Content uploaded by Taylor Curley
Author content
All content in this area was uploaded by Taylor Curley on Sep 28, 2017
Content may be subject to copyright.
Modeling Junctional Tourniquet Skills from Empirical
Data
Bauchwitz, B.1, Kwan, C.2, Curley, T.3 Niehaus, J.1, Pugh, C.2, and Weyhrauch, P.1
Presented at the Military Health System Research Symposium (MHSRS), Kissimmee
FL (August 2017)
Background
The control of junctional hemorrhage, a serious life-threatening issue facing combat
personnel, has benefited from recent technological advances such as the development of
the SAM® Junctional Tourniquet (SJT) and Abdominal Aortic Junctional Tourniquet
(AAJT™). Applying these devices correctly is challenging, and existing training processes
are highly subjective, leading military medical providers to lack confidence in their ability to
use these devices. Military medical training systems that instruct on the use of these
tourniquets would benefit from an empirical model of the required skills, enabling targeted
instructional design and objective assessment. The Methodology for Annotated Skill Trees
(MAST), which has been previously described (Perez et al. 2013), enables hierarchical
decomposition of tasks with annotations such as skill descriptions, metrics, and information
requirements. Here we describe the use of MAST to create a model of the skills needed to
control inguinal bleeding using the SJT. The SJT skill model was developed based on video
evidence from a data collection with 46 EMS professionals, interviews with military medical
instructors, and instructional material produced by SAM Medical, the manufacturers of the
SJT.
Methods
To construct a MAST skill model for applying the SJT, we created a hierarchical skeleton of
the steps needed to stop inguinal bleeding with the SJT using instructional material
produced by SAM Medical. Using these and other instructional materials, we annotated
each step in the hierarchy with knowledge required to perform that step (such as landmarks
to identify the iliac artery) and decision-making rules (such as when to stop inflating the
hand pump). We then held discussions with experts from SAM Medical and the University of
Wisconsin School of Medicine to identify the critical skills at each step. Finally, we
conducted a data collection in which we filmed EMS professionals as they attempted to
apply the SJT during manikin simulations. During data analysis, participants were
categorized as experts or non-experts according to their professional experience and
familiarity with junctional tourniquets. Videos were reviewed in detail to determine the types
of errors made by the participants, and were then scored on a variety of metrics, including
the number and types of errors made as well as time to complete the simulation. Based on
the analysis of these videos, we further annotated each step in the skill model by identifying
common errors, techniques, learning curves, and metrics for assessing skill mastery, and
associating these with different levels of expertise.
Results
The result of this process was a hierarchical MAST skill model describing the tasks for
applying the SJT to stop inguinal hemorrhage. This model was annotated with references to
instructional material for training each task, decision-making rules for selecting tasks,
required knowledge and critical skills for completing each task, learning curves for the
critical skills, and metrics for assessing them.
Conclusions
We were able to model the skills for using the SJT to manage inguinal hemorrhage based
on empirical data. This skill model includes objective assessment measures to correlate
performance with expertise, enabling its use as a training tool.
1 Charles River Analytics
2 University of Wisconsin-Madison
3 Georgia Institute of Technology