Page 42 - JSOM Summer 2018

TABLE 1 Six Performance Metrics, Which Are Composite Results of Categorical Data
Metric A B C D E F
Metric component 1 Trial Patient Ease of Use Bleeding Tourniquet Tourniquet
Status: Status: Score ≥3 Control: Placement: Pressure:
Go A Go B (Neutral or Easier) Yes Good Good
Metric component 2: No No No No No No
Tourniquet damage
Metric component 3: No No No No No No
Manikin damage
Metric component 4:
User symptoms No No No No No No
No. of C-A-T tests 73 76 77 77 77 73
No. of SXT tests 58 58 58 59 60 58
No. of tests favoring C-A-T 13 18 19 18 17 15
A Trial status go was when the manikin determined that the performance was satisfactory. Patient status go was when the manikin determined
B
that the patient was stable (neither bleeding nor dead). C-A-T, Combat Application Tourniquet; SXT, SAM Extremity Tourniquet.
TABLE 2 Five Performance Metrics, Which Are Composite Results TABLE 3 Five Metrics, Which Are Composite Combinations of
of Continuous Data Categorical and Continuous Data
Metric F G H I J Metric K L M N O
Metric component 1: Metric component 1:
Time to determination of ≤30 ≤60 ≤90 ≤120 ≤120 Patient status Stable Stable Stable Stable Stable
bleeding control, seconds Metric component 2:
Metric component 2: ≤60 ≤90 ≤120 ≤150 ≤180 Time to determination of ≤30 ≤60 ≤90 ≤120 ≤120
Total trial time, seconds bleeding control, seconds
Metric component 3: ≤800 ≤800 ≤800 ≤800 ≤800 Metric component 3: ≤60 ≤120 ≤150 ≤180 ≤180
Blood loss volume, mL Total trial time, seconds
Metric component 4: Metric component 4:
Bleeding rate, mL/s ≤9.4 ≤9.4 ≤9.4 ≤9.4 ≤9.4 Ease of use (1, very difficult; ≥4 ≥4 ≥4 ≥4 ≥3
Metric component 5: 2, difficult; 3, neutral; 4,
Pressure, mmHg ≥250 ≥250 ≥250 ≥250 ≥250 easy; 5, very easy)
Metric component 6: Metric component 5: No No No No No
Composite score ≥4 ≥4 ≥4 ≥4 ≥4 Tourniquet damage
No. of C-A-T tests 0 24 39 43 43 Metric component 6: No No No No No
Manikin damage
No. of SXT tests 0 21 38 40 41
Metric component 7:
No. of tests favoring C-A-T 0 3 1 3 2 User symptoms No No No No No
C-A-T, Combat Application Tourniquet; SXT, SAM Extremity No. of C-A-T tests 0 41 71 75 75
Tourniquet.
No. of SXT tests 0 14 26 29 53
The first minor finding was a surprisingly large difference in No. of tests favoring C-A-T 0 27 45 46 22
blood loss due to donning gloves and unwrapping a tourni- C-A-T, Combat Application Tourniquet; SXT, SAM Extremity
quet. For pretime, a period of uncontrolled bleeding, the Tourniquet.
bleeding rate was approximately 10.4 mL/second. Over an
average of 39 seconds, loss was calculated to be 406mL. Some strategy was “the race is to be to the swift.” To race through
educators consider such a volume as double the minimally im- the step in assessment of whether the tourniquet worked felt
portant amount indicating tourniquet use in the first place, too obvious to question. On the flipside, one user went slower
specifically, half of a 12-oz can of soda (6 oz [177mL]). Pars- to reduce errors, whether in carefully checking or in stopping
17
ing loss by time segment, 59% (240mL) was due to donning the bleed. His strategy was “to be flawless.” In fact, both strat-
gloves, and 41% (166mL) was due to unwrapping the tour- egies were flawed, because the steps are evaluated differently
niquet. These points are actionable by educators for recom- than the task. This insight did not appear to us until we saw all
mendations of whether users are to don gloves or to unwrap the data analyzed, especially by segmenting times and blood
tourniquets beforehand. No user commented on mitigating losses. Although the task is an aggregate of its steps, for the
blood losses such as when donning because the amount at- task to be optimal, some steps should be faster than others.
tributed to the preparation phase only became obvious when In hindsight, that difference in ideal speed by step was obvi-
the analysis was complete. Thus, no user recorded a recom- ous, but in foresight, no one saw it. Notably, error-free glove
mended workaround for a first responder to minimize blood donning and tourniquet unwrapping resulted in more patient
loss before actually applying the tourniquet. hemorrhage. No one’s intuitive judgment was right because
each implicitly evaluated the task and its steps identically. No
The second minor finding was also surprising: All users made one had thought to switch speeds stepwise. No one recalled
the same mistake in their implicit strategy applied both to such an idea from first aid instruction, from a lesson, or from
the task of tourniquet use and each of its steps. Users spon- expert advice. Now, our advice is to move as fast as practical
taneously self-selected their own strategy. For example, three until bleeding is controlled, then slow down to avoid errors. In
users raced through whether the aim was to minimize bleed- this way, only one change of speed is advised to occur—at the
ing or to assess if the intervention controlled bleeding. Their pivot: control. Both speed and deliberate thinking have value,

40 | JSOM Volume 18, Edition 2/Summer 2018