Lifetime prevalence of polyvictimization among older adults in Sweden, associations with ill-heath, and the mediating effect of sense of coherence

Table 4 Final model for the association between violence victimization and mental health status. A moderated (social support) mediation (sense of coherence) effect was found for victimization on mental health status. Social support had a near significant moderation effect also on the direct effect of victimization on mental health status


		Effect	95% CI
Relative conditional indirect effect (ab) (Victimization – SOC – Mental health status, moderated by social support)
With social support	One form of victimization	−0.42	−1.62	0.76
	Low polyvictimization	−1.14	−2.29	−0.21
	High polyvictimization	−4.32	− 7.17	−2.02
No social support	One form of victimization	−0.58	−3.28	1.95
	Low polyvictimization	−5.13	−9.75	−1.34
	High polyvictimization	−4.81	−7.59	−2.29
Significance test for moderating effect of social support: Index of moderated mediation for low polyvictimization: 3.98; CI 0.09–8.39
Relative conditional direct effect (c’) (Victimization - Mental health status, moderated by social support)
With social support	One form of victimization	−1.16	−3.83	1.51
	Low polyvictimization	0.08	−2.12	2.28
	High polyvictimization	−1.65	−5.09	1.79
No social support	One form of victimization	−5.38	−10.02	−0.73
	Low polyvictimization	−6.96	−12.79	−1.12
	High polyvictimization	−6.19	−10.75	−1.63
Significance test for moderating effect of social support: Test of highest order unconditional interaction: p = 0.06 Test of equality means: with social support p = 0.67, no social support p = <0.01

Note: Model summary R² = 0.36. Included in analyses = 471, Missing cases = 136. Confidence intervals are computed using bootstrapping with 5000 sample. No moderating effect of SOC was found (interaction effect p = 0.53)

ISSN: 1471-2318