Page 12 - JSOM Fall 2023

the importance of accurate infusions, most U.S. Army medics generator (Research Randomizer, https://www.randomizer.org)
returning from deployment identified medication administra- to decide whether they performed infusion testing with the
tion as an essential skill, but also stated that they would have TM or DripAssist device first. Each volunteer participated in
benefitted from additional training prior to deployment. 9 an isolated environment with a standardized set-up for each
infusion rate tested (Figure 3). Volunteers were allotted five
The DripAssist (Shift Labs, Seattle, WA) is a small 3.8-ounce, minutes to familiarize themselves with the DripAssist prior to
battery-powered, in-line device that monitors the infusion rate testing. They were then asked to achieve infusions at rates of
of an IV medication, providing a user interface that displaces 250mL/hr and 125mL/hr based on prior literature, as well as
10
the rate and volume infused (Figure 2). Medics are still re- 40mL/hr based on recommended sedation rates for PCC. 10,12
quired to manipulate the rate via the roller clamp, however the
DripAssist device calculates the infusion rate for them, thereby
eliminating much of the calculations required in the TM. Lim-
ited research demonstrates that non-physician medical provid-
ers find the use of the DripAssist to be easier for infusion rate
manipulation than the TM, but provides no direct comparison
for device impact on time to achieve infusion rates or drip
10
rate accuracy. A single case report published the use of the
DripAssist in an austere setting to manage infusion rates in
a casualty in Syria. There was a perceived benefit among the FIGURE 3
medical team, however no quantitative data compared infu- Standardized
testing set-up.
sion rates to other methods. 11
FIGURE 2 DripAssist device features.

Time began when volunteers were handed the prescribed rate
and stopped when the participant stated they had established
the required rate. Once volunteers verbalized the established
rate, a single investigator counted the number of drops in the
drip chamber over 30 seconds and then doubled the number
counted to evaluate for rate accuracy. We defined accuracy as
within ten percent based on prior studies of infusion rate ac-
curacy. 6,13 Once volunteers completed three infusions for their
initial method, they proceeded to their second method type.
We sought to directly compare the DripAssist to the TM for Upon completion of both methods, volunteers completed a
time to achieve infusion rates, infusion rate accuracy, and end post-participation survey.
user feedback amongst U.S. Army medics.
Continuous variables, including time, rate accuracy, volume
accuracy, and user confidence, were analyzed using a Mann-
Methods
Whitney U test, while ordinal data of user appraisal was ana-
The Regional Health Command – Pacific Institutional Review lyzed using a Wilcoxon Signed-Rank test. Demographic data
Board determined protocol #221047 exempt from board re- was evaluated using Microsoft Excel (Microsoft, https://www
view. We conducted a prospective, randomized, crossover .microsoft.com/en-us/microsoft-365/excel), while all other data
study utilizing a convenience sample of active-duty U.S. Army was analyzed with SPSS statistical software (IBM, https://www
medics stationed at Joint Base Lewis-McChord, Washington. .ibm.com/spss).
Volunteers were excluded if they self-reported any prior for-
mal advanced medical training, such as medical school, Special Results
Operations Combat Medic, paramedic, physician assistant, li-
censed practice nurse, registered nurse, or if they had any phys- A total of 22 U.S. Army medics initially volunteered for study
ical limitations that would prevent them from participation. participation, however one was excluded due to prior advanced
training, and one withdrew due to scheduling conflicts. The
Given a dearth of literature, investigators agreed upon a mean resulting 20 medics completed 120 total infusions over three
clinically important difference (MCID) of 60 seconds between days of testing at a single outpatient clinic. Median volunteer
modalities for achieving infusion rates. Utilizing a cross- age was 22 years old; most were in the rank of E-4 (65%) and
over design, a power analysis established a sample size of 20 most participants were male (85%, Table 1). Four volunteers
volunteers. (20%) reported never having established a rate-specific infu-
sion in a live patient with the TM. Of those who reported IV
Volunteers received a brief introduction to the study, then com- infusion experience, 65% reported performing between one
pleted a demographic survey, including prior IV infusion ex- and five infusions prior to the study, with smaller amounts
perience. They then were randomized using a random number reporting more. One volunteer (5%) reported experience with

10 | JSOM Volume 22, Edition 3 / Fall 2023