Assessing the co-occurrence of intimate partner violence domains across the life-course: relating typologies to mental health

Figures & data

Table 1 Victimisation subscale endorsements across the life-course (N=264)

	Parents’ relationships, n (%)	Adolescence (13–17), n (%)	Adulthood (18 +), n (%)
Psychological aggression	205 (78.2)	183 (75.6)	190 (79.2)
Physical assault	73 (29.8)	58 (23.8)	73 (30.5)
Injury	15 (5.7)	12 (4.9)	18 (7.5)
Sexual coercion	–	40 (16.4)	64 (26.4)

Sexual coercion within parental relationships was not assessed in the current study. Values are representative of those endorsing one or more experiences in the relevant subscale domain. Categories are not mutually exclusive. %=percent of those who responded.

Fig. 1 Latent profile plot of CTS subscales across exposure to inter-parental, adolescent, and adult victimisation.

Table 2 Fit indices for a two-class, three-class, four-class, and five-class solution using latent class analysis for the 11 CTS subscales

Model	AIC	BIC	SSABIC	Entropy	Lo-Mendell–Rubin adjusted LRT (p)
2 class	5701.310	5783.469	5710.548	0.888	781.170 (0.0002)
3 class	5504.658	5629.684	5518.717	0.892	217.400 (0.0015)
4 class	5434.549	5602.441	5453.428	0.877	92.722 (0.6257)
5 class	–	–	–	–	–

AIC=Aikaike Information Criteria; BIC=Bayesian Information Criteria; SSABIC=sample size adjusted BIC.

Table 3 Results of univariate ANOVAs using class membership as the independent variable and a number of severity scores on psychological correlates as dependent variables

	Class 1: global victimisation (GV)		Class 2: parental victimisation (CPVE)		Class 3: psychological victimisation (CLPA)
Variable	M	SD	M	SD	M	SD	F	Partial, η ^2
PTSD (PSS-SR-5)	14.10 (n=50)	13.50	13.37 (n=27)	12.62	7.14 (n=114)	11.46	6.99*	0.07
Depression (PHQ-9)	8.38 (n=63)	6.94	8.61 (n=36)	6.50	5.83 (n=155)	4.76	6.63*	0.05
Alcohol use (AUDIT)	5.62 (n=61)	4.14	4.16 (n=31)	4.10	4.01 (n=142)	3.81	3.70*	0.03
Anger (DARS-7)	14.37 (n=62)	5.91	13.97 (n=36)	5.48	11.81 (n=154)	4.77	6.61*	0.05

*p<0.05.

Classes differ in size due to variability of missing data across the different measures.