Escolar Documentos
Profissional Documentos
Cultura Documentos
E S E A R C H
E P O R T
258
.15; 95% confidence interval, 0.06 0.40) and 69% less for
teacher clinicians (odds ratio, .31; 95% confidence interval, 0.11 0.88) than for basic researchers. Clinical researchers did not differ from basic researchers in the
likelihood of being at higher rank. Similarly, compared
with basic research faculty, the adjusted odds of being
more satisfied with progress towards academic promotion
were 92% lower for academic clinicians and 87% lower for
teacher clinicians.
Conclusions. Clinician educator faculty were less
likely to be at higher rank at this institution than were
faculty in research paths. Differences in rank may be
explained by lower rank at hire for faculty in these career
paths, time available for scholarly activities, or other
resources available to support scholarship. Retaining clinician educators will require further exploration of barriers to promotion inherent to these career paths and
methods of modifying these barriers.
Acad Med. 2004;79:258 264.
their faculty to increase their specialization, not only in medical content areas
but also in the broader missions of research, education, and patient care.
Most schools have responded by recruiting faculty dedicated primarily to the
patient care and education missions,1
and these clinician educator faculty are
Correspondence and requests for reprints should be addressed to Dr. Thomas, 9033 1830 E. Monument Street,
Baltimore, MD 21287; e-mail: pathomas@jhmi.edu.
For articles on related topics, see pp. 205213, 214
218, and 250 257.
METHOD
Survey Development
We began developing the questionnaire
using a nominal group of departmental
faculty representing several divisions
who worked with previously published
definitions to write mutually exclusive
descriptors for four career paths.13,14
The basic researcher career track was
defined as those spending more than
50% of work time engaged in basic science research and having more than
50% of salary supported through extramural grants. The remainder of a basic
researchers time could be spent engaged in patient care, teaching, and
administrative activities. Clinical researchers were defined as being engaged in clinical research for more than
50% of their time and having more than
50% of their salary supported through
extramural grants. Less than 50% of
their effort was engaged in direct patient
care and as much as 10% could be devoted to administrative or teaching duties. Academic clinicians devoted
70% to 90% of their time to patient care
with teaching activities occupying the
remainder. The teacher clinician
pathway was defined as those spending
less than 50% of their time engaged in
patient care with the remainder devoted
to educational administration and
teaching. Based on these definitions,
faculty were asked to identify the singlebest descriptor of their career path.
Nine job satisfaction items were
drawn from other published surveys of
physicians job satisfaction as well as suggestions from the working group.1517
Other items requested demographic information. Faculty were asked to select categories for years on faculty, rank, years at
rank, years at previous rank, hours worked
per week, and shared family responsibility. The questionnaire (available from
the authors upon request) was pretested
for clarity by 12 faculty. After development, the questionnaire was mailed in
1999 to all full-time MD faculty in the
department at the rank of instructor and
above (268). The questionnaires were
coded for confidentiality and tracking.
Nonresponders received a second mailing and a final phone call and mailing.
Demographic data for the entire department of medicine and JHUSOM were
obtained from the faculty management
systems maintained by the dean of the
School of Medicine.
Statistical Analysis
We used the chi-square test to compare
the demographic characteristics of
respondents and nonrespondents and
assess bivariate associations between
categorical items. We used the KruskalWallis test to investigate differences in
Likert-scale items by career path. We
constructed a global satisfaction score
by averaging the possible responses to
the nine satisfaction items and used a
nonparametric test for trend to assess
correlations between it and items ranking likelihood and importance of remaining at JHUSOM.
We used ordinal logistic regression
methods18 using the proportional odds
model to assess the relationship between academic rank and career path
while preserving the ordinal nature of
the four level-outcome of academic
rank. The coefficients from the proportional odds model provided the log odds
of being at a higher rank versus a lesser
rank, as a function of predictor variables. We added the covariates of age,
gender, years at rank and global satisfaction score to the model. We performed a
separate ordinal logistic regression analysis to investigate the relationship between ordered four-level satisfaction
with progress in career goals toward academic promotion and career path.
259
Table 1
Gender, Age, and Rank Comparisons of Responders and Nonresponders to a Promotion and Career
Path Survey, Johns Hopkins University School of Medicine, 1999
% of Respondents by Career Path
Total
Basic
Clinical Academic Teacher
Department School
Researcher Researcher Clinician Clinician Nonresponders Responders Faculty Faculty
Male
74
75
82
76
82
77
Age 45 yrs.
70
61
63
60
57
63
Rank
Instructor
Assistant
Associate
Professor
9
41
24
26
7
35
33
25
29
32
26
13
8
48
24
20
11
34
29
25
Total (no.)
46
69
38
25
79
RESULTS
Of the 268 faculty surveyed, nine faculty had left the institution. Of the 259
faculty remaining, 180 responded (response rate 69%). Two faculty declined to identify a career path, so comparisons by career path are based on 178
responding faculty. We found no significant differences between 79 nonresponders and 180 responders by gender,
age, or rank. Table 1 shows the responders characteristics compared with faculty of the entire department and school
of medicine. There were no statistically
significant differences in age or gender
by career path.
Workload and Family Responsibility
The proportion of academic clinicians
working more than 60 hours per week
was greater than for other career paths
(see Table 2). Open-ended responses
indicated that academic clinicians averaged 1.4 months of general medicine
attending per year and 2.5 months of
subspecialty attending; teacher clinicians averaged 1.5 months of general
260
78
71
12
37
28
22
12
36
29
23
13
38
27
22
180
259
1508
Table 2
Characteristics of Workload and Family Burden by Career Path for Department of Medicine Faculty,
Johns Hopkins University School of Medicine, 1999
% of Respondents by Career Path
Basic
Researcher
Clinical
Researcher
Academic Clinician
TeacherClinician
50
50
66
34
38
62*
52
48
Marital status
Single
Married
Divorced or separated
9
87
4
10
85
5
5
92
3
28
72
0
Number of children
0
12
3 or more
22
54
24
12
52
36
24
47
29
33
25
42
76
24
83
17
77
23
90
10
Total (no.)
46
69
38
25
Characteristic
*Chi-square test comparison, p .05. Remaining chi-square comparisons were not significant (p .14 for marital status, p
.13 for no. of children, p .51 for share of family responsibility).
DISCUSSION
To our knowledge, this is the first report
of a survey comparing faculty satisfaction and rank across career paths within
one institution. Sixty-four faculty mem-
261
Table 3
Characteristics of Workload and Family Burden by Gender and Career Path for Department of
Medicine Faculty, Johns Hopkins University School of Medicine, 1999
% of Respondents by Career Path
Basic
Researcher
Clinical
Researcher
Academic
Clinician
Teacher
Clinician
p Value*
56
33
42
12
63
57
58
20
.261
.117
Married
Men
Women
94
67
88
75
94
86
79
50
.297
.334
13
60
11
38
10
83
0
50
.544
.319
Total (no.)
Men
Women
Overall
34
12
46
52
17
69
31
7
38
19
6
25
Characteristic
*Chi-square test.
Table 4
Results of Ordinal Logistic Regression Analysis on Promotion and Career Path for Department of
Medicine Faculty, Johns Hopkins University School of Medicine, 1999
Unadjusted
Career Path
OR
95% CI
1.00
1.18
.43
.80
.602.31
.19.96
.331.93
1.00
1.20
.24
.33
.612.35
.10.57
.12.86
Adjusted
p Value
OR
95% CI
p Value
.63
.04
.62
1.00
.53
.15
.31
.241.20
.06.40
.11.88
.13
.01
.03
.60
.01
.02
1.00
.39
.08
.13
.17.89
.03.23
.04.42
.03
.01
.01
*OR is interpreted as the odds of being at a higher rank in a certain career path as compared to the odds in basic science
researchers, adjusting for age, gender, years at rank and global satisfaction score.
OR is interpreted as the odds of being more satisfied with progress in career goals toward promotion in a certain career path
as compared to the odds in basic science researchers.
262
rent single-track system has not eliminated a sense of elitism. The potential
consequences of failing to retain clinician educators in midlevel and senior positions at academic medical centers have
been well-described by others.19
We did not find disproportional representations of women in the clinical
paths or differences in hours worked or
family burdens that would explain potential outside influences on scholarly
productivity. Other barriers to promotion may exist, however, that are more
amenable to interventions. Studies of
women faculty may be helpful models in
understanding these barriers. A multiinstitutional study of women faculty reporting low satisfaction with career
progress found that women had less institutional support (e.g., research funding and secretarial support).20 Another
study noted that women faculty assigned a lower value than men did to
leadership and national recognition;
women faculty had the least time for
scholarly activity and the poorest understanding of promotion criteria.21
Similar themes have appeared in studies
of clinician educators. For instance,
one study found that physicianfaculty
who spent the majority of their time in
clinical activities had less time, mentoring, and resources for an academic career.22 So-called protected time is often
lost in the busy work week of clinical
faculty. Sheffield et al.23 found that clinician educators spent significantly less
time on scholarly activities than is designated for this work. Each of these
differences is a potential opportunity for
institutional intervention.
A major limitation of our study was
its focus on one clinical department
within one institution. Although our
respondents did not differ from nonrespondents demographically, the results
may have been affected by response
bias. Satisfied and successful faculty may
have felt less inclined to complete the
questionnaire. The questionnaire was
developed from some items for which
reliability was not tested, but most items
REFERENCES
1. Barchi RL, Lowery BJ. Scholarship in the
medical faculty from the university perspective: retaining academic values. Acad Med.
2000;75:899 905.
2. Levinson W, Rubenstein A. Mission criticalintegrating clinician-educators into academic medical centers. N Engl J Med. 1999;
342:840 4.
3. Viggiano TR, Shub C, Giere RW. The Mayo
Clinics Clinician-Educator Award: a program to encourage educational innovation
and scholarship. Acad Med. 2000;75:940 3.
4. Lovejoy FH, Clark MB. A promotion ladder
for teachers at Harvard Medical School: experience and challenges. Acad Med. 1995;70:
1079 86.
5. Simpson DE, Marcdante KW, Duthie EH,
Sheehan KM, Holloway RL, Towne JB.
Valuing educational scholarship at the Medical College of Wisconsin. Acad Med. 2000;
75:930 4.
263
264