Feature Article

Toward a new paradigm in hospital-based pediatric education: The

development of an onsite simulator program*
Peter H. Weinstock, MD, PhD; Liana J. Kappus, MED; Monica E. Kleinman, MD; Barry Grenier, BA, RRT;
Patricia Hickey, RN, MS, MBA; Jeffrey P. Burns, MD, MPH

Objective: The low incidence of crises in pediatrics, coupled
with logistic issues and restricted work hours for trainees, hin-
ders opportunities for frequent practice of crisis management and
teamwork skills. We hypothesized that a dedicated simulator
suite contiguous to the intensive care unit (ICU) would enhance
the frequency and breadth of critical-incident training for a range
of clinicians.
Design: Descriptive study.
Setting: A tertiary-care pediatric teaching hospital.
Measurements and Main Results: A realistic pediatric simula-
tor suite was constructed 100 feet from the ICU, at a total base
cost of $290,000. The simulation room is an exact replica of an
ICU bed space, incorporating high-fidelity mannequin simulators.
To capture an even wider audience, a portable unit was also
created. Leaders from seven departments— critical care, cardiac
intensive care, emergency medicine, transport medicine, anes-
thesia, respiratory care, and general pediatrics— completed in-
structor training to ensure effective debriefing techniques. Pedi-
atric staff, including 100% of critical care fellows, 86% of nurses,
90% of respiratory therapists, and 74% of pediatric house staff,
participated in >1500 learning encounters per year. All individ-
uals were trained during their normal workday in the hospital.
Courses in crisis resource management, skills acquisition, annual
review, orientation, and trauma management (1116, 98, 90, 60,
and 60 encounters per year, respectively) were all designed by a
multidisciplinary committee to ensure goal-directed education to
a range of audiences. Annual costs were on par with those at
other centers (approximately $44 per trainee encounter).
Conclusions: An onsite and comprehensive simulation pro-
gram can significantly increase the opportunities for clinicians
from multiple disciplines, in the course of their daily routines, to
repetitively practice responses to pediatric medical crises. After
an initial capital investment, the training appears to be cost-
effective. Hospital-based simulator suites may point the way
forward as a new paradigm for the effective education of today’s
busy clinicians. (Pediatr Crit Care Med 2005; 6:635–641)
KEY WORDS: pediatrics; critical care; simulation; volume-out-
come; education

A troublesome “paradox” con-
fronts educators in pediatric
medicine. Medical crises are
rare events in pediatrics. Al-
though this is unquestionably a positive
marker of the health of any society, par-
adoxically this reality places clinicians at
*See also p. 712.
From the Department of Anesthesia, Harvard Med-
ical School (PHW, MEK, JPB); and Departments of
Anesthesia (PHW, LJK, MEK, JPB), Respiratory Care
(BG), and Nursing (PH), Children’s Hospital Boston,
Boston, MA.
The onsite pediatric simulator center was con-
structed with financial support from the CEO of Chil-
dren’s Hospital, with additional funds provided by the
Chief of Anesthesia and philanthropic donations.
The authors have no financial interests associated
with the completion of this study.
Address requests for reprints to: Peter H. Wein-
stock, MD, PhD, MSICU offices, Pavilion 5, Boston
Children’s Hospital, 300 Longwood Ave., Boston, MA
02115.
Copyright © 2005 by the Society of Critical Care
Medicine and the World Federation of Pediatric Inten-
sive and Critical Care Societies
DOI: 10.1097/01.PCC.0000185489.07469.AF
risk of being unprepared, hesitant, and
highly anxious when such events do oc-
cur. Schoenfeld et al. (1) reviewed 80,000
admissions to the pediatric emergency
department of a major teaching hospital
and found that only 0.23% of patients
required treatment in a resuscitation
room. Reports such as this underscore
the lack of confidence most prehospital
and in-hospital personnel experience
when faced with a critically ill child or a
pediatric resuscitation (2– 4).
Strong evidence to support the apho-
rism that “practice makes perfect” has
emerged in the literature. Originally de-
scribed by Luft (5), research documents
the fact that high volume is often associ-
ated with better health outcomes (6 – 8).
This relationship appears to hold true es-
pecially for infrequent, high-risk proce-
dures (9) and has led many to argue for
regionalization of health care. However,
crisis-management and teamwork skills
are infrequent, high-risk procedures that,
by necessity, cannot be regionalized. In-
stead, all clinicians must undertake an
adequate volume of training to ensure
excellent outcomes universally (10).
Utilizing simulation to increase train-
ing in the management of a crisis has
only recently emerged as a science.
Thirty years ago Crew Resource Manage-
ment was developed in response to find-
ings that the root cause of airplane acci-
dents was poor team performance rather
than lack of expertise (11). Trainees were
videotaped during flight simulations and
performance was then evaluated through
debriefing sessions led by trained facilita-
tors (12). The program proved very suc-
cessful; crew performance and confidence
improved significantly (11, 13, 14). Ob-
serving many parallels between the in-
dustries, health care educators, most no-
tably within the departments of
anesthesia, followed suit (15) and created
analogous curricula dubbed Crisis Re-
source Management (CRM).
Over the past 10 yrs, ⬎450 simulation
centers have been built throughout the
United States and abroad (16), yet broad
exposure by clinicians remains challeng-

Pediatr Crit Care Med 2005 Vol. 6, No. 6 635

ing and limited because of logistic issues.
Inflation in health care expenditures con-
tinues to place relentless demands on cli-
nicians, resulting in limited time for ed-
ucational efforts. Off-site simulation-
learning centers impose the added
burden of travel time, further limiting
opportunities for repetitive, sequential
skill development. Morgan et al. (17) pub-
lished survey results from 60 centers and
reported lack of time and human re-
sources as the leading problem limiting
use of simulation technologies.
In an effort to capitalize on the vol-
ume-outcome relationship for pediatric
crisis training and thereby mitigate the
“pediatric training paradox,” we hypoth-
esized that a dedicated simulator suite
located on site, next to the Medical-
Surgical Intensive Care Unit (MSICU),
would provide a significant increase in
repetitive, sequential skill development
in pediatric crisis resource management
for clinicians in multiple disciplines as
part of their daily work routine.
MATERIALS AND METHODS
Development. Children’s Hospital Boston
is a 325-bed comprehensive center for pediat-
ric health care. The 18-bed MSICU, with ap-
proximately 1800 admissions annually, pro-
vides all critical care services for a wide array
of programs. To provide these services, MSICU
clinicians include a wide range of physicians
(attendings, fellows, residents, medical stu-
dents), nurses, respiratory therapists, social
workers, and support staff. The Simulation
Center for On-Site Pediatric Education
(SCOPE) at Children’s Hospital Boston was
developed as an extension of the MSICU in
2002 and was based on ground-breaking work
by the Harvard Center for Medical Simulation,
in Cambridge, MA. The conceptual foundation
for onsite simulation evolved from several ob-
servations regarding cardiopulmonary resus-
citations: they are infrequent events when
compared with the incidence in adult institu-
tions; there are rising societal expectations
that trainees have limited responsibility for
such events; and financial pressures and work-
hour limitations restrict potential concurrent
teaching opportunities by more senior clini-
cians. The onsite pediatric simulator center
was constructed with financial support from
the CEO of Children’s Hospital, with addi-
tional funds provided by the chief of anesthe-
sia and philanthropic donations.
Staffing. A full-time Simulator Suite Coor-
dinator (SSC) with a masters degree in edu-
cation was hired to provide expertise in cur-
riculum development and to oversee day-to-
day operations. A staff of ten clinician
educators in various critical care disciplines
also attended a 12-hr course on simulation-
636
based education at the Harvard Center for
Medical Simulation (funded by a Harvard Uni-
versity Risk Management Foundation Grant).
Personnel costs amounted to the annual salary
for the program coordinator; facilitators’ sal-
aries were absorbed by their individual depart-
ments.
Curriculum Development. Curricula were
developed in four phases: identification of a
departmental clinician-educator; identifica-
tion of department-specific goals; categoriza-
tion of programs; and course design, including
scenario development and execution. Depart-
mental clinician-educators worked in collabo-
ration with the SSC in developing programs in
one or more categories: CRM training, skills
acquisition, orientation programs, and annual
review. For each category, between one and
four sessions, lasting one to several hours,
were designed and tailored to the specific au-
dience. Each training session is facilitated by
the SSC and a clinician-educator and revolves
around a preprogrammed clinical scenario.
The SSC and clinician educator collaborated
with local experts in the field to design sce-
narios. To enhance realism, actual cases from
the specific clinical setting (e.g., the MSICU,
Cardiac Intensive Care Unit [CICU]) were
used, including relevant radiologic images and
laboratory data whenever possible. Scenarios
were revised as a continuous process based on
formal written feedback obtained from train-
ees. Each CRM training course begins with a
thorough introduction to simulation and the
mannequin itself. We found that trainees in-
teracted more realistically and became more
engaged with the simulator when they were
fully introduced to its capabilities before par-
ticipating in an actual scenario. To conclude,
trainees debrief with trained facilitators in the
nearby conference room, using predefined
“ground rules” such as confidentiality and
avoidance of individual critique. The core
learning principle of the debriefing is selective
use of the videotape to foster the cooperative
development of insights by the adult learners,
with the avoidance of didactic presentations by
the facilitators. CRM principles, such as effec-
tive leadership, organization, and communi-
cation, are discussed extensively (18).
RESULTS
Location and Layout. SCOPE was
constructed from an empty office space
located 100 feet from the MSICU (Fig. 1).
Although the space was relatively small,
we believed that location and proximity
were paramount considerations. In its
entirety, SCOPE is 436 square feet (sq ft),
which encompasses simulator, control,
and debriefing/conference rooms. The
simulation room, designed to the exact
specifications of an MSICU bed space, is
164 sq ft, including a 20-sq-ft storage
space, and is centered on a high-fidelity
human patient simulator—initially the
Figure 1. Location and layout of the Simulation
Center for On-Site Pediatric Education (SCOPE).
A, SCOPE is located 100 feet from the medical-
surgical intensive care unit (MSICU), on the fifth
floor of the hospital along with staff offices. The
center comprises a 164-sq-ft simulation room (B)
and a detached 240-sq-ft video conference room
that lies 145 yards away (C).
adult SimMan mannequin (Laerdal Med-
ical Corporation, Stavanger, Norway) and
more recently the automatic physiologic
PediaSim pediatric simulator (Medical
Education Technologies, METI, Sarasota,
FL). The adjacent control room is 32 sq ft
and contains computer and video periph-
erals. Lying 145 feet apart, the simulation
and conference rooms are electronically
linked for video-conferencing. Storage of
equipment is shared between the simula-
tion room and the standard storage areas
for ventilators and other equipment
within the MSICU. To further its useful-
ness, a portable component program was
included (pSCOPE), comprising a human
patient simulator, video equipment, and
necessary accessories that can be trans-
ported to various locations throughout
the hospital.
Start-Up and Operational Costs. Total
cost of initial set-up of the SCOPE
amounted to $292,000, the bulk of which
was attributed to physical site construc-
tion ($200,000; 66%), followed by equip-
ment (Table 1). With the addition of the
second mannequin ($180,000), total cost
rose to $472,000. Costs associated with
the bed space itself revolved around in-
vestments to enhance realism and in-
cluded gas and electrical outlets for med-
ical equipment, an intercom system, a
stocked code cart, intravenous equip-
ment and medications, a defibrillator,
and a vital-sign monitor identical to that
used within the MSICU. A high-definition
liquid crystal display monitor displays on-
Pediatr Crit Care Med 2005 Vol. 6, No. 6
Table 1. Start-up and Operations Costs
Start-Up
Equipment
Simulation suite construction
Human patient simulators
METI PediaSim
Laerdal SimMan
Major Equipment
Code cart, Waterloo Healthcare
Defibrillator, HP CodeMaster
Monitor, Philips MP90
Monitor display for multimedia images
Syringe pumpsa
Ventilatora
Video equipment
Wireless communication system
Code cart contents
Medications, airway, intravenous equipment
Total
Operating
Simulator maintenance (includes warranty)
Scenario supplies (central venous line/chest tube kits, etc.)
SCOPE coordinator
CO2 and nitrogen cylinders
Media (videotapes)
Office equipment and supplies
Facilitator salary
Total
a
Equipment that is borrowed and therefore free; b cost per student encounter equals cost divided
by 1,548 encounters per year.
Table 2. Time Utilization
Activity
Curriculum development: goals, scenario design/programming
Operations: tours, scenario set-up, manuscript preparation
Administrative: scheduling, ordering supplies, accounting
Courses
Crisis Resource Management (CRM)
Annual Review
Orientation
Critical Skills
Trauma
demand radiographs, echocardiograms,
and real-time laboratory results. Given
SCOPE’s location, ventilators, echocar-
diographic machines, and intravenous
pumps were easily borrowed at no extra
cost. Three video cameras were installed
and attached to editing equipment within
the control room. The majority of opera-
tion costs were attributed to the SSC
salary ($57,500; 85%), followed by main-
tenance of the human patient simulator
($7,000; 10%). Average cost per student
or learning encounter was $44. This was
calculated by dividing total annual cost
by 1548 encounters per year (see Usage
Demographics below).
Time Utilization. Thirteen hours per
Pediatr Crit Care Med 2005 Vol. 6, No. 6
per simulation) were calculated for com-
parison. As shown in Table 3, the major-
Cost ($)
ity of usage was by the MSICU (50%; 108
participant-hours per month), followed
200,000 by CICU (18%; 40); general pediatrics
(18%; 40); respiratory therapy (5%; 11.5);
180,000 emergency medicine (5%; 10); anesthesia
30,000 (3%; 6); and transport medicine (1%; 2).
1,200 A wide range of specialists came from
7,800 multiple departments throughout the
30,000 hospital to actively participate in onsite
800 simulation. Table 3 shows the distribu-
0
0
tion of individuals trained per month.
20,000 Again, participant-hours per month were
1,400 calculated for each group and were then
compared with the total. Trainees came
800 from four groups; most were registered
472,000
nurses (44%), followed by residents
(30%), fellows (21%), and respiratory
Annual Per Student Encounterb
therapists (5%). Each participant trained
7,000 4.52 in one or more courses. A total of 129
1,400 0.90 learning encounters occurred at SCOPE
57,500 37.00 each month (1548 encounters/yr). Forty-
1,200 0.78 five nurses trained each month. The ma-
275 0.18
500 0.32 jority came from the MSICU (50%) and
0 0 CICU (33%), with the remainder from the
67,875 43.70 wards (13%), transport, and emergency
department. Thirty-seven fellows from
five departments were trained each
month (444 encounters/yr), over half
(53%) of whom came from the MISCU.
Thirty-seven pediatric residents of both
Hrs/wk the MSICU and general pediatric ward
services participated each month (56%
8 and 44%, respectively). Ten respiratory
11 therapists underwent simulation-based
8
13
training each month within the setting of
5 a variety of courses.
3 Yearly participation rates by specialty
2 were also analyzed. From September
2.6 2003 to September 2004, the vast major-
0.1
Total 40 ity of MSICU staff (100% of critical care
fellows [n ⫽ 12]; 86% of critical care
nurses [n ⫽ 94]; 90% of respiratory ther-
apists [n ⫽ 44]) and pediatric residents
week (33%, 1.6 days/wk) are spent ac- (70%; n ⫽ 48) received simulation-based
tively engaged in course/educational ac- training in one or more courses. Of 129
tivities (Table 2). Another large portion of trainees in the Boston Combined Resi-
time (11 hrs/wk; 28%) is spent in opera- dency Program, 96 (74%) received CRM
tional activities, particularly setting up training during either their MSICU or
and preparing for each scenario. ward rotations at some point during the
Usage Demographics. Usage demo- 2003–2004 academic year.
graphics were analyzed by department as SCOPE Courses. Due to the location of
well as by participant type (e.g., nurse, SCOPE we were able, within a relatively
fellow, resident). Seven departments— short period of time, to fully develop 13
critical care (MSICU), CICU, anesthesia, courses within five different main catego-
emergency medicine, transport medicine, ries: CRM, trauma, skills acquisition, ori-
respiratory care, and general pediatrics— entation, and annual review. Table 4 de-
participate in SCOPE programs. For each scribes each course in detail. The variety
department, participant-hours per month of courses included pediatric health care
(total number of participants per month staff from multiple departments. As in-
⫻ number of simulations ⫻ hours spent tended, the majority of SCOPE time is
637
Table 3. Usage Demographics for the Simulation Center for On-Site Pediatric Education (SCOPE), by
Department and Participant Group
Departmenta
MSICU
CICU
General pediatrics
Respiratory care
Emergency medicine
Anesthesia
Transport medicine
Participantsb No./Mo Participant Hrs/Mo (%)
Fellows 37 46 (21)
Residents 37 64 (30)
RNs 45 96 (44)
RTs 10 12 (5)
Total 129 218
a
Seven departments participate: medical-surgical intensive care unit (MSICU), cardiac intensive
care unit (CICU), general pediatrics (GP), respiratory therapy (RT), emergency medicine (EM),
anesthesia, and transport medicine (TM). Half of SCOPE’s usage is devoted to training of MSICU staff.
Values are reported as participant hours per month (see text); b trainees come from four groups: fellow,
residents, registered nurses (RNs), and respiratory therapists (RTs).
devoted to CRM (59%), followed by An-
nual Review for nurses (17%). Within de-
briefings, incorporating relevant re-
sources reinforces key topics. For
instance, “fixation and failure of adequate
leadership” are demonstrated during the
debriefing by the screening of a landmark
NOVA episode entitled Why Planes Crash
(original broadcast date, January 27,
1987), which highlights the crash of
Eastern Flight 401 in December, 1972.
CRM courses were developed for mul-
tidisciplinary teams and also tailored for
individual departments such as respira-
tory therapy. General pediatric staff
(8 –12 residents and four nurses) undergo
CRM training for 1 hr/wk. The unique
course was motivated by the Children’s
Hospital CPR Committee, which identi-
fied a need for training specifically for
events that occur before the arrival of the
formal code team at a resuscitation.
Learning objectives therefore focus on
improvements in team performance dur-
ing the first 3 to 5 mins of a code, before
help arrives. pSCOPE is transported to
the emergency department, where trau-
ma-based CRM occurs bimonthly. With
use of airway adjuvants and fiberoptics
from the nearby operating rooms, skills-
638
routines, to practice responses to pediat-
ric crises that would otherwise not be
attainable. The center is contiguous with
Participant Hrs/Mo (%)
our MSICU and was built at a base cost
108 (50) lower than at many other reported cen-
40 (18) ters. The success of such a program
40 (18) stems from a multidisciplinary commit-
11.5 (5) tee whose leaders were trained in simu-
10 (5)
6 (3)
lation/debriefing. Multiple courses were
2 (1) designed and tailored to the needs of in-
dividual departments. Despite its rela-
Department Participant Hrs/Mo (%) tively small size, the SCOPE’s location
within the hospital leads to increased
MSICU 24 (53) scheduling flexibility, minimal interrup-
CICU 8 (17) tion of patient care, and, consequently,
EM 8 (16)
high participation rates among all depart-
Surgery 1 (1)
Anesthesia 5 (13) ment/divisions. Utilization of existing
MSICU 36 (56) hospital space and borrowed equipment
Ward 28 (44) results in reduced operational costs and
MSICU 48 (50) improved participant numbers.
CICU 32 (33)
Ward 12 (13) Although the simulation program
TM 2 (2) start-up costs we report are on a par with
EM 2 (2) those reported by others, our per-trainee
RT 12 (100) expenses appear to be more cost-effective.
Kurrek et al. reported on an onsite Cana-
dian Simulation Center similar to ours
(3) with total construction costs of ap-
proximately $800,000 (19). Loyd reported
on a 5000-sq-ft center at the University of
Louisville with a cost of $2 million. The
center anticipated providing approxi-
mately 4500 “learning encounters” each
acquisition programs have been produced year. Our center cost approximately 75%
for anesthesia fellows, residents, and less, yet we conducted ⬎1500 such en-
transport team nurses. counters, in large part because of acces-
Orientation courses have been devel- sibility. In a comparison of both centers
oped for both MSICU and CICU nurses. in their first year, cost per encounter at
Since September 2003 all new hires (n ⫽ our onsite facility was equal to or less
21) in the MSICU have completed simu- than that at the University of Louisville
lator training. The orientation program ($330 vs. $440 per encounter). The
for CICU nurses includes cardiac-specific WISER Institute at the University of
skills and, to date, has trained all new Pittsburgh, one of the largest simulation
hires (n ⫽ 11). As part of mandatory centers in the world, occupies approxi-
continuing education for MSICU nurses, mately 11,000 sq ft and was constructed
annual review sessions occur weekly. at a cost in the millions. WISER reported
Three-hour sessions focus on competen- 8000 learning experiences in the 2003–
cies such as assistance with procedural 2004 fiscal year (20). With adjustments
sedation and effective communication for size, SCOPE trains more individuals
“up the chain of command.” To date, per sq ft (3.55 vs. 0.7). These comparisons
nearly all critical care nurses (76/81; further underscore the effectiveness of
86%), excluding new hires, have com- onsite simulation.
pleted this training. One notable finding in our study is the
large number of clinicians, from a wide
DISCUSSION range of disciplines, we were able to pro-
vide with simulation training in 1 yr.
To our knowledge, this is the first pub- Before SCOPE, we were able to train be-
lished report of an onsite, high-fidelity tween five and eight critical care fellows
simulation suite at a pediatric teaching per year at an off-site facility only 15 mins
hospital. Our findings indicate that onsite from our campus. Logistic issues were
simulation has significantly increased the significant and mostly related to patient
opportunities for clinicians in multiple care responsibilities associated with
disciplines, in the course of their daily transporting fellows away from the unit.
Pediatr Crit Care Med 2005 Vol. 6, No. 6
Table 4. Simulation Center for On-Site Pediatric Education (SCOPE) Courses

Monthly Sessions, No.

Course (No. of Encounters/Yr) Audience (Duration of Each) Participants (No.), Content

CRM: Crisis Resource Management MSICU
(1116; 59%)
Annual Review (98; 17%)
Skills Acquisition (90; 11%)
Orientation (60; 9%)
a
Course utilizes pSCOPE (portable-component program) and is taught within the actual clinical environments. MSICU, medical-surgical intensive care
unit; CICU, cardiac intensive care unit; Gen Peds, general pediatrics house staff; RT, respiratory therapists; EM, emergency medicine; HPS, human patient
stimulator; ECMO, extracorporeal membrane oxygenation.
4 (1 hr) Pediatric residents (5), critical care fellows (6), critical care nurses (2),
and a respiratory therapist (1)
Wk 1: Review of CRM principles, orientation to HPS
Wk 2: Medical hierarchy, communication, utilization of resources
Wk 3: Effective response to a suspected inappropriate plan of care
Wk 4: Global assessment, “fixation,” and responsibilities of nonleader
participants
CICUa 1 (4 hrs) Critical care nurses (4), cardiology fellows (2), respiratory therapist (1)
and ECMO specialists (1)
Gen Peds 4 (1 hr) General pediatric residents (7) and nurses (3)
Improvements in team performance during the first 3–5 mins of a code,
before help arrives
RT 1 (4 hrs) Respiratory therapists (2)
Crises involving artificial airway emergencies, high-frequency
ventilation; veteran therapists participate in scenarios as an education
resource
Transport Team 1 (1 hr) Transport nurses (3) and paramedics (3)
EM 1 (1 hr) Emergency medicine fellows (1 or 2)
MSICU 4 (3 hrs) Critical care nurses (2–4)
Competencies, including assistance with procedural sedation, use of
rapid infusers, identification and management of transfusion
reactions, transport of critically ill patients, defibrillation technique,
and effective communication “up the chain of command”
MSICU 4 (1 hr) Pediatric residents (4 or 5)
Use of difficult airway equipment, central venous and intraosseous lines,
chest tubes, defibrillation, and transcutaneous pacing within the
context of simulated clinical scenarios
Anesthesia 2 (1 hr) Anesthesia fellows (1)
One-on-one technical skills tutorial in difficult airway management,
instructed by an expert anesthesiologist at Harvard Medical School
Transport 1 (1 hr) Transport nurses (2 or 3) and paramedics (2 or 3)
MSICU 6 (1 hr) Critical care nurses (1 or 2)
Insertion and management of invasive lines, troubleshooting monitors,
reading electrocardiograms, preparing for intubations, performing
assessments, preparing medications, defibrillation technique,
addressing airway emergencies, and practicing various roles during
code situations
CICU 4 (1 hr) Critical care nurses (1–4)
As above, with addition of cardiac-specific skills (e.g., treatment of
arrhythmias, pulmonary hypertension)

The same was true for other departments.
Teamwork training involving physicians,
nurses, and respiratory therapists was se-
verely limited. Orientation and skills
training for nurses and other groups oc-
curred primarily as didactic experiences
out of clinical context. Historically, the
hardest population to reproducibly train
is house staff because of their highly vari-
able schedules and new work-hour re-
strictions. With the availability of onsite
simulation, we have provided training op-
portunities for ⬎70% of residents
through our program. Despite the large
number of centers worldwide, usage de-
mographic data remain sparse, a circum-
stance making comparisons difficult.
Lighthall et al. (21) reported on an inter-
disciplinary CRM course that trained 181
participants over a 3-yr period (60/yr).
Taught in an off-site center, the course
was well very received by medical and
surgical house staff as well as ICU nurses
and respiratory therapists. Attendings
and fellows provided clinical coverage
during the sessions. By comparison, de-
spite its relatively small size, SCOPE
trained ⬎1,100 encounters/yr in CRM
alone.
Our findings indicate that, in addition
to the benefits of an onsite location, the
high utilization of the program was due
to an authentic and challenging curricu-
lum that allowed clinicians to suspend
disbelief in the simulation exercise. Fur-
thermore, we strenuously avoided situa-
tions in which participants might feel
that they were being humiliated or en-
trapped by arcane or rare scenarios. We
realized early on that proximity alone
would not be enough to attract busy cli-
nicians and that well-designed course of-
ferings must be made available (22). Cur-
rently, the simulation literature is
devoted primarily to CRM training alone.
We report here on the development of
five individual courses, ranging from
CRM to orientation programs, taught
within a single center. Specialists were
able to easily “step away” and engage in
simulation without disrupting their daily
work routines. In particular, 1-hr prob-
lem-solving sessions for respiratory ther-
apists and nurses were met with great

Pediatr Crit Care Med 2005 Vol. 6, No. 6 639

enthusiasm. The pSCOPE brought simu-
lation “on demand” to various sites
throughout the hospital, allowing us to
provide courses for busy groups by bring-
ing the center to them.
Our study has several limitations. Al-
though onsite simulation promises to
train large numbers of individuals, as
compared with off-site training, potential
disadvantages do exist, supporting the
use of both approaches. Off-site simula-
tion often occurs over a full day, allowing
more time for both simulation and de-
briefing. We have begun to address this
by having all debriefings run by trained
facilitators in order to provide for the
most educational and therapeutic use of
time. When the inverse is considered, one
benefit of more frequent albeit shorter
sessions is time for trainees to reflect on
performance during the days between ex-
ercises. Some claim traveling off-site pro-
motes less interruption by beepers and
pagers. Although this finding has been
primarily anecdotal, we have addressed
this potential limitation by providing cov-
erage for all involved trainees. Clinical
coverage for onsite simulation is simpli-
fied by the absence of travel and by the
relatively short times that any individual
is away from a post.
Whether simulation actually improves
outcomes in a crisis has been difficult to
measure (23). Nevertheless, despite the
inability to specifically measure skill im-
provements in pilots who undergo yearly
crew-resource management, the training
technique has been fully embraced by
most airline carriers (11). Similarly, de-
spite a paucity of data validating the effect
of medical simulation on outcomes (24 –
27), a rapidly growing number of educa-
tors believe in the power of simulation as
a pedagogical tool (28), supported by
qualitative analyses in which trainees de-
scribe simulation as enjoyable and re-
warding (21, 29 –31). Mayo et al. (32)
recently reported on both the short- and
long-term benefits of simulation-based
teaching of airway skills.
Cost analysis, such as that described
here, is isolated to set-up and operational
costs but does not address the more in-
teresting question of the cost-utility ratio
of simulation. Such analyses, even with
well-designed observational studies, will
be inherently difficult for pediatric crises
in which events are infrequent and the
confluence of important variables makes
measurement more challenging. Another
important limitation relates to the con-
ceptual underpinning of simulation itself,
640
that “practice makes perfect.” Critics of
the volume-outcome relationship note
that little is known about whether, for a
given procedure or condition, a volume
threshold exists above which outcomes
are more favorable but do not continue to
improve with further increases in volume
(33). In our center, studies are currently
under way in which onsite simulation is
used to determine the effectiveness of
high-volume simulation training on both
clinician preparedness and confidence as
well as on the delivery of care at the
bedside.
CONCLUSIONS
In summary, we have reported a major
improvement in the volume of critical
incident training via onsite simulation
within a pediatric teaching hospital, ac-
companied by detailed demographic and
cost analyses. Many of the challenges
faced by the medical profession today—
disaster response, decreasing societal tol-
erance for trainee participation in patient
care, and work-hour restrictions that de-
crease opportunities for learning—will
best be addressed by adopting a new par-
adigm in hospital-based education. In
keeping with the insights of the recent
roadmap initiative of the National Insti-
tutes of Health (34), hospital-based sim-
ulator suites may more readily enhance
multidisciplinary collaboration in patient
care through easily accessible, repetitive,
sequential team training. Hospital-based
simulator suites, such as the one de-
scribed here, may point the way forward
as the new paradigm for the effective ed-
ucation of clinicians, current and future.
REFERENCES
1. Schoenfeld PS, Baker MD: Management of
cardiopulmonary and trauma resuscitation
in the pediatric emergency department. Pe-
diatrics 1993; 91:726 –729
2. Nadel FM, Lavelle JM, Fein JA, et al: Assess-
ing pediatric senior residents’ training in re-
suscitation: Fund of knowledge, technical
skills, and perception of confidence. Pediatr
Emerg Care 2000; 16:73–76
3. Kurrek MM, Devitt JH: The cost for construc-
tion and operation of a simulation centre.
Can J Anaesth 1997; 44:1191–1195
4. McQuillan P, Pilkington S, Allan A, et al:
Confidential inquiry into quality of care be-
fore admission to intensive care. BMJ 1998;
316:1853–1858
5. Luft HS, Bunker JP, Enthoven AC: Should
operations be regionalized? The empirical re-
lation between surgical volume and mortal-
ity. N Engl J Med 1979; 301:1364 –1369
6. Hannan EL, Racz M, Kavey RE, et al: Pedi-
atric cardiac surgery: The effect of hospital
and surgeon volume on in-hospital mortal-
ity. Pediatrics 1998; 101:963–969
7. Hannan EL, Popp AJ, Tranmer B, et al: Re-
lationship between provider volume and
mortality for carotid endarterectomies in
New York State. Stroke 1998; 29:2292–2297
8. Hannan EL: Measuring hospital outcomes:
Don’t make perfect the enemy of good!
J Health Serv Res Policy 1998; 3:67– 69
9. Begg CB, Cramer LD, Hoskins WJ, et al:
Impact of hospital volume on operative mor-
tality for major cancer surgery. JAMA 1998;
280:1747–1751
10. Rogers PL, Grenvik A, Willenkin RL: Teach-
ing medical students complex cognitive skills
in the intensive care unit. Crit Care Med
1995; 23:575–581
11. Billings C, Reynard W: Human factors in
aircraft incidents: Results of a 7-year study.
Aviat Space Environ Med 1984; 55:960 –965
12. Helmreich RL, Foushee HC, Benson R, et al:
Cockpit resource management: Exploring
the attitude-performance linkage. Aviat
Space Environ Med 1986; 57:1198 –1200
13. Helmreich RL, Wilhelm JA, Gregorich SE, et
al: Preliminary results from the evaluation of
cockpit resource management training: Per-
formance ratings of flight crews. Aviat Space
Environ Med 1990; 61:576 –579
14. Helmreich RL: Does CRM training work? Air
Line Pilot 1991; 60:17–20
15. Gaba D, Fish K, Howard SK: Crisis Manage-
ment in Anesthesiology. New York, Churchill
Livingstone, 1994
16. McIndoe A, Jones A: Simulation Centers Es-
tablished World Wide 1994 –2004. Graph.
Available online at http://www.bris.ac.uk/
Depts/BMSC (go to links/geographical-
breakdown/graph). Accessed December 9,
2004
17. Morgan PJ, Cleave-Hogg D: A worldwide sur-
vey of the use of simulation in anesthesia.
Can J Anaesth 2002; 49:659 – 662
18. Mort TC, Donahue SP: Debriefing: The Ba-
sics. In: Simulators in Critical Care Educa-
tion and Beyond. Dunn WF (Ed). Des Plaines,
IL, Society of Critical Care Medicine, 2004,
pp 76 – 83
19. Morgan PJ, Cleave-Hogg DM: Cost and re-
source implications of undergraduate simu-
lator-based education. Can J Anaesth 2001;
48:827– 828
20. Grenvik A, Brindis C, Schaefer J: The Peter
M. Winter Institute for Simulation, Educa-
tion and Research FY 2003–2004 Annual Re-
port. Available online at: http://www.wiser.
pitt.edu/aboutus/annualreports.htm. Ac-
cessed December 12, 2004
21. Lighthall GK, Barr J, Howard SK, et al: Use of
a fully simulated intensive care unit environ-
ment for critical event management training
for internal medicine residents. Crit Care
Med 2003; 31:2437–2443
22. Kyle RR: Technological resources for clinical
simulation. In: Simulators in Critical Care
Pediatr Crit Care Med 2005 Vol. 6, No. 6
 Education and Beyond. Dunn WF (Ed). Des
Plaines, IL, Society for Critical Care Medi-
cine, 2004, pp 95–113
23. Cavanaugh S: Computerized simulation
technology for clinical teaching and testing.
Acad Emerg Med 1997; 4:939 –943
24. Morgan PJ, Cleave-Hogg DM, Guest CB, et al:
Validity and reliability of undergraduate per-
formance assessments in an anesthesia sim-
ulator. Can J Anaesth 2001; 48:225–233
25. Morgan PJ, Cleave-Hogg D, McIlroy J, et al:
Simulation technology: A comparison of ex-
periential and visual learning for undergrad-
uate medical students. Anesthesiology 2002;
96:10 –16
26. Morgan PJ, Cleave-Hogg D: Evaluation of
medical students’ performance using the an-
aesthesia simulator. Med Educ 2000; 34:
42– 45
27. Devitt JH, Kurrek MM, Cohen MM, et al:
Pediatr Crit Care Med 2005 Vol. 6, No. 6
Testing internal consistency and construct
validity during evaluation of performance in
a patient simulator. Anesth Analg 1998; 86:
1160 –1164
28. Gordon JA, Wilkerson WM, Shaffer DW, et al:
“Practicing” medicine without risk: Stu-
dents’ and educators’ responses to high-
fidelity patient simulation. Acad Med 2001;
76:469 – 472
29. O’Brien G, Haughton A, Flanagan B: Interns’
perceptions of performance and confidence
in participating in and managing simulated
and real cardiac arrest situations. Med Teach
2001; 23:389 –395
30. Howard S, Gaba D, Fish K, et al: Anesthesia
crisis resource management training: Teach-
ing anesthesiologists to handle critical inci-
dents. Aviat Space Environ Med 1992; 63:
763–770
31. Hammond J, Bermann M, Chen B, Kushins
L: Incorporation of a computerized human
patient simulator in critical care training: A
preliminary report. J Trauma 2002; 53:
1064 –1067
32. Mayo PH, Hackney JE, Mueck JT, et al:
Achieving house staff competence in emer-
gency airway management: Results of a
teaching program using a computerized pa-
tient simulator. Crit Care Med 2004; 32:
2422–2427
33. Hewett M: Interpreting the Volume-Out-
come Relationship in the Context of Health
Care Quality: Workshop Summary (2000).
Washington, DC, Institute of Medicine, 2000
34. National Institutes of Health. NIH Roadmap
Initiatives: Theme: New Pathways to Discov-
ery. Available online at http://nihroadmap.
nih.gov/initiatives.asp. Accessed January 27,
2005
641