CPR chest compressions are often suboptimal. CPR coaching, such as guiding rate with a metronome, can correct compression and ventilation rates. Measurement and feedback of compressions moderately improve depth but add cost and complexity. We evaluated the effectiveness of a simple vocal prompt on compression depth.Methods: We randomized critical care nurses to perform CPR on an intubated manikin with an AED either with metronome guidance alone (control), or with metronome plus the prompt, “Remember to push hard” at the beginning of each minute (intervention). Each nurse did compressions for a 4-min period, followed by a 2-min rest, and 2 more minutes of compressions.Results: The participants did not differ in mean weight, height, total number of compressions (595 ± 37 vs. 616 ± 46, mean ± standard deviation), or compression rate (99 ± 6 min−1 vs. 103 ± 8 min−1) for control (n = 29) and intervention (n = 19) groups. Compression depth over all 6 min of compressions was 31 ± 9 mm control vs. 33 ± 10 mm intervention. Depth improved most following the 1-min prompt, when the 3-s mean depth increased from 33.8 ± 9.2 mm to 36.6 ± 9.5 mm after the prompt (n = 19, p < 0.005, paired t-test). The number of participants performing any compressions with excessive depth or incomplete release was not different between groups. Overall, 42 of 48 subjects found the metronome useful; 15 of 19 thought “Remember to push hard” was useful.Tabled 1MinuteControlInterventionDelta (mm)95%CI (mm)%diff135.2 ± 8.136.2 ± 9.41.0[−4.1,6.2]3.0%231.8 ± 8.134.4 ± 9.32.7[−2.5,7.8]8.4%329.6 ± 9.432.6 ± 9.23.0[−2.5,8.6]10.3%428.4 ± 9.330.9 ± 10.02.5[−3.2,8.2]8.8%Pause732.6 ± 8.634.6 ± 9.92.0[−3.4,7.4]6.1%829.4 ± 9.330.4 ± 9.41.0[−4.5,6.5]3.4% Open table in a new tab CPR chest compressions are often suboptimal. CPR coaching, such as guiding rate with a metronome, can correct compression and ventilation rates. Measurement and feedback of compressions moderately improve depth but add cost and complexity. We evaluated the effectiveness of a simple vocal prompt on compression depth. Methods: We randomized critical care nurses to perform CPR on an intubated manikin with an AED either with metronome guidance alone (control), or with metronome plus the prompt, “Remember to push hard” at the beginning of each minute (intervention). Each nurse did compressions for a 4-min period, followed by a 2-min rest, and 2 more minutes of compressions. Results: The participants did not differ in mean weight, height, total number of compressions (595 ± 37 vs. 616 ± 46, mean ± standard deviation), or compression rate (99 ± 6 min−1 vs. 103 ± 8 min−1) for control (n = 29) and intervention (n = 19) groups. Compression depth over all 6 min of compressions was 31 ± 9 mm control vs. 33 ± 10 mm intervention. Depth improved most following the 1-min prompt, when the 3-s mean depth increased from 33.8 ± 9.2 mm to 36.6 ± 9.5 mm after the prompt (n = 19, p < 0.005, paired t-test). The number of participants performing any compressions with excessive depth or incomplete release was not different between groups. Overall, 42 of 48 subjects found the metronome useful; 15 of 19 thought “Remember to push hard” was useful.