Table 1. Hypotheses of the relationship between patient-initiated violence and psychological variables.

					Dependent variable
Model type	Independent variable	Type of variable	Value	Ref.	Emotional exhaustion (EE)	Job satisfaction (JS)		Intention to leave (IL)
						Satisfaction with work (JS-1)	Satisfaction with career (JS-2)	Leave front-line clinical services(IL-1)	Leave the hospital (IL-2)	Quit practicing medicine (IL-3)
MT-I	Verbal abuse(VA)	binomial	0:none	none	+	-	-	+	+	+
			1: occurrence
MT-I	Threats of assault(TA)	binomial	0:none	none	+	-	-	+	+	+
			1:occurrence
MT-I	Physical assault(PA)	binomial	0:none	none	+	-	-	+	+	+
			1:occurrence
MT-II	Verbal abuse(VA)	ordinal	0:none	none	low freq.:+	low freq.: -	low freq.: -	low freq.:+	low freq.:+	low freq.:+
			1:low freq.		high freq:++	high freq: - -	high freq: - -	high freq:++	high freq:++	high freq:++
			2:high freq.
MT-II	Threats of assault(TA)	ordinal	0:none	none	low freq.:+	low freq.: -	low freq.: -	low freq.:+	low freq.:+	low freq.:+
			1:low freq.		high freq:++	high freq: - -	high freq: - -	high freq:++	high freq:++	high freq:++
			2:high freq.
MT-II	Physical assault(PA)	ordinal	0:none	none	low freq.:+	low freq.: -	low freq.: -	low freq.:+	low freq.:+	low freq.:+
			1:low freq.		high freq:++	high freq: - -	high freq: - -	high freq:++	high freq:++	high freq:++
			2:high freq.

Note. Control variables considered in MT-I and MT-II included personal characteristics and work-related factors (see Table 2);

+: positive, -:negative, ++: increasingly positive, --:increasingly negative.