See

discussions, stats, and author profiles for this publication at: [Link]

Predictors of Failure in the Advanced Trauma

Life Support Course
ARTICLE in THE AMERICAN JOURNAL OF SURGERY MAY 2015
Impact Factor: 2.41 DOI: 10.1016/[Link].2015.03.007

DOWNLOADS

VIEWS

15

60

10 AUTHORS, INCLUDING:
Bernardino Castelo Branco

Bellal Joseph

The University of Arizona

The University of Arizona

82 PUBLICATIONS 731 CITATIONS

143 PUBLICATIONS 289 CITATIONS

SEE PROFILE

SEE PROFILE

Donald J Green

Peter Rhee

University of Southern California

The University of Arizona

44 PUBLICATIONS 188 CITATIONS

434 PUBLICATIONS 7,925 CITATIONS

SEE PROFILE

SEE PROFILE

Available from: Bernardino Castelo Branco

Retrieved on: 15 September 2015

The American Journal of Surgery (2015) -, --

Predictors of failure in the Advanced Trauma Life

Support course
Matthew Mobily, M.D.a,b, Bernardino C. Branco, M.D.a,b,
Bellal Joseph, M.D.a,b, Nancy Hernandez, R.N.c,
Richard D. Catalano, M.D.c, Daniel G. Judkins, M.P.H.a,b,
Donald J. Green, M.D.a,b, Narong Kulvatunyou, M.D.a,b,
Peter Rhee, M.D.a,b, Andrew L. Tang, M.D.a,b,*
a

Department of Surgery and bDivision of Trauma, Critical Care, Emergency Surgery and Burns,
University of Arizona, Tucson, AZ, USA; cDepartment of Surgery, Loma Linda University, Loma Linda,
CA, USA

KEYWORDS:
Advanced Trauma Life
Support participant
course;
Failure;
Predictors;
Trauma and Surgical
Critical Care;
Emergency medicine

Abstract
BACKGROUND: Over 1 million healthcare providers have participated in the Advanced Trauma Life
Support course. No studies have evaluated factors that predict course performance. This study aims to
identify these predictors.
METHODS: All participants taking the course at 2 centers over a 4-year period were identified.
Demographics, background, and performance data were extracted. Participants who failed were
compared with those who did not. Stepwise logistic regression analysis was used to identify independent risk factors for failure.
RESULTS: Seven hundred forty-four healthcare providers participated in the course; 89.5% passed
and 10.5% failed. Failure rates were lowest (.0%) among Trauma/Surgical Critical Care (SCC)
providers and highest among pediatric providers (28.6%). Stepwise logistic regression identified age
greater than 55, English as a second language, pretest score less than 75, and non-Trauma/SCC and
non-Emergency Medicine background as predictors of failure.
CONCLUSIONS: A failure rate of 10.5% was demonstrated among the course participants. Age
greater than 55, English as second language, pretest score less than 75, and non-Trauma/SCC and
non-Emergency Medicine backgrounds were associated with failure. These subgroups may benefit
from performance improvement measures.
2015 Elsevier Inc. All rights reserved.

There were no relevant financial relationships or any sources of support in the form of grants, equipment, or drugs.
The authors declare no conflicts of interest.
Presented at the Annual Meeting of the Academic Surgical Congress, February 57, 2013, New Orleans, Louisiana, USA.
* Corresponding author. Tel.: 11-520-626-5056; fax: 11-520-626-5016.
E-mail address: atang@[Link]
Manuscript received November 19, 2014; revised manuscript February 18, 2015
0002-9610/$ - see front matter 2015 Elsevier Inc. All rights reserved.
[Link]

The American Journal of Surgery, Vol -, No -, - 2015

2
Table 1

Demographic and professional background data between study groups

Age (years) . 55, n (%)

Male, n (%)
M.D. or D.O., n (%)
ESL, n (%)
Refresher course, n (%)
Pretest score, mean 6 SD
Pretest score R 75%, n (%)
Resident, n (%)
Fellow, n (%)
Attending, n (%)

Total
(n 5 744)

Failed
(n 5 78)

Passed
(n 5 666)

P value

32 (4.3)
531 (71.4)
709 (95.3)
81 (10.9)
185 (24.9)
80.5 6 11.0
486 (68.1)
345 (60.5)
12 (2.1)
213 (37.4)

7 (9.0)
53 (67.9)
71 (91.0)
25 (32.1)
8 (10.3)
73.7 6 12.4
31 (41.9)
40 (76.9)
0 (.0)
12 (23.1)

25 (3.8)
478 (71.8)
638 (95.8)
56 (8.4)
177 (26.6)
81.3 6 10.6
455 (71.1)
305 (58.9)
12 (2.3)
201 (38.8)

,.001*
.480
.068
,.001*
.002*
,.001*
,.001*
.011*
.267
.025*

The P values for categorical variables were derived from chi-square or Fishers exact tests; P values for continuous variables were derived from unpaired
Student t or MannWhitney U tests. In the analysis of failure rates according to the level of training, a total of 570 M.D. or D.O. with available data were
included in the analysis.
D.O. 5 Doctor of Osteopathy; ESL 5 English as second language; M.D. 5 Doctor of Medicine; SD 5 standard deviation.
*P values are significantly different (P , .05).

Since its introduction in 1978, the Advanced Trauma Life

Support (ATLS) course has been taught to over one million
doctors in more than 60 countries worldwide. In the United
States alone, over 500,000 healthcare providers from various
specialties and diverse medical backgrounds have taken the
course.1 The goal of this course is to provide an effective,
safe, and structured approach to management of patients
who have sustained traumatic injuries. In previous examinations, the implementation of an ATLS program has been
shown to improve trauma patient outcomes in a variety of
trauma systems both within the United States and worldwide.24 More recently, competence in ATLS has become
a core component of certain postgraduate healthcare
curricula.5 ATLS certification is now required for starting
postgraduate training among some surgical and nonsurgical
training programs including general surgery, orthopedics,
emergency medicine, family practice, and pediatrics. For
trauma care providers, the course offers a unique opportunity
to enhance academic and practical knowledge on trauma patient resuscitation, initial management, and workup.
As the pool and diversity of prospective applicants
expand, it is important to identify factors associated with
successful performance and those that may suggest an
impediment. A review of the current literature failed to
identify contemporary studies that have assessed participant
variables associated with failure in the ATLS course.
Identification of these factors may allow for preventive
strategies to maximize participant success. Hence, the
purpose of this study was to identify predictors of failure
in the ATLS course.

Patients and Methods

A retrospective review of all ATLS participant courses
sponsored by the Arizona and California states American
College of Surgeons Committee on Trauma at 2 distinct

sites (University of Arizona Medical Center [UMC],

Tucson, AZ, and Loma Linda University [LLU] Medical
Center, Loma Linda, CA) was performed. All participants
taking the ATLS course between January 1, 2007 and
December 31, 2011 were enrolled.
The ATLS course is composed of a cognitive skill test
including 40 multiple choice questions constructed by
trauma experts and a clinical skills assessment in which
2 standardized trauma patient scenarios using live patient
models are presented. These patient encounters last 15 minutes each during which the testees are judged by their
appropriateness and thoroughness to trauma workup by a
checklist system.
Data including sex, age, citizenship status, first spoken
language, academic degrees, fellowships, current employment, and course participant performance were extracted.
Failure in the ATLS participant course was defined as a
written test score less than 75% or a performance below
standards in the simulated practical section scored by an
ATLS-certified instructor. Participants who failed ATLS
course were compared with those who did not. Dichotomous variables were compared using chi-square or Fishers
exact tests, while continuous variables were compared
using unpaired Student t or MannWhitney U tests. Values
are reported as means 6 standard deviation for continuous
variables and as percentage for categorical variables.
Factors potentially associated with failure in the ATLS
course were examined for their effect using bivariate
analysis. To identify independent risk factors for failure
in the ATLS course, factors with P value less than .2 on
bivariate analysis were entered in a stepwise logistic regression. The summary data are presented as a raw percentage
or mean 6 standard deviation. The P values were significantly different at P less than .05.
All analyses were performed using the Statistical Package for Social Sciences (SPSS Mac), version 18.0 (SPSS,
Inc, Chicago, IL).

M. Mobily et al.
Table 2

Failure in the Advanced Trauma Life Support course

Predictors of failure in ATLS course

Step

Variable

Failed
ATLS (%)

r2

Adjusted OR
(95% CI)

Adjusted P

1
2
3
4

Age . 55 years
ESL
Pretest , 75%
Nontrauma and SCC/non-EM

27.3
48.1
21.0
12.4

.07
.05
.02
.01

4.6
1.6
2.3
2.1

,.001*
.002*
.010*
.007*

(1.911.1)
(1.42.7)
(1.24.1)
(1.23.6)

Variables entered in the regression are as follows: age greater than 55 years, sex, ESL, pretest less than 75%, residency level of training, and medical
specialty. A total of 544 (73.1%) subjects with complete data were included in the model.
ATLS 5 Advanced Trauma Life Support; CI 5 confidence interval; EM 5 emergency medicine; ESL 5 English as second language; OR 5 odds ratio;
SCC 5 Surgical Critical Care.
*P values are significantly different (P , .05).

Results

Comments

Over the 4-year study period, a total of 744 healthcare

providers participated in the ATLS provider course (450%
to 60.5% LLU and 294% to 39.5% UMC). There were 158
(21.2%) participants in 2007, 141 (19.0%) in 2008, 171
(23.0%) in 2009, 145 (19.5%) in 2010, and 129 (17.3%)
in 2011. The mean number of participants per course was
20.1 6 4.5 (range 16 to 24). Overall, 585 (78.6%) were
Doctor of Medicine (M.D.), 108 (14.5%) Doctor of
Osteopathy (D.O.), 14 (1.8%) Physician Assistants, and
37 (5.1%) had other or multiple degrees.
Of all participants, 89.5% (666) passed and 10.5% (78)
failed the ATLS course (71 failed the written test,
3 practical, and 4 both). There was no significant difference in failure rates between sites (LLU 9.1% vs UMC
12.5%, P 5 .130). When demographics and professional
background data were compared between both groups,
those who failed the ATLS course were more likely to
be aged above 55 years (9.0% vs 3.8%, P , .001) and
to have English as second language (32.1% vs 8.4%,
P , .001). There were no significant differences in sex
(male: 67.9% vs 71.8%, P 5 .480) or between M.D. and
D.O. degrees (91.0% vs 95.8%, P 5 .068). Those who
failed the ATLS course had lower pretest scores (73.7 6
12.4 vs 81.3 6 10.6, P , .001) and were more likely to
be at the residency level of training (76.9% vs 58.9%,
P 5 .011) (Table 1).
Among physicians, failure rates ranged from 5.6%
among attendings to 11.8% among residents (P , .001).
Failure rates were .0% among Trauma and Surgical Critical
Care (SCC) providers, 4.1% Family Medicine, 5.6% Emergency Medicine, 7.1% Internal Medicine, 10.8% General
Surgery, 24.0% Anesthesiology, and 28.6% Pediatrics.
A stepwise logistic regression identified age greater than
55 years, English as second language, pretest less than
75, and non-Trauma and SCC and non-Emergency Medicine providers as independent predictors of ATLS course
failure (Table 2). Fig. 1 depicts failure rates according to
the number of predictors for ATLS failure. Failure rates
ranged from 4.5% for patients without risk factors for
ATLS failure to 100% for those with 4 risk factors.

The assurance of high-quality trauma outcomes begins

with the provision of competent and protocolized trauma
care. More so than ever, in our current practice environment
of evolving trauma care in the face of reduced trauma
exposure during training, the importance of a structured,
intensive, trauma-specific education program has become
ever more critical. Designed by the American College of
Surgeons Committee on Trauma, the ATLS course remains
the standard teaching tool for educating healthcare providers of all levels of clinical practice involved with
treating injured patients. ATLS certification has become a
mandatory component of training for a wide variety of
surgical and nonsurgical residencies across the country.
ATLS is a 2-day course utilizing a systematic methodology composed of interactive lectures and practicums
focusing on initial patient assessment and stabilization. In
this course, 2 distinct assessment tools are employed to
evaluate clinical competency: a cognitive skill test and a
clinical skills assessment using live patient models. Failure
of the ATLS course is defined as a written test score less
than 75% or if the clinical performance is judged to be
unacceptable in the practical section.

Figure 1 Failure rates according to the number of predictors for

ATLS failure.

The American Journal of Surgery, Vol -, No -, - 2015

To date, only a single study evaluated factors that

influence participant performance in the ATLS course.
The study by Ben-Abraham et al included 4,475 physicians
participating in the Israeli training program between 1990
and 1996. Only written examinations were analyzed. The
authors demonstrated that physicians younger than 45 years
of age or with a surgical specialty were more likely to
successfully complete the ATLS course.6 In addition, the
authors also found that passage rate could be improved
by translating the reading material to Hebrew as English
language proficiency was a suspected contributing factor
in ATLS course failures.6
This study is the largest contemporary investigation on
risk factors for failure in the ATLS course. All participants
taking ATLS at 2 high-volume centers with similar
participant demographics and professional background
data over a 4-year study period were enrolled. One of the
strengths of this study is the inclusion of a wide range of
healthcare providers (78.6% were M.D., 14.5% were D.O.,
1.8% Physician Assistants, and 5.1% had other or multiple
degrees). Among physicians, emergency medicine (43.8%),
surgery (29.3%), and family medicine (8.6%) comprised
the vast majority of participants. Sixty one percent were
residents, 37.3% attendings, and 1.7% were fellows. Overall, 10.5% (78) failed ATLS, which is slightly lower than
previously demonstrated by other centers.2,7 The vast majority did so on the written test. When independent risk factors for ATLS failure were analyzed, age greater than
55 years, English as second language, pretest score less
than 75, non-Trauma and SCC and non-Emergency Medicine providers were identified as independently associated
with course failure.
There are a variety of potential explanations for these
observations. As far as age predicting ATLS failure, it
seems most likely that the results reflect the gradual loss
of knowledge that occurs because of infrequent exposure
to trauma patients, particularly when participants are
removed from an intense learning environment. This has
been well documented by Ali et al.8 They evaluated the
attrition of cognitive and trauma management skills of
60 practicing physicians pre-ATLS, immediately postATLS, at 6 months, 2 years, 4 years, and 6 years after
the course. They demonstrated that the cognitive skills
of physicians completing the ATLS course, as well as
the practical trauma management skills, undergo attrition
as measured by multiple choice and the objectively structured clinical examinations. By 6 months, only 50% of
physicians achieved the 80% pass mark. Subsequent to
this, no physician was able to maintain an 80% pass
mark in the written test.8 This issue has been demonstrated to be true for other examinations as well including
United States Medical Licensing Examination (USMLE)
and American Board of Surgery.9,10
English as a second language has also been demonstrated in other studies to impact success rates on intraining examinations.11,12 Schenarts et al found that 24%
and 19% of foreign medical graduates (FMG) required

multiple attempts to pass the USMLE Step 1 and Step 2

compared with 9% and 11% of domestic graduates, respectively. In addition, the mean USMLE Step 2 score was also
lower among the FMG (203 vs 213).
Our data on ATLS failure rates are alarming. Although
the performance of trauma and emergency medicine
providers reached expectation, almost a quarter of the
participants specialized in pediatrics, anesthesiology,
vascular surgery, oral and maxillofacial surgery, and
orthopedics failed the ATLS examination. Although the
specific reasons for these findings remain unclear, the
literature from the broader medical education community
has also investigated clinical performance measures
in various medical specialties. This has been most studied
in the field of pediatrics. In a study of pediatric senior
residents training in resuscitation, Nadel et al found
that only 11% of these trainees were able to successfully
perform vascular access procedures such as intraosseous
and femoral central line placement using the Seldinger
technique. None of the 28 third-year residents were able
to perform both basic and advanced airway maneuvers.
Of note, only 34% verified if bag-valve mask was
available and functioning before attempting endotracheal
intubation.13 Another study by Hunt et al14 evaluated
delays and errors in cardiopulmonary resuscitation and
defibrillation by pediatric residents during simulated cardiopulmonary arrests. The proportion of residents who
started 2 simple measures was assessed: (1) chest compressions within 1 minute of cardiopulmonary arrest and
(2) defibrillation within 3 minutes. Overall, 66% failed
to start chest compressions within 1 minute of pulselessness and 33% never started compressions. Furthermore,
only 54% of residents defibrillated the mannequin within
3 minutes of pulseless ventricular tachycardia. They
concluded that pediatric residents do not meet performance standards in resuscitation set by the American
Heart Association.14
Despite its large sample size and diligent data collection,
this study has several limitations. As the only centers
involved were from the southwest United States, generalizations to all healthcare providers at all training centers in
this country cannot be assumed. In addition, we were
unable to capture additional important data points which
were likely to affect failure rates among ATLS participants
including year of graduation, current practice (private vs
academic vs mixed), location of practice (rural vs urban),
teaching responsibilities, and academic productivity. Those
variables should be captured in future studies evaluating
ATLS course performance.
ATLS will remain a vital tool in training future
generations of healthcare providers in the initial assessment
and management of trauma patients. Its interactive format
rewards the participant who prepares by reviewing the
topics before arrival. The degree of precourse preparation,
as evidenced by pretest scores in this analysis, is directly
related to success rates in the actual examination. For those
at risk of failure, implementation of measures such as

M. Mobily et al.

Failure in the Advanced Trauma Life Support course

precourse review sections, computer-based on-line realtime simulations, and extended on-site practice may
improve performance rates in the actual test.
In summary, among ATLS course participants to 2 highvolume sites, a failure rate of 10.5% was demonstrated.
Age greater than 55 years, English as second language,
pretest less than 75, and non-Trauma and SCC and
non-Emergency Medicine backgrounds were independently
associated with failure. While this study has outlined the
problem of ATLS course failure, determination of solutions
can be challenging because of the specific predictors we
identified. Age, specialty, and English as a second language
are nonmodifiable factors. Of note, English as a second
language has implications as both a language barrier and in
regards to possible differences in medical training for those
physicians who are FMG. It may be feasible to provide or
direct participants toward supplementary materials in other
languages, particularly where those languages are common.
As previously noted, this intervention was shown to
improve passage rates in Israeli training programs. Additionally, pretest performance is a modifiable component.
While, to our knowledge, there is no other literature that
demonstrates how to improve performance on the ATLS
course, there is a substantial body of evidence concerning
performance improvement measures for another test
required in general surgery training programs, the American Board of Surgery In-service Training Examination.
Although the course material and test structure are different
from the ATLS course, these data may provide insight into
potential methods for improving performance. A recent
Cochrane review demonstrated that structured reading
programs and mandatory remedial programs improved the
American Board of Surgery In-service Training Examination performance.15 Extrapolation of these findings suggests that for participants who score below a certain
threshold on the pretest, additional study, for example, in
the form of individually structured reading programs or
remedial teaching sessions, may improve their performance
on the formal test. This study was not designed to identify
what the pretest cutoff should be. However, this is another
potential intervention to improve ATLS passage rates.
Finally, the findings of this study may help individual
ATLS instructors with early identification of students with
high failure potential. Prompt identification may result in
earlier intervention in the form of individualized attention
during the course and practical. Certainly, further research
is needed to determine which methods are most effective at

lowering failure rates of the ATLS course and improving

overall understanding of trauma care.

References
1. ATLS Subcommittee, American College of Surgeons Committee on
Trauma, International ATLS Working Group. Advanced trauma life
support (ATLS(R)): the ninth edition. J Trauma Acute Care Surg
2013;74:13636.
2. Aboutanos MB, Rodas EB, Aboutanos SZ, et al. Trauma education and
care in the jungle of Ecuador, where there is no advanced trauma life
support. J Trauma 2007;62:7149.
3. Blumenfeld A, Ben Abraham R, Stein M, et al. The accumulated experience of the Israeli Advanced Trauma Life Support program. J Am
Coll Surg 1997;185:812.
4. Amaraegbulam PI, Nwankwo OE. The level of knowledge of the
advanced trauma life support protocol among nonspecialist doctors
involved in trauma care in Enugu metropolis. Niger J Clin Pract
2013;16:6770.
5. Ben Abraham R, Stein M, Kluger Y, et al. The impact of advanced
trauma life support course on graduates with a non-surgical medical
background. Eur J Emerg Med 1997;4:114.
6. Ben-Abraham R, Stein M, Paret G, et al. Factors contributing to physicians success in the advanced trauma life support program in Israel.
Isr Med Assoc J 1999;1:16971.
7. Ali J, Cohen RJ, Gana TJ, et al. Effect of the advanced trauma life support program on medical students performance in simulated trauma
patient management. J Trauma 1998;44:58891.
8. Ali J, Cohen R, Adam R, et al. Attrition of cognitive and trauma management skills after the Advanced Trauma Life Support (ATLS)
course. J Trauma 1996;40:8606.
9. Malangoni MA, Jones AT, Rubright J, et al. Delay in taking the American Board of Surgery qualifying examination affects examination performance. Surgery 2012;152:73843; discussion, 7436.
10. Haist SA, Wilson JF, Elam CL, et al. The effect of gender and age on
medical school performance: an important interaction. Adv Health Sci
Educ Theor Pract 2000;5:197205.
11. Kanna B, Gu Y, Akhuetie J, et al. Predicting performance using background characteristics of international medical graduates in an innercity university-affiliated internal medicine residency training program.
BMC Med Educ 2009;9:42.
12. Schenarts PJ, Love KM, Agle SC, et al. Comparison of surgical residency applicants from U.S. medical schools with U.S.-born and
foreign-born international medical school graduates. J Surg Educ
2008;65:40612.
13. Nadel FM, Lavelle JM, Fein JA, et al. Assessing pediatric senior residents training in resuscitation: fund of knowledge, technical skills,
and perception of confidence. Pediatr Emerg Care 2000;16:736.
14. Hunt EA, Vera K, Diener-West M, et al. Delays and errors in cardiopulmonary resuscitation and defibrillation by pediatric residents during
simulated cardiopulmonary arrests. Resuscitation 2009;80:81925.
15. Kim RH, Tan TW. Interventions that affect resident performance on
the American Board of Surgery In-Training Examination. J Surg
Educ 2015;72:41829.