Abstract 108: Effect of a Medical Decision Support Device on Quality of Basic Life Support
Background: As cardiopulmonary arrest (CPA) can occur anywhere, laypersons may need to initiate CPR. However, recent literature reveals the quality of basic life support (BLS) provided during CPA is poor. There is increasing interest in whether devices can support laypeople, as well as others, to manage medical emergencies.
Methods: We performed a prospective, randomized, controlled trial of simulated out of hospital CPA in which each subject was asked to manage a standardized event using the Laerdal Resusci-Anne simulator. A study member played a second rescuer to help as directed. The intervention was a device with audiovisual prompts designed to assist with decision making. Study subjects included laypeople and nursing and medical students. The primary outcome measure was the proportion of subjects who performed compressions and ventilations at a 30:2 ratio. Secondary outcome measures were proportion of subjects who insisted on switching roles with the second rescuer after 2 minutes and other measures of quality BLS.
Results: Thirty-one subjects were enrolled, 16 in the intervention arm and 15 in the control arm, with similar baseline characteristics. Subjects in the intervention arm were more likely to perform compressions to ventilations at a ratio of 30:2 than controls, [15/16 (94%) vs. 4/15 (27%), p<0.001], even with stratification by rescuer background: laypeople [8/9 (89%) vs. 0/4 (0%), p=0.02], medical or nursing student [7/7 (100%) vs. 4/11 (36%), p=0.01]. Intervention subjects were also more likely to insist that the second rescuer switch roles with them than controls, [12/16 (75%) vs. 2/15 (13%), p=0.001]. However, there was a delay in mean time (± standard deviation) to initiation of compressions in the intervention arm, [159.5 (±53) vs. 78.2 (± 20) seconds, p < 0.001] and the no-flow fraction was very high in both arms, [72.2% (± 0.1) vs. 75.4 (± 0.8), p = 0.35].
Conclusions: The device resulted in a behavior change as intervention subjects were more likely to perform 30:2 and switch providers. Unfortunately, the device slowed the initial response time. These results are promising, but future studies should concentrate on decreasing latency time, periods without compressions and possible incorporation into automated external defibrillators.