels of emotional exhaustion (p <
0.001; correlation coefficient, 0.66), General Health Questionnaire-12 scores
(p < 0.0075; correlation coefficient, 0.57), and depersonalization (p <
0.01; correlation coefficient, 0.54).
As faculty made more time for hobbies, their emotional exhaustion levels
diminished (p < 0.04; correlation coefficient, —0.41), personal
achievement levels increased (p < 0.03; correlation coefficient, 0.44), and
psychiatric morbidity (General Health Questionnaire-12) levels declined (p
< 0.07; correlation coefficient, —0.37).
This study is the first, as far as we know, to examine job stress and
satisfaction and the psychological and social functioning of orthopaedic
residents and faculty. Our study group was small and included residents from
only one institution, so it is difficult to generalize our results. Our
results showed a great disparity in burnout and psychiatric morbidity between
the resident group and the faculty group. Whether this variance arises from
generational differences or stage of career development cannot be determined
from these results alone. Age may influence the vulnerability to burnout. In a
study by Campbell et al. that surveyed 582 practicing surgeons, substantially
higher levels of burnout were identified in younger
surgeons8.
Longitudinal follow-up of this population is planned to evaluate the changes
over time; furthermore, we hope to expand the current study to include other
orthopaedic residency programs as well as orthopaedic surgeons in private
practice in an effort to fortify our findings.
Our results revealed a substantial level of burnout in this resident
population, with scores in the upper third on both emotional exhaustion and
depersonalization. These results reflect a tendency toward cynicism and a view
of patients as inanimate objects by doctors who feel overwhelmed, drained, and
depleted9. An
average level of personal achievement was maintained in this population;
however, a study by Schaufeli and van Dierendonck supported a two-dimensional
conception of burnout including only emotional exhaustion and
depersonalization22.
Thus, the preservation of an average level of personal achievement may not
temper the resident burnout level to any substantial degree. Indeed, despite
maintaining a reasonable sense of personal accomplishment, the respondents who
showed greater levels of burnout also reported increased levels of anxiety
concerning their own clinical competence.
Other factors associated with burnout among the residents in our study
included increases in work hours per week; conflict between work and home
life; stress in relationships with faculty, nursing staff, and senior
residents; debt load; and perceptions of work as stressful. Protective factors
included being a parent, spending more time alone with a spouse, having a
father who is or was a physician, and deriving greater satisfaction from
speaking about concerns with colleagues, friends, and family.
The faculty group, by contrast, showed low levels of burnout, scoring below
average for emotional exhaustion, within the average range for
depersonalization, and above average for personal achievement. These findings
contradict those of Campbell et al., who noted that orthopaedists are more
likely than other surgeons to suffer from high degrees of burnout,
particularly on the subscale of
depersonalization8.
It is tempting to attribute the low level of burnout in the faculty group to
practicing in an academic setting; however, Campbell et al. also investigated
the impact of caseload, practice setting, and percentage of the patient
population covered by health maintenance organizations but found no
correlation between these practice characteristics and the prevalence or
degree of
burnout8.
Factors associated with burnout among the faculty members in our study
included increases in anxiety with regard to clinical competence, worry about
the future because of the growing supply of orthopaedic surgeons, stress in
relationships with other faculty, financial concerns, perceptions of work as
stressful, alcohol use, concerns with regard to drug and/or alcohol abuse, and
conflict between work and home life. Protective factors included increases in
the number of children, spouse's work hours per week, perceptions of support
from colleagues and colleagues' families, overall quality of the marriage or
relationship, and time spent on hobbies.
Residents showed high levels of psychiatric morbidity, as assessed by the
General Health Questionnaire-12, in comparison with the faculty group.
However, the finding of significant psychiatric abnormalities in one-third of
our resident group is similar to levels reported among other resident
groups3. In a 1991
study of anesthesia, pediatric, and psychiatry residents at Jefferson Medical
College, Samuel et al. reported that the prevalence of psychiatric morbidity,
as assessed by the Beck Depression Inventory, was only
17%23. However,
they suspected that symptoms were underreported because of participant
statements questioning the anonymity of the study or revealing a belief that
"it was best to always present themselves in a positive light for fear
of a negative performance
evaluation."23
The finding of similar levels of psychiatric morbidity in other resident
populations is not intended to diminish the importance of such dysfunction
occurring in one-third of the resident group in the present study. It does,
however, lend support to the idea that such a dysfunction is a consequence of
residency, as a construct, rather than an isolated outcome for a particular
group of trainees in a particular program. Indeed, the low prevalence of
psychiatric morbidity found in our faculty group (8%) suggests that, beyond
the training period, orthopaedic surgery may even be associated with a
decreased prevalence of psychiatric dysfunction in comparison with that among
physicians in other
specialties7,8.
The frequency of positive feedback from patients and their families, the
relative reliability of good and excellent outcomes of orthopaedic procedures,
and the current high income potential for orthopaedic surgeons may help to
protect job satisfaction and reduce mood disorders. In fact, the sole factor
correlating with psychiatric morbidity in our faculty group was a perception
of increasing numbers of practicing orthopaedic surgeons. It is of interest
that only the few faculty physicians who reported depression and anxiety also
reported anxiety-provoking situations such as a hostile marketplace. This
finding suggests that it is the concern about competition that may be
associated with psychiatric dysfunction in some individuals.
The disparity in burnout and psychiatric morbidity between orthopaedic
residents and the faculty who teach them may arise from the contrasting levels
of control and autonomy experienced by the two groups. The
demand-control-support model of the impact of work characteristics on
psychological health is predicated on the notion that jobs with high
psychological demands, minimal autonomy, and low levels of support from
superiors are associated with a higher prevalence of psychological health
complaints6. This
model is supported by extensive research in industry, a few studies of
health-care professionals, and one study of 166 medical residents in the
Netherlands6. It is
widely recognized that resident physicians are subject to high demands and
afforded little discretion over their work content or
schedules3,6,7,13,14,21.
Although our instrument did not specifically address autonomy and control, our
results revealed that difficult relationships with faculty and senior
residents correlated with increased levels of emotional exhaustion and
psychiatric morbidity among junior residents. Additionally, having a physician
father correlated with a decreased level of burnout among residents possibly
because of enhanced perceptions of support.
Inflated financial pressures on current residents may also contribute to
the observed differences in quality of life and burnout between residents and
faculty. Increased anticipated debt at the completion of training correlated
strongly with increased depersonalization among our resident group. The
faculty population is not immune to financial stressors, but they are afforded
some protection by their substantially higher income level. In addition, debt
load at the completion of training has dramatically increased in the recent
past1,3,12.
Furthermore, while salaries have increased over the past two decades,
purchasing power (in 1967 dollars) has
decreased1-3,24.
It has been estimated that comfortably paying off a debt of $75,000 would
require a salary of $140,000 per
year24. Our
resident population reported a mean debt of greater than $75,000, yet their
current average salary is under $40,000 per year.
On a positive note, the majority of resident and faculty marriages and
relationships appeared to be doing well. This finding suggests that, despite
their current levels of stress and emotional dysfunction, these residents had
in general maintained the ability to sustain an intimate, rewarding personal
relationship. It is of concern, however, that the marriages or relationships
of almost one in five residents and more than one in four faculty members
scored within the distressed range. It is of note that this was the only
negative parameter in which the faculty group scored higher than the resident
group.
Our results concerning the current quality of life and psychological
wellbeing of orthopaedic residents raised the question of appropriate
intervention. Compared with other specialties, surgical residencies offer the
fewest supports3. In
a study of the anesthesia, pediatric, and psychiatry residents at Jefferson
Medical College, Samuel et al. reported that 80% of the residents were
interested in support groups for themselves or their spouses and that 30% to
60% of the residents were interested in individual counseling or
psychotherapy23.
The ideal intervention would improve resident quality of life and
psychological well-being without detracting from the quality of the education
or the level of patient care. A limit on work hours per week has been shown to
improve resident quality of life, increase reading time, increase intraining
examination scores, and increase the volume of surgical cases per resident in
a nonorthopaedic surgical training
program11. Whether
these benefits generalize to orthopaedic programs will become evident given
the recent implementation of resident work-hour restrictions.
Alternate strategies for improving quality of life among residents include
increased support mechanisms provided by the residency program, the
incorporation of stress management and effective emotional management training
into medical school and residency curricula, and the creation of national
financial assistance for medical
training1,3,5,24,25.
Support mechanisms provided by some residency programs include formal or
informal support groups and individual counseling; financial counseling to
address loan repayment, tax preparation, and insurance issues; professional
and career counseling; legal advisors to address malpractice and contract
issues; spousal support to address both emotional adjustment and practical
guidance for negotiating a new city; and even child
care3. Our study did
not assess the utilization of psychiatric services or resident assistance
programs offered by the medical school, although neither group reported use of
counseling in general. The current mentoring program, instituted with the
intention of providing professional and emotional support to residents,
matches each resident with a faculty mentor and requires biannual meetings.
Unfortunately, our results demonstrated little perceived benefit to either
residents or faculty. Our findings indicate that current program efforts to
provide support are viewed as inadequate by the residents; however, our data
also revealed low expectations of benefit from additional support efforts.
Furthermore, evaluation of the impact of stress management training on burnout
levels has indicated that associated benefits are transitory, expiring at
approximately six months unless frequent refresher training is
provided25.
In conclusion, our data revealed high levels of burnout among surveyed
orthopaedic residents, yet low levels among surveyed orthopaedic faculty.
One-third of our population of residents reported symptoms of psychiatric
morbidity. Factors decreasing the quality of life and increasing the emotional
dysfunction among residents included workload, debt load, and difficult
relationships with superiors. Current support strategies employed by the
training program are perceived as insufficient or ineffectual. Faculty data
obtained in this study suggested the possibility that low job satisfaction and
quality-of-life assessments improve after the completion of training, at least
among those engaged in full-time academic practice.
The primary deficit of this study was the small study population. It is
hoped that our findings of high levels of burnout and psychiatric morbidity
among orthopaedic residents will stimulate additional research in this area.
Expansion of the subject group to include additional residency programs,
incoming residents, and graduates is planned. In addition, creation of a
Web-based survey instrument and a companion instrument to be completed by the
physician's spouse or significant other is under consideration. Some
alteration of the survey instrument may assist in eliciting variance within
and between programs. Regarding mentorship, inquiry was limited to whether the
subject had a mentor and whether speaking about concerns with that mentor
helped the subject to cope with stress. An expanded inquiry might provide
insight into the deficiencies of the current mentorship program. Inclusion of
more items specifically related to residency and orthopaedics, as well as
pre-training experiences and functioning, would assist in identifying
protective resident characteristics as well as risk factors for
decompensation. Such an instrument might enable a program to identify
residents at risk and intervene preemptively.
Readministration of the questionnaire to the original population is planned
in an effort to provide for longitudinal evaluation of variations in stressors
and quality of life with progression through and beyond residency.
Longitudinal information may also highlight indicators of the hardiness of
residents and elucidate the characteristics that allow some residents to pass
through residency unscathed while others founder. One of the primary strengths
of this study is its provision of a benchmark against which to evaluate the
impact of residency changes such as the recently implemnted resident work-hour
initiative.