Page 54 - JSOM Winter 2019

a manikin with three tourniquet models of the individual’s FIGURE 1 A Step Wheel: The test steps as ordered.
maximal, moderate, or minimal familiarity. In this framing,
any intermodel difference in performance was seen through
a lens of interoperability to assess whether the difference is
beneficial or costly as performance is improved or impaired.
The interval from study conception to the beginning of data
collection was a 1-hour period during which the test setup and
data collection sheet were made. This hour’s work prohibited
preparations like mental rehearsals.

All three models—Combat Application Tourniquet (C-A-T;
C-A-T Resources Inc., http://combattourniquet.com/, gen-
eration 7), Special Operations Forces Tactical Tourniquet
(SOFTT, Tactical Medical Solutions, https://www.tacmedsolu-
tions.com/, wide version), and Military Emergency Tourniquet
(MET, Tier-One Quality Solutions, http://www.tqsresponse.
com/)—had a band-and-rod design. Among them, the user’s
familiarity with C-A-T was maximal, whereas familiarity with
SOFTT and MET was moderate and minimal, respectively. In
all, the study had a total of 30 tests with 10 tests for each
model. The test order was randomized by model, and one de-
vice was used per model.

By removing wrappers or bags, each tourniquet was arranged
in its ready-to-go configuration as shipped. The paper instruc-
tions as a product insert accompanied each tourniquet were
integral to the configuration as routinely stowed in a first-aid
kit. The configurations of SOFTT and MET included the use
of rubber bands. The diagram is a clockwise depiction of step order in a test. The test
was designed to assess performance of a basic set of skills needed in
For the setup, three tourniquets were laid on a table next to first aid. The development of the diagram was made to label, describe,
the manikin. A test began with the user starting a timer (Fig- categorize, and graph concepts. Time segmentation was portrayed
ure 1), starting the manikin, picking up the model, removing graphically to aid in understanding the steps (named, listed, ordered,
and associated) and the categories of concepts such as the burden of
items (instructions and rubber bands), opening the loop in the overall time (test duration [step 15 minus step 1]), clinical effect (time
band, passing the loop around the end of the limb, placing the to stop bleeding [occurs during step 9 after starting in step 2]), and
tourniquet on the limb, pulling slack out of the band, turning post control steps (assessing control [during steps 9 and 10]), securing
the rod, securing the rod and band, troubleshooting problems, the rod and band [troubleshooting was associated with securing prob-
stopping the manikin, removing the tourniquet, configuring lems], and removing and configuring a tourniquet). The depiction of
the steps in order eases the perception of the sequence for learners, aids
the tourniquet, stopping the timer, and recording data and in naming each step for easy communication among coworkers, and
notes of performance. The manikin collected data during its invites intermodel comparisons and contrasts by step. For example,
operation, which was called a trial. A test encompassed a trial among the three models, only one has to have both its rod ends unse-
plus its preceding interval, which began at the smartphone cured in removal while the others only have one. The wheel is to be a
common operating method to help all be on the same sheet of music.
timer plus a following interval that ended with the timer.
The manikin was previously described elsewhere. Briefly, a times required to teach 10 people could be proportional to
9
HapMed Leg Tourniquet Trainer (CHI Systems, http://www. the sums of test durations, and the sum estimates a burden of
chisystems.com) simulated a limb amputation. The trial sce- training 10 people in man-minutes.
nario included a single adult with a medium build, standard
first aid with tourniquet placement 2–3 inches above the Ease of use was on a 5-point Likert scale, with 1 being very
wound, and a bleed-to-death time of 120 seconds (Figure 2). difficult; 2, difficult; 3, neutral; 4, easy; and 5, very easy. De-
scriptive statistics were used to analyze the data and present
The manikin measured total trial time and time to stop bleed- results. Analyses were performed using JMP (v13.1.0, SAS
ing. The manikin determined patient status (bleeding, stable, Corporation, Cary, NC) and Excel 2003 (Microsoft, www.mi-
or dead), trial status (go [satisfactory] or no go), tourniquet crosoft.com). While one-way analyses of data were performed
placement, pressure, and blood loss. Data collected by the user by nonparametric comparison of paired means (± standard
included ease of use and times (start test, end test). Calcula- deviation) using Wilcoxon method, contingency table analyses
tions included test duration (test end minus test start), post were used for categorical data.
control time (total trial time minus time to stop bleeding), and
time of removal and configuration (test duration minus total Results
trial time).
Timesaving Strongly Favored the Model With Which the
The costs or benefits of interoperability were measured in units User Had Maximal Familiarity
of time, ease, blood, and pressure. Summing of cost or benefit In comparison of interoperability burdens, sums of test dura-
by model was made by summing its 10 tests. For example, tions by tourniquet model for C-A-T, SOFTT, and MET were

52 | JSOM Volume 19, Edition 4 / Winter 2019