Page 13 - JSOM Fall 2023

TABLE 1 Volunteer Demographics For all infusion rates, utilization of the DripAssist resulted in
Characteristics Total (%) significantly lower median percentages of error compared to
Sex the TM (Table 3). The median percentage error for volume
Male 17 (85) infused over 60 minutes at a 40mL/hr rate was significantly
Female 3 (15) lower compared to the TM (65% vs. 26.5%, p < .001, Table
Age (years) 4). Additionally, investigators noted that 70% of infusions per-
18–21 6 (30) formed with the TM were more than 50% off from the desired
22–25 10 (50) rate, compared to 20% performed with the DripAssist.
26–29 1 (5)
30–33 2 (10)
>34 1 (5) TABLE 4 Percent Error of Volume Infused Over 60 Minutes
Rank TM % error DA % error
E-1 0 (0) Rate median median p value
E-2 0 (0) 40mL/hr 65 26.5 <.001
E-3 5 (25)
E-4 13 (65) DA = DripAssist; hr = hour; mL = milliliter; TM = traditional method
E-5 2 (10)
TM infusion experience Volunteers reported significantly higher confidence in the
(iterations performed) DripAssist compared to the TM on the Bandura Scale (median
None 4 (20)
1–5 13 (65) 80 vs. 47.5, p <.001). Most volunteers preferred the DripAssist
6–10 1 (5) to establish a rate-based infusion over the TM (90% vs. 10%).
11–15 1 (5) Volunteers likewise evaluated the DripAssist as significantly
>15 1 (5) easier to use and more likely to be learned, remembered, and
TM = traditional method retained over the TM.

an infusion pump but not performing calculations or infusions Discussion
on a live patient.
Our study found significantly shorter times to establish a rate
Volunteers demonstrated significantly faster times to achieve a infusion with fewer rates of error using the DripAssist com-
specific infusion rate with the DripAssist compared to the TM pared to the TM. There was a higher preference and end user
(median 146.5 seconds vs. 207.5 seconds, p = .003, Table 2). appraisal for the DripAssist in a convenience sample of U.S.
Although there was no significant difference in median time to Army medics.
infusion with period effect from repeat interactions of IV in-
fusion, median time for infusion rate achievement was signifi- Limited prior research evaluated mean times to establish infu-
cantly faster with sequence of the TM as the second method sion rates using the DripAssist, without comparison to other
tested compared to the TM as the first method tested. (159.5 infusion rate calculation methods. Our findings demonstrate
10
vs. 179.5, p = .033). relatively slower times to infusion rate achievement across all
infusion rates. However, it is important to note the compara-
TABLE 2 Comparison of Median Time and Period and Sequence tive sample populations. Prior research utilized a large non-
Effects for Time to Achieve Infusion Rate
medic population as part of their sample, including advanced
Median time (sec), paramedics, physician assistants, and nurses to establish infu-
Method (IQR) p-value sion rates. Each of these populations had advanced training
TM 207.5 (136, 350) 0.003 compared to our population. Although direct comparison is
DripAssist 146.5 (100, 206) difficult given that we did not obtain demographic information
Period and Sequence Effect Testing regarding years of practice in our sample, 80% of our sample
Median Time to Achieve Infusion Rate (sec) was below the age of 26 years old. The study by Couperus et
Period 1 (TM1 & DA1) 196 0.26 al. found an average medical practice experience of over 18
Period 2 (TM2 & DA2) 168.5 years across the paramedics, physician assistants, and nurses,
10
Sequence DA1 (DA1 & TM2) 159.5 0.033 which comprised 75% of their sample population. The av-
Sequence TM1 (TM1 & DA2) 179.5 erage years of experience in medics comprising 25% of that
DA = DripAssist; DA1 = DripAssist first; DA2 = DripAssist second; study population was approximately three years. There was
IQR = interquartile range; sec = seconds; TM = traditional method; no subgroup analysis provided for their medic cohort to create
TM1 = traditional method first; TM2 = traditional method second a direct comparison. Advanced experience in their population

TABLE 3 Median Drop, Volume Rates, and Proportional Error Compared to Ideal Infusion Rates
Infusion rate drops/min TM drops/min [IQR] DA drops/min [IQR] TM error DA error
(mL/min) (mL/min) (mL/min) (%) (%) p-value
62.5 47 [20,93] 59 [53,60] 61 7 <.001
(250mL/hr) (188mL/hr) (236mL/hr)
31.0 23 [14,32] 30 [28,32] 39 3 <.001
(125mL/hr) (92.7mL/hr) (121.0mL/hr)
10.0 8 [3,10] 10 [10,12]
(40mL/hr) (32mL/hr) (40mL/hr) 50 20 0.004
Overall median error 46 5 <.001
DA = DripAssist; hr = hour; IQR= interquartile range; mL = milliliters; TM = traditional method

DripAssist and Traditional Methods for Rate Infusions | 11