Table 3.

Descriptive statistics of the quality of general counseling and suicide safety planning skills based on nurses’ performance on 30-minute role-plays with a patient actor rated using the automated coding system Lyssn Advisor human coding with the Safety Plan Intervention Rating Scale, Version 3 (SPIRS)^a.

	Values, n (%)b	Role-play length, mean (SD)	Lyssn Advisor empathy, mean (SD; 95% CI)c	Lyssn Advisor collaboration, mean (SD; 95% CI)c	SPIRS generald, mean (SD; 95% CI)c	SPIRS steps, mean (SD; 95% CI)c
Postdidactic	17 (89)	24.51 (6.20)	4.01 (0.37; 3.76-4.26)	3.53 (0.59; 3.26-3.80)	7.94 (3.29; 6.37-9.51)	7.88 (1.69; 7.08-8.68)
Posttraining	14 (74)	28.81 (7.20)	4.14 (0.62; 3.83-4.45)	3.62 (0.66; 3.27-3.97)	6.93 (2.43; 5.66-8.20)	8.07 (1.77; 7.15-8.99)
6-months follow-up	10 (53)	29.00 (4.32)	4.15 (0.43; 3.88-4.42)	3.72 (0.45; 3.45-3.99)	9.10 (2.60; 7.49-10.71)	8.10 (1.66; 7.06-9.14)

^aRole-plays were completed after (1) the didactic portion of the training in which nurses learned about safety planning and watched demonstration videos of safety planning, (2) practicing general counseling microskills with the simulated patient Client Bot Emily, and (3) viewing automated coding scores for general counseling skills through Lyssn Advisor.

^bPercentages are out of 19 nurses who provided consent.

^c95% CIs presented are based on the SEM.

^dSPIRS general includes the sum of 6 items of general skills for the delivering suicide safety planning and SPIRS steps includes the sum of 6 items of completing the safety planning steps. Both scales are rated from 0 to 3 with higher scores indicating higher quality of skills. Lyssn Advisor empathy and collaboration scores are on a scale from 0 to 5; higher scores indicate higher quality skills and cutoff of 3.5 for what may be considered basic ability and 4 for advanced ability.