2009; 31: 375–378
Burnout assessment in house officers:
Evaluation of an intervention to reduce stress
JAY M. MILSTEIN, BONNIE J. RAINGRUBER, STEPHEN H. BENNETT, ALEXANDER A. KON,
CYNTHIA A. WINN & DEBORA A. PATERNITI
University of California Davis Health System, Sacramento, USA
Background: Medical house officers are at increased risk for stress related symptoms leading to professional burnout.
Aims: Measure burnout in house officers and establish whether utilization of a psychotherapeutic tool individually by physicians
reduces symptoms characteristic of burnout.
Method: Two groups of pediatric house officers at the University of California Davis Health System completed a Maslach Burnout
Survey (MBS) at the beginning and end of a three-month period in 2003. An Intervention group (7 of 15 enrolled) was trained
in the use of a self-administered psychotherapeutic tool. Outcome Measures were MBS scores and a qualitative interview of
intervention group members.
Results: There were no significant differences between the two groups, prior to the study or over time. Qualitative interviews
revealed that subjects experience stressors in relation to their professional activities, but already utilize some elements of the tool
and were too busy to implement the entire tool systematically.
Conclusions: Pediatric trainees did not seem to manifest burnout symptoms based upon the MBS; interviews suggested that some
do experience significant stress, although manifestations and responses were varied, some may be at risk. Methods identifying
individuals at risk for burnout, and interventions to cope with stress may be valuable to their training.
Physicians are vulnerable to both short and long-term stress-
related problems that can lead to burnout, and it is possible
that it can be initiated during residency. Indeed, even with
the advent of a limited work-schedule (Gopal et al. 2005),
medical residents remain susceptible to stress from pressure
and fatigue, which ultimately affects their job performance
(Feddock et al. 2007). The short-term effects of stress may
manifest itself with self-limited cognitive, behavioral, physical
or emotional symptomatology (Benson & Stuart 1993);
however, the long-term effects may be more significant
including professional burnout and ineffectiveness, organic
disease, as well as an increased risk of suicide, drug and
alcohol addiction (Flahtery & Richman 1993: Lindeman et al.
1996: Hsu and Marshall 1997: Tyssen et al. 2000). Therefore,
the early identification of stress in the medical teaching
environment by both students and teachers, and an effective
means of intervention well ahead of burnout, should be
considered a useful measure or precaution of a comprehensive
training program that would contribute to the long-term
success of its students. At the same time, it is unclear what
resources of a training program should be encumbered to both
monitor and intervene, if indeed, burnout becomes manifest
(McCue & Sachs 1991).
Hospitals, emergency medical service programs and fire
departments to assist workers in coping with such stress to
prevent burnout now commonly use debriefing techniques,
specifically critical incident stress debriefing (Mitchell 1983:
Mitchell & Everly 1997). While use of debriefing techniques
targeting both the victims as well as the caregivers has become
commonplace in the face of the myriad of catastrophic events,
its use in the stresses associated with day-to-day care of sick
patients remains limited. We previously reported use of the
debriefment process in combination with a psychotherapeutic
tool with an emergency medical team and a small group of
physicians in-training involved in the care of a critically ill
infant (Milstein et al. 2002). Since the effects of the debriefment
process appeared favorable, this led us to the hypothesis that
utilization of a psychotherapeutic tool by medical residents
in training may reduce symptoms characteristic of stress and
Inventory is a useful instrument for both program and
. In absence of house officer burnout at the program level
and an active program to teach stress reduction and
coping skills, the instrument is of utility in assessing
individuals at risk.
Correspondence: Jay M. Milstein, MD, Department of Pediatrics, 2516 Stockton Boulevard, Sacramento, CA 95817, USA. Tel: 916-734-8921;
fax: 530-752-6215; email: firstname.lastname@example.org
ISSN 0142–159X print/ISSN 1466–187X online/09/0040375–4 ? 2009 Informa UK Ltd.
This study was approved by the Institutional Review Board at
the University of California Davis (UCD) and conducted at the
University of California Davis Health System, in Sacramento
CA. Volunteer subjects were recruited from the entire house
staff from the UCD Pediatric Residency Training Program
to ‘complete surveys to assess responses to stress’. The study
was performed during the first academic quarter of 2003,
beginning in July, when house officers were restricted to work
less than 80 hours per week and no more than 30 hours of
continuous duty by the Accreditation Council for Graduate
Medical Education (ACGME) (Gopal et al. 2005).
The residents were randomly assigned to either a control or
study group, and then completed the Maslach Burnout
Inventory (MBI). The MBI assesses the risk of burnout relative
to a large population of physicians, which is expressed in
scales related to degree of emotional exhaustion (EE),
depersonalization (DP), and sense of personal accomplish-
ment (PA) (Maslach & Jackson 1996). The combination of High
co-incident scores, EE?27, and DP?10, plus a low score
PA?33, identified individuals at the highest risk for burnout.
After survey completion, the control group was dismissed.
The study group received 45-minute instruction in the use
of a psychotherapeutic technique, BATHE, adapted from
The Fifteen Minute Hour (Stuart & Lieberman 2002).
Following the acronym, one reflects on the background of
the situation that may have generated stress professionally,
examines one’s affect, or how one feels about a situation,
analyses the most troublesome aspects of the situation,
individual’s expertise, reflects upon how one handled the
situation, and provides oneself empathy (supportive com-
ments)for one’s responses
(Appendix 1, available at www.medicalteacher.org). The
participants were encouraged to use the technique approxi-
mately three times per week over the subsequent three-month
period and particularly when experiencing professionally
related stress. Three months later, the study and control
groups were gathered separately to complete the MBI survey a
second time (t2). Qualitative interviews were then conducted
in the study group volunteers to establish the utility,
effectiveness and deterrents of the BATHE technique.
Quantitative analysis of the MBI was performed utilizing
a two-factor (group and time) ANOVA with repeated
measures where scores were interpreted on a continuous
scale based on the instrument developer’s design (Zar 1984;
Maslach & Jackson 1996). Qualitative phenomenological
interviews were conducted in which participants in the
intervention group were asked to talk about (1) stresses they
experienced, (2) how those stresses showed up for them
(physical, behavioral, emotional, or cognitive manifestations),
and (3) whether the BATHE technique was or was not helpful
in managing stress. Interviews were audiotape recorded
and transcribed verbatim. Ultimately five coding categories
were identified: (a) stressors, (b) manifestations of stress,
(c) ways of handling stress, (d) the utility and effectiveness
of BATHE, and (e) deterrents to using BATHE.
There were 33 potential subjects amongst the entire pediatric
house officer population. Some were not able to volunteer in
response to our study invitation because they were on service
in other facilities or were subject to the house officer hour
restrictions and could not be on hospital grounds. Fifteen of
the subjects enrolled with 8 (4F/4M) randomized to a control
group and 7 (4F/3M) to a study group.
The findings from the MBI survey are summarized in
Figure 1. For Emotional exhaustion, both groups scored in
the average range relative to other physicians. There were no
significant differences between the two groups, prior to or
after the three-month interval. For the Depersonalization,
members of the control group scored in the average range
while the members of the study group scored higher relative to
other physicians, but the difference was not significant
between the study and control groups. For the sense of
Personal accomplishment, the control group scored in the
average range relative to other physicians, whereas the
study group scored in the low to average range over time.
Again, there were no statistical differences in the perception
of personal accomplishment between the two groups at the
outset of this study or over time. Although the surveys indicated
that there were no significant differences in stress or burnout
and control) were in average range for all facets of burnout. The Assessment identifies tercile thresholds identifying particular
individuals at burnout risk at a given time: identified by high scores for emotional exhaustion (EE?27), depersonalization
(DP?10) plus a low score for personal accomplishment (PA?33) designated by open circle/square (Maslach & Jackson 1996).
Maslach burnout assessment in pediatric residents. Combined mean scores (mean?SD) for both groups (treatment
J. M. Milstein et al.
between groups, the survey did identify individuals at risk for
burnout (Figure 1, available at www.medicalteacher.org).
Six of the seven in the study group participated in the
qualitative interviews. Although no one in the study group
reported using the BATHE tool in a systematic conscious
manner, several participants mentioned they did reflect on their
practice, tried to step back and put things in perspective, or
reviewedtheir dayintheir mind.Thispractice wasanautomatic
way they had developed to work on managing stress. In
general, the study subjects did utilize partial elements of the
BATHE technique as strategies to reduce stress (Table 1).
Our results indicate that our pediatric house officers experi-
enced a level of stress and burnout that is, on average, at the
same level as that common to the population of physicians in
general. In anticipation that residents were indeed experien-
cing a high level of stress on their rotation, we simultaneously
measured burnout and tested a simple intervention that might
be used as part of a house officer-training program. In the
absence of a comprehensive training program to reduce
House officer stress and document need, our intent was to
measure the burnout in our population, and at the same time,
establish whether a brief intervention, taught in a didactic
fashion, would be an effective psychotherapeutic tool to
potentially reduce house officer stress. Overall, our sample of
the Pediatric House officers indicated that the population as a
resources for a comprehensive program was not justified at
this time. In addition, the intervention we chose was not
effective in reducing the average level of apparent stress in
the study group. Yet, the burnout surveys indicated that some
individuals may be at risk, and the qualitative results indicate
that house officers evaluated in this study do experience
stressors in relation to their professional activities. Thus, the
ability to identify individuals at risk, along with the factors
that are responsible for these stressors, provide a basis for
devising specific interventions that can reverse or prevent the
manifestation of burnout.
House officers are particularly vulnerable to the effects of
stress during the transition from student to physician status,
which can influence their job performance (Feddock et al.
2007) and the development of their sense of professionalism
(West & Shanafelt 2007). Organizational factors, excessive
responsibility, and perceptions of overwork, are amongst the
most stressful factors (Paice et al. 2002). However, new
activities and responsibilities may have had similar effects on
both control and study groups (Paice et al. 2002). Martini et al,
suggest that an intervention be used to reduce the risk of
house officer burnout, such as workshops to improve coping
skills. However, the necessity for intervention must be
demonstrated to administration responsible for resource
allocation, and it is possible that a given program may
improve, but not significantly reduce, stress in the House
officer population (McCue & Sachs 1991). Alternatively, simple
didactic methods seemed at first promising (Martini et al.
2004). Our working hypothesis was that individual utilization
of a psychotherapeutic tool by house officers would reduce
the physical, emotional, behavioral and cognitive symptoms
of stress compared to a control group of house officers
who were not instructed in this intervention. In addition, we
hypothesized that utilization of the same tool would reduce
the symptoms characteristic of burnout, an increased sense of
emotional exhaustion and depersonalization and decreased
sense of personal accomplishment over a three-month period.
The Maslach survey indicated that the house officers were not
particularly stressed, on average, suggesting that the environ-
ment was not unduly stressful. Under these circumstances it
would be unlikely that a stress reduction method would be
effective in significantly reducing the level of stress leading to
burnout. Also, while our study subjects did not manifest
significant symptoms of stress or burnout and thus likely did
not have the motivation to utilize the tool, the qualitative
inventory of subjects in the study group indicated that they
already utilized elements of the BATHE technique.
While the house-officer environment may not have been
unduly stressful, the survey was an important monitoring tool
that identified individuals at risk relative to their peers
(Figure 1). A limitation of the Maslach burnout inventory is
that it does not identify factors likely responsible for burnout,
which generally arise from personal and environmental
sources (West & Shanafelt 2007). While our study findings do
not compel us to utilize the BATHE technique on a systematic
basis, it would be premature to discount its value as part of a
more comprehensive individual program designed to decrease
stress and burnout risk in health care providers. For example,
one source of personal stress that cannot be predicted
adequately, but is necessary for professional development, is
the resident’s ability to deal with patients’ suffering and death
(Meier et al. 2001). In this regard, some faculty do intervene
utilizing the BATHE technique in an attempt to train and help at
risk individuals deal with their responses. In particularly
stressful situations application of the tool by another may be
more effective than self-administration. Thus, a challenge to the
discipline of medical education may be to develop methods
identifying individuals at risk for burnout and significant stress,
along with interventions to cope.
Notes on contributors
JAY MILSTEIN, MD, Professor of Clinical Pediatrics, Medical Director
Neonatal Units, and Associate Member of the Bioethics Group, University
of California, Davis Health System. His research interests include
holistic care of infants and their families with a particular focus on
healing, physician welfare, bioethics, and fetal to neonatal cardiovascular
BONNIE RAINGRUBER, RN, PhD, Professor of Nursing at California State
University, Sacramento as well as Nurse Researcher and Associate
Adjunct Professor of Hematology and Oncology, University of California,
Davis Health System. Her research interests are in alternative treatments
and palliative care for cancer survivors and retention of health care
STEPHEN BENNETT, BA, Staff Researcher in Department of Pediatrics. His
research interests are in fetal to neonatal cardiovascular adaptation.
ALEXANDER A. KON, MD, Associate Professor of Clinical Pediatrics and
Bioethics, University of California Davis Health System. His creative work
focuses on physician-patient interactions, decision-making in pediatric
care, pediatric research ethics, and education and training in ethics and
professionalism in graduate medical education.
House Officer Burnout
CYNTHIA WINN, LCSW, Pediatric Clinical Social Worker, University of
California, Davis Health System. She has had a long-standing interest in
child abuse and neglect as well as patient and staff care issues in intensive
care settings. She has written a treatment manual for the National Center
on Child Abuse and Neglect (NCCAN) and has written about the role of
the therapist involved in child abuse and neglect.
DEBORAH PATERNITI, Associate Professor of Sociology and Deputy
Director, Center for Healthcare Policy and Research of the University of
California, Davis Health System. Her research interests include patient
expectations and decision making, health communication bioethics and
Benson H, Stuart E. (1993). The Wellness Book. Simon And Schuster.
Feddock C, Hollein A, Wilson J, Caudill T. 2007. Do pressure and fatigue
influence resident job performance. Med Teach 29:495–497.
Flahtery J, Richman J. 1993. Substance use and addiction among
medical students, residents, and physicians. Psychiatr Clin N Am
Gopal R, Glasheen J, Miyoshi T, Prochazka A. 2005. Burnout and
internal medicine resident work-hour restrictions. Arch Intern Med
Hsu K, Marshall V. 1997. Prevalence of depression and distress in a large
sample of canadian residents, interns, and fellows. Am J Psychiatr
Lindeman S, Laara E, Hakko H, Lonnqvist J. 1996. A systematic review on
gender-specific suicide mortality in medical doctors. Br J Psychiatr
Martini S, Arfken C, Churchill A, Balon R. 2004. Burnout comparison among
residents in different medical specialties. Acad Psych 28:240–242.
Maslach C, Jackson S. 1996. Maslach Burnout Inventory. Palo Alto:
Psychologist Press Inc.
McCue J, Sachs C. 1991. A stress management workshop improves
residents coping skills. Arch Intern Med 151:2273–2277.
Meier D, Black A, Morrison R. 2001. The inner life of physicians and care
of the seriously Ill. Jama 0:286.
Milstein J, Gerstenberger A, Barton S. 2002. Healing the caregiver. J Alt
Comp Med 8:917–920.
Mitchell J. 1983. When disaster strikes: The critical incident debriefing
process. Am J Emergency Services 3:6–26.
Mitchell J, Everly G. 1997. Critical Incident Stress Debriefing. Ellicott City
Md: Chevron Pub.
Paice E, Rutter H, Wetherall M, Winder B, Mcmanus I. 2002. Stressful
incidents, stress and coping strategies in the pre-registration house
officer year. Med Educ 36:566–65.
Stuart M, Lieberman J. 2002. The Fifteen Minute Hour: Practical
Therapeutic Interventions in Primary Care. Philadelphia, PA: Saunders.
Tyssen R, Vahlum P, Gronvold N, Ekeberg O. 2000. The impact of job stress
and working conditions on mental health problems among junior house
officers: A nationwide prospective cohort study. Med Educ 34:374–384.
West C, Shanafelt T. 2007. The influence of personal and environmental
factors on professionalism in medical education. Bmc Med Ed 7:29–38.
Zar H. 1984. Biostatistical Analysis. Engelwood Cliffs: Prentice-Hall.
J. M. Milstein et al.
Page 5 Download full-text