Quality of audio-assisted versus video-assisted dispatcher-instructed bystander cardiopulmonary resuscitation

This study aimed to conduct a systematic review and meta-analysis comparing the effect of videoassistance and audio-assistance on quality of dispatcher-instructed cardiopulmonary resuscitation (DICPR) for bystanders.
Five databases were searched, including PubMed, Cochrane library, Embase, Scopus and NIH clinical trial, to find randomized control trials published before June 2017. Qualitative analysis and metaanalysis were undertaken to examine the difference between the quality of video-instructed and audioinstructed dispatcher-instructed bystander CPR.
The database search yielded 929 records, resulting in the inclusion of 9 relevant articles in this study. Of these, 6 were included in the meta-analysis. Initiation of chest compressions was slower in the video-instructed group than in the audio-instructed group (median delay 31.5 s; 95% CI: 10.94–52.09). The difference in the number of chest compressions per minute between the groups was 19.9 (95% CI: 10.50–29.38) with significantly faster compressions in the video-instructed group than in the audioinstructed group (104.8 vs. 80.6). The odds ratio (OR) for correct hand positioning was 0.8 (95% CI: 0.53–1.30) when comparing the audio-instructed and video-instructed groups. The differences in chest compression depth (mm) and time to first ventilation (seconds) between the video-instructed group and audio-instructed group were 1.6 mm (95% CI: −8.75, 5.55) and 7.5 s (95% CI: −56.84, 71.80), respectively.
The study concludes that video-instructed DI-CPR significantly improved the chest compression rate compared to the audio-instructed method, and a trend for correctness of hand position was also observed. However, this method caused a delay in the commencement of bystander-initiated CPR in the simulation setting.