47 / 240
cutoff point at which a subject is considered ‘‘burned out.’’
Rather, scores are compared to normative data and grouped
into low, average, and high degrees of EE, DP, and PA,
reflecting a continuum of potential responses to work-related
stress. Although the syndrome of burnout is readily identified
with the MBI-HSS, the value of the survey is its ability to
assess a subject’s place along a spectrum of responses to
stress ranging from low to high degrees of burnout, in con-
trast to a dichotomous characterization of burnout as either
‘‘present’’ or ‘‘absent.’’
In the present study, we measured burnout in alumni of
the University of Iowa Hospitals and Clinics (UIHC) resi-
dency program in otolaryngology using the MBI-HSS. In
conjunction with the survey, we also collected demographic
information from survey participants, and correlation
between demographic data and burnout was assessed.
Materials and Methods
Study Design and Participants
The design of this investigation was a questionnaire-based
study of alumni of the UIHC otolaryngology program who
were registered with our alumni relations office as of 2008.
The survey was distributed to a total of 236 alumni.
Survey Administration
A single postal mailing containing the MBI-HSS and a demo-
graphic data sheet was sent. Each mailing included the survey,
the demographic data sheet, a postage-paid return envelope, an
instruction sheet, and a cover letter broadly explaining the
study’s purpose. To maintain confidentiality, survey partici-
pants were instructed not to mark any identifying information
on the survey or return envelope. Participation in the study
was completely voluntary. Completed surveys were returned
by mail and stored anonymously by secretarial staff otherwise
uninvolved with the study. The study protocol, survey instru-
ment, and demographic survey were reviewed and approved
by the University of Iowa Institutional Review Board.
MBI-HSS
The MBI-HSS evaluates the 3 subjective components of
burnout—namely, PA, EE, and DP—through a brief 22-item
inventory. We administered the full MBI-HSS, including all
22 questions, among 3 subscales: 9 questions assess emo-
tional exhaustion, 8 evaluate personal accomplishment, and 5
score depersonalization. Questions regarding emotional
exhaustion include ‘‘I feel like I am at the end of my rope,’’
and ‘‘I feel burned out from my work.’’ Questions such as ‘‘I
have accomplished many worthwhile things in this job’’
assess personal accomplishment, whereas questions such as
‘‘I feel I treat some of my faculty and residents as if they
were impersonal objects’’ measure depersonalization. The
items are listed in no particular order, and respondents are
instructed to assign a frequency to these feelings on a scale
ranging from
never
to
once a day
. Survey respondents link
each statement to a score on a 6-point Likert scale (0 =
never;
1 =
a few times a year or less;
2 =
once a month or
less;
3 =
a few times a month;
4 =
once a week;
5 =
a few
times a week;
6 =
every day
), relating the statement or feeling
to the incidence of its perception. In scoring the survey,
responses were grouped according to category (EE, DP, and
PA) based on a key and added together to generate a score
for each category.
Demographic Data Survey
The demographic data survey consisted of a total of 8 ques-
tions. The survey was designed to collect basic demographic
information, including age (by decade), marital status, length
of time married, and number of children in the home. It also
collected information about the survey participants’ practice,
including type of practice (academic vs private, group vs
solo, single vs multispecialty group), number of hours
worked per week, and number of years in practice. One ques-
tion addressed spirituality by asking respondents to indicate
how religious they are on a 7-point Likert scale (1 =
very
religious;
7 =
not at all religious
).
Statistical Analysis
Statistical analysis was performed using SAS version 9.1 for
Windows (SAS Institute, Cary, North Carolina). Pearson
correlation coefficients were used to summarize associations
between continuous variables. Univariate and multivariate
linear regression was used to measure crude and adjusted
associations for categorical and continuous risk factors for
burnout. Associations and comparisons of means were con-
sidered statistically significant if
P
.05.
Results
Demographic Information
Of 236 surveys distributed, 115 were returned completed
(49% participation rate). Of note, 94% of the study popula-
tion indicated that they were married, with the average
number of years married being 16.3 (range, 1-57 years), and
56% (64/115) of participants indicated that they had chil-
dren living with them. Only 14% (16/115) were younger
than age 40, and equal proportions of respondents were in
their fourth and fifth decades: 23% (27/115) aged 41 to 50
years and 23% (26/115) aged 51 to 60 years. The majority
of the study population was in private practice 56% (64/
115) with 34% (39/115) in academic medicine.
MBI-HSS
Percentages of subjects stratified into low, moderate, and
high levels for each subscale are listed in
Table 1
. Levels
Table 1.
Maslach Burnout Inventory Subscale Stratification
(% of Subjects in Each Stratum)
Low Moderate
High
Emotional exhaustion
71
15
19
Depersonalization
56
17
21
Personal accomplishment
10
24
57
Fletcher et al
25