Page 27 - Journal of Special Operations Medicine - Summer 2016

Preliminary Measures of Instructor Learning

in Teaching Junctional Tourniquet Users

John F. Kragh Jr, MD; James K. Aden 3rd, PhD;
Stacy A. Shackelford, MD; Michael A. Dubick, PhD

ABSTRACT

Background: The objective of the present study was to of data available after completion of another study. We
1
assess the effect of instructor learning on student perfor- used a manikin (Combat Ready Clamp Trainer Manikin;
mance in use of junctional tourniquets. Methods: From Operative Experience, Inc., http://operativeexperience
a convenience sample of data available after another .com) that allowed for assessment of bleeding control
study, we used a manikin for assessment of control of from a right groin gunshot wound. Blood loss was mea-
bleeding from a right groin gunshot wound. Blood loss sured by the instructor while training users of junctional
was measured by the instructor while training users. tourniquet models. The capacity of the one instructor
The data set represented a group of 30 persons taught to affect the performance of users was a surrogate of
one at a time. The first measure was a plot of mean instructor learning.
blood loss volumes for the sequential users. The second
measure was a plot of the cumulative sum (CUSUM) of The first measure was a plot of mean blood loss volumes
mean blood loss (BL) volumes for users. Results: Mean for the sequential users. The second measure was to plot
blood loss trended down as the instructor gained experi- a cumulative sum (CUSUM) of mean blood loss volumes
ence with each newly instructed user. User performance for users. CUSUM is a way to detect changes in trends
continually improved as the instructor gained more ex- and it looks at the sequential change in the cumulative
perience with teaching. No plateau effect was observed sum of data plotted by the sequence (i.e., the order of
within the 30 users. The CUSUM plot illustrated a turn- the data points acquired); the details are described in
ing point or cusp at the seventh user. The prior portion Equation 1.
of the plot (users 1–7) had the greatest improvement;
performance did not improve as much thereafter. The
improvement after the seventh user was the only change Equation 1 Description of the Cumulative Sum (CUSUM)
detected in the instructor’s trend of performance. Con- Technique Used
clusions: The instructor’s teaching experience appeared k
to directly affect user performance; in a model of junc- CUSUM (k) = ∑ (BL(i) – BL)
tional hemorrhage, the volume of blood loss from the
manikin during junctional tourniquet placement was a i=1
useful metric of instructor learning. The CUSUM tech- CUSUM entails summing a data sequence (∑ = summation). The cu-
nique detected a small but meaningful change in trend mulative difference of each data point from the benchmark value,
where the instructor learning curve was greatest while in this case the mean value (BL) of the data set, is typically used in
working with the first seven users. monitoring change detection (e.g., a step in a time series). The dif-
ference is calculated between each measurement (user mean) and the
overall mean (mean of all users), and the differences are cumulatively
Keywords: first aid; damage control; hemorrhage, preven- summed. The summing begins with the initial (i) data point and pro-
tion and control; shock; tourniquet; resuscitation; emer- ceeds incrementally through each successive data point (called the kth
value). If learning is absent, measurements do not deviate significantly
gency medical services from the mean, so measurements greater than the mean and those
less than the mean roughly average each other out, and the CUSUM
value would vary narrowly around the mean. If learning was pres-
ent, measurements would trend away from the mean. CUSUM is an
established technique that is useful in statistical monitoring to detect
he objective of this study was to assess the effect of changes in trends of sequential data, such as in learning or quality
Tinstructor learning on student performance in use of control. CUSUM is typically portrayed in plots indicating if or where
small changes are detected visually. Although CUSUM is uncommonly
junctional tourniquets. We used a convenience sample used in medicine, some investigators advocate for its increased use.

13