Table 2.

Actions performed from the beginning of simulation until diagnosis of cardiac arrest

	All simulations	Fastest group 20 (16; 26) s	Slowest group 78 (60; 118) s
Always necessary
Talks loudly to manikin	94% (n = 34)	92% (n = 11)	92% (n = 11)
Shakes manikin	83% (n = 30)	83% (n = 10)	75% (n = 9)
Checks for breathing	67% (n = 24)	92% (n = 11)	58% (n = 7)
Sometimes necessary
Removes bed rail(s)	58% (n = 21)	50% (n = 6)	33% (n = 4)
Activates internal alarm	36% (n = 13)	33% (n = 4)	33% (n = 4)
Never necessary
Prepares to measure blood pressure and/or saturation	11% (n = 4)	0	33% (n = 4)
Pain stimulates manikin	8% (n = 3)	8% (n = 1)	8% (n = 1)
Calls doctor	8% (n = 3)	0	8% (n = 1)
Gives verbal handover to colleagues without performing other actions	8% (n = 3)	0	17% (n = 2)
Removes duvet	8% (n = 3)	25% (n = 3)	0
Exposes manikin’s chest	8% (n = 3)	8% (n = 1)	8% (n = 1)
Checks for pulse	6% (n = 2)	0	8% (n = 1)
Leaves the room to get help	6% (n = 2)	8% (n = 1)	0
Moves the bed	6% (n = 2)	0	17% (n = 2)
Removes pillow	6% (n = 2)	0	8% (n = 1)
Raises the bed	3% (n = 1)	0	8% (n = 1)
Collects equipment	3% (n = 1)	0	8% (n = 1)

Data presented as median (Q1; Q3) or percentages (n). The fastest group included the 1/3 of simulation with the fastest time to diagnose cardiac arrest, whereas the slowest group included the 1/3 of simulation with the longest time to diagnose cardiac arrest