Emotional-behavioral resilience and competence in preschool children exposed and not exposed to intimate partner violence in early life

Abstract

Exposure to intimate partner violence (IPV) during childhood is a risk factor for poor emotional-behavioral functioning. Despite this, many children show resilience in the face of IPV exposure. The current study aimed to identify characteristics associated with positive emotional-behavioral outcomes in 4-year-old children exposed to IPV in early life. Data were drawn from the Maternal Health Study (MHS), a prospective study of women during pregnancy and following the birth of their first child. Women were recruited from six Melbourne public hospitals between 2003 and 2005. Mother–child dyads (n = 1060) were included in the study using data collected during pregnancy; 12 months postpartum; and four years postpartum. Of the children exposed to IPV in early life, 38% displayed emotional-behavioral resilience at four years. Maternal physical wellbeing, mothers’ return to work or study and no exposure to IPV at four years were associated with child resilience. These results highlighted the importance of prioritizing mothers’ physical wellbeing and access to employment in promoting positive outcomes for their children. The results also reinforced the significant role of early intervention; when exposure to IPV stops at an early age, children are more likely to experience emotional-behavioral resilience.

Keywords

Resilience children intimate partner violence emotional-behavioral functioning

In Australia, approximately one million children are exposed to IPV each year (Australian Domestic and Family Violence Clearinghouse, 2011). These children have an increased risk of emotional-behavioral difficulties (Kitzmann, Gaylord, Holt, & Kenny, 2003). While some children exposed to IPV experience adverse outcomes, research shows other children in seemingly similar circumstances do not experience poor health or development (Holt, Buckley, & Whelan, 2008). There is still much to learn about what assists these children to display resilience despite exposure to IPV.

The Impact of Exposure to IPV on Children’s Development

Young children are highly vulnerable to the effects of exposure to IPV due to their developmental vulnerability (Lupien, McEwen, Gunnar, & Heim, 2009) and high dependence on caregivers (Howell, 2011). Specifically, children exposed to IPV experience an increased risk for depression and anxiety (Holt et al., 2008), attentional problems, conduct disorders (Flach et al., 2011), and trauma symptoms (Chemtob & Carlsonalso, 2004). However, many children do continue to develop according to typical expectations (Graham-Bermann, Gruber, Howell, & Girz, 2009; Howell, Graham-Bermann, Czyz, & Lilly, 2010). A meta-analysis found that 37% of children exposed to IPV displayed outcomes which were similar or better than children who had not been exposed to IPV (Kitzmann et al., 2003). These findings showed that some children maintain normal health and development displaying resilience.

Resilience Theory

Resilience is successful adaptation in the face of significant adversity (Luthar, Cicchetti, & Becker, 2000) and is determined by complex interactions between the individual and their environment (Cicchetti, 2013). Protective and resource factors help understand why some children adapt to certain adversities. Protective factors are those providing unique protective effect in high-risk settings, which is not apparent in low-risk settings (Luthar et al., 2000). This differs from resource factors, which are associated with positive outcomes in high and low-risk settings (Luthar et al., 2000).

Bronfenbrenner’s Ecological Theory of Development (1986) emphasizes the influence of individual, family, community, and socio-political risk factors on child development and provides a useful framework for identifying protective and resource factors. The child’s environment consists of several systems: (a) the individual system (i.e., child’s characteristics such as temperament); (b) the microsystem (i.e., immediate environment such as the family and school setting); (c) the mesosystem (i.e., interactions between two or more settings from the microsystem); (d) the exosystem (i.e., societal structures such as parental workplace and access to services); and (e) the macrosystem (i.e., wider socio-cultural context including the policy and community beliefs) (Bronfenbrenner, 1986). Protective and resource factors operating at each level can help to better understand how children display resilience despite exposure to IPV (Luthar et al., 2000).

Factors Associated with Resilience in Children Exposed to IPV

Research on resilience in children exposed to IPV has focused on maternal factors, which fall within the microsystem. Maternal mental health was identified as a protective factor in a study of 190 mother–child dyads defining resilience as the maintenance of non-clinical levels of emotional-behavioral problems on the Child Behavior Checklist (CBCL; Martinez-Torteya, Bogat, von Eye, & Levendosky, 2009). The protective role of maternal mental health has been further supported in studies using the CBCL (Cohodes, Chen, & Lieberman, 2017; Howell et al., 2010). It is likely mothers experiencing good mental health are better able assist their children to learn emotion regulation by responding adaptively when they experience intense emotional responses (Howell et al., 2010). Contrary to past findings, a study of 7,743 families did not find maternal mental health to be protective (Bowen, 2017). This study however, examined boys and girls separately, looking soley at resilience in externalizing behavior. These differences may account for the variation in results. Further investigation is needed to understand the role of maternal mental health in promoting resilience in children exposed to IPV.

Research into individual child characteristics which buffer the risk associated with IPV exposure has been scarce. Martinez-Torteya, Bogat, von Eye, and Levendosky (2009) found child’s temperament to be protective for children exposed to IPV. Few studies have explored factors in the broader family environment. Having greater in-home family support (i.e., number of significant others residing in family home) has been found to predict better outcomes in a community and shelter population of children exposed to IPV (n = 120) (Miller, VanZomeren-Dohm, Howell, Hunter, & Graham-Bermann, 2014). Higher maternal educational attainment has also been found to be associated with children’s emotional-behavioral functioning across multiple studies (Bowen, 2017). Investigation of a broader range of factors in the exosystem such as mother’s return to work is yet to be investigated.

Key Methodological Considerations

Several methodological challenges in resilience research are worth noting. First, the lack of a valid measure of childhood resilience creates measurement difficulties. A common approach has been to define resilience as the absence of difficulties in one domain of functioning for children experiencing adversity. Most studies have assessed emotional-behavioral resilience using the CBCL (Martinez-Torteya et al., 2009; Miller et al., 2014) or the SDQ (Bowen, 2017). Although there are other domains of functioning affected by IPV, the present study assessed emotional-behavioral resilience as this is integral to children’s learning, language, peer, mental and physical health outcomes (Fergusson, Horwood, & Ridder, 2005).

Second, there are two main analytical approaches to investigating resilience. Variable-centered approaches focus on identifying factors associated with positive outcomes for all children, and how these interact with exposure to high or low adversity. Although studies taking this approach have contributed significantly to the field, resilience researchers are encouraging the use of person-centered approaches (Cicchetti, 2013; Masten & Obradović, 2006). These approaches are increasingly popular (Bowen, 2017; Giallo et al., 2018; Hopkins, Zubrick, & Taylor, 2014; Martinez-Torteya et al., 2009), and seek to identify a group of children displaying resilience in one or more domains within the context of adversity. This classification approach enables the identification of factors which are uniquely protective for resilient individuals (Hopkins et al., 2014; Martinez-Torteya et al., 2009). Despite the different approaches taken, there is some evidence suggesting they yield similar results (Miller-Lewis, Searle, Sawyer, Baghurst, & Hedley, 2013). In addition, person-centered approaches offer strong potential for informing interventions for individuals with similar experiences, such as exposure to IPV (Swartout, Swartout, & White, 2011). For these reasons, a person-centered approach was chosen for the current study. To provide a comparison of our person-centered findings, we also conducted a variable-centered approach as a sensitivity analysis.

With respect to IPV research specifically, there are several limitations worth noting. First, many studies rely on samples from domestic violence shelters (Howell et al., 2010; Miller et al., 2014) where children are likely to have experienced recent and severe violence, and experience exposure to community violence (Holt et al., 2008), limiting the extent to which results might be generalized to broader populations. Second, many studies used measures of IPV which only take into consideration physical and sexual violence (Cohodes et al., 2017). A broader definition including psychological violence should be used as these acts are also detrimental to women and children (Pico-Alfonso et al., 2006). Additionally, many studies have used cross-sectional designs (Cohodes et al., 2017; Howell et al., 2010; Miller et al., 2014), which do not account for sleeper effects of IPV exposure, where impacts are not immediately apparent, but experienced at a later stage of development (Holmes, 2013). Similarly, few studies have assessed the impact of exposure to IPV in early life representing a time of developmental vulnerability. Finally, the sample sizes have been small, with low statistical power.

Study Aims

The current study sought to address a number of current knowledge gaps and methodological concerns by drawing on data from an Australian longitudinal cohort study of over 1,000 first-time mothers. First, this study aimed to determine the proportion of children exposed to IPV within the first 12 months postpartum that displayed emotional-behavioral resilience at four years of age. Second, the study aimed to identify psychosocial factors (i.e., maternal mental health and maternal physical wellbeing) associated with emotional-behavioral resilience among 4-year-old children exposed and not exposed to IPV in the first year postpartum. By identifying psychosocial factors associated with emotional-behavioral resilience, we can better understand what support will be beneficial to children and their families experiencing IPV.

Method

Study Design and Sample

Data for this study were drawn from the Maternal Health Study (MHS), a prospective study investigating the health and wellbeing of mothers during pregnancy and following the birth of their first child. Ethics approval for the study was sought from the participating hospitals, La Trobe University, and the Royal Children’s Hospital, Melbourne. A detailed record of the study design and methods can be found in the published study protocol (Brown, Lumley, McDonald, & Krastev, 2006). In summary, women registered to give birth at six Melbourne metropolitan hospitals between April 2003 and December 2005 were invited to participate. Women were eligible if they were (a) 18 years or older; (b) nulliparous; (c) had an estimated gestation of up to 24 weeks at the time of enrolment; and (d) were sufficiently proficient in English to complete questionnaires and telephone interviews. Hospital staff mailed eligible women an invitation pack comprising an invitation letter, consent form, prepaid return envelope, and baseline questionnaire. This was followed up with a single reminder postcard. Follow-up data were collected during pregnancy at 30–32 weeks’ gestation, as well as 3, 6, 9, 12, and 18 months postpartum and when their first child was 4 years of age. This paper draws on data collected at enrolment, at 3 months, 12 months, and 4 years postpartum.

Approximately 6,000 eligible women were invited to take part in the study. As hospital staff mailed out invitation packs on behalf of the MHS, an exact response rate was unable to be determined. A conservative response rate was estimated at between 29% and 31% with 1,507 eligible women returning the baseline questionnaire. From enrolment in pregnancy to 18 months postpartum, 126 participants (8.4%) withdrew or were lost to follow-up. Reasons included: participant declined further participation (n = 58); participant unable to be contacted (n = 29); participant too busy (n = 21); undisclosed reason (n = 7); stillbirth (n = 6); infant death (n = 2); infant ill health (n = 2); and maternal ill health (n = 1). The number of surveys returned at the 3-, 6-, 12-, 18-months and 4-year follow-ups were 1,431, 1,400, 1,357, 1,327, and 1,102, respectively. Selective attrition was observed with participants who completed the 4-year follow-up being older, tertiary educated, less likely to be receiving a government benefit and report IPV or depressive symptoms.

A further 42 cases were excluded from the current analysis due to either (a) missing data on the Composite Abuse Scale (CAS) at 12 months, or (b) no SDQ parent-report information at four years. The final sample comprised 1,060 women. The characteristics of the sample were compared with routinely collected data for nulliparous women giving birth in Victorian public hospitals during the data collection period to assess for representativeness. The sample included fewer younger women (aged 18–24 years) and fewer women born overseas from non-English speaking backgrounds. Despite this, the sample was representative in terms of method of birth and infant birth weight. Education levels within our sample were compared with 2006 Victorian census data from the Australian Bureau of Statistics (ABS) (2007). Mothers with a tertiary qualification were overrepresented within our sample with 75% of participants holding a tertiary qualification compared to 42% of Victorian women aged 15–55 years (ABS, 2007). Five women within the sample identified as Aboriginal and/or Torres Strait Islander. This was too few to enable meaningful comparisons. Demographic characteristics of the sample and how this compares with the routinely collected Victorian data are displayed in Table 1.

Table 1.

Sample Demographic Characteristics.

Characteristics	Study participants n = 1,060, n (%)	Victorian public patients^a n = 40905, n (%)
Maternal Age
< 25 years	105 (9.9)	12,216 (29.9)
≥25 years	955 (90.1)	28,679 (70.1)
Country of birth
Australia	826 (77.9)	29,791 (72.8)
Overseas	234 (22.1)	10,847 (26.7)
Relationship Status
Married	658 (62.1)	22,790 (55.7)
Living with partner	361 (34.1)
Divorced/separated	5 (.5)
Single	36 (3.4)
Highest Level of Education
Year 12 or less	260 (24.5)	n/a^b
Tertiary	795 (75)
Employment Status
Paid employment	916 (86.4)	n/a^b
Not in paid employment	125 (11.8)
Not reported	19 (1.8)
Personal income (AUS $)
>60,000	153 (14.4)	n/a^b
40,001–60,000	352 (33.2)
20,001–40,000	368 (34.7)
<20,000	122 (11.5)
Not reported	65 (61)
Low income (recipient of health care card)	55 (5.2)	n/a^b

Note. ^aVictorian Perinatal Data Collection Unit (Reiley, Davey, & King, 2005).

^bn/a indicates this information was not routinely collected by Victorian hospitals and therefore not available.

Measures

Intimate partner violence

Measured at 12 months and four years postpartum using the shortened CAS (Hegarty, Bush, & Sheehan, 2005; Hegarty, Sheehan, & Schonfeld, 1999). The shortened CAS comprises 18 items of actions by a partner constituting emotional or physical violence. Examples of items include “Slapped me” and “Blamed me for causing their violent behavior.” Women are asked how frequently they experienced each specific behavior during the last 12 months (never, only once, several times, once per month, once per week, daily). Scores of ≥ 3 for emotional items are indicative of emotional violence, and a score of ≥ 1 on physical items indicative of physical violence (Hegarty et al., 2005). The CAS has strong psychometric properties (Cronbach alpha of .80) and was developed and validated in Australia (Hegarty et al., 2005).

Children’s emotional-behavioral resilience

Assessed using the parent-report of the SDQ (Goodman, 1997) at age four. The 25 items assess emotional and behavior symptoms such as sadness, worries, attention difficulties, and fighting. Items are rated on a 3-point scale ranging from 0 = not true, 2 = certainly true, with higher scores indicating more emotional-behavioral difficulties. The SDQ comprises five subscales: Hyperactivity/Inattention; Emotional Symptoms; Peer Problems; Prosocial Behavior; and a Total Difficulties Scale. Using the Australian normed cut-off scores, children were classified as resilient if they scored in the normal range on all five subscales. The SDQ has been evaluated within Australian samples with moderate to strong internal consistency and test–retest reliability (Hawes & Dadds, 2004).

Potential Predictor Variables at the Individual Level

Child gender

Reported at three months postpartum.

Potential Predictor Variables at the Microsystem Level

Maternal mental health

Assessed at four years postpartum by the Edinburgh Postnatal Depression Scale (EPDS) (Cox, Holden, & Sagovsky, 1987) The 10 items assess depressive symptoms such as “I have been so unhappy that I have been crying.” Women are asked to indicate the extent that they have experienced the symptoms in the last week (0 = never, 1 = not often, 2 = sometimes, and 3 = most of the time). A cut-off score of ≥ 13 is recommended when screening for major depression (Murray & Cox, 1990). The EPDS has been well validated for use in pregnancy, in non-postnatal women, as well as in Australian populations (Cox, Chapman, Murray, & Jones, 1996; Murray & Cox, 1990).

Maternal physical wellbeing

Assessed at four years postpartum using the Physical Health component score from the Short Form-36 (SF-36; Ware, Snow, Kosinski, & Gandek, 1993). This provides a general measure of health and quality of life across the four subscales: physical functioning; role limitations due to physical health; bodily pain; and general health. Higher scores on each scale indicate better physical health and wellbeing. The SF-36 has been validated in both clinical and general populations (Ware & Gandek, 1998).

Maternal involvement in home activities

Assessed at four years postpartum using a five-item scale from the Growing up in Australia: Longitudinal study of Australian children (National Center for Education Statistics, 2002). Mothers were asked to indicate how many days in the past week (1 = none, 2 = 1–2 days, 3 = 3–5 days, 4 = 6–7 days) they had engaged in a range of activities with their children such as read to them and involved them in everyday activities. Scores were summed with higher scores indicating greater maternal involvement in home activities.

Stressful life events

Measured at four years postpartum using 20 items drawn from the Pregnancy Risk Assessment and Monitoring System (PRAMS) (Gilbert, Shulman, Fischer, & Rogers, 1999). Responses were categorized as 0 = no exposure to stressful life events, 1 = exposure to one or two stressful life events, 2 = exposure to three or more stressful life events.

Maternal confidence in looking after baby

Measured at 12 months postpartum using a single item as (0 = very confident to 1 = less confident).

Demographic factors

Demographic factors of interest were measured at the baseline questionnaire in pregnancy and include maternal age (≤24 years, ≥25 years), maternal education level, and maternal income.

Potential Predictor Variables at the Exosystem Level

Return to work or study

At 12 months postpartum this was recorded as yes = returned to work or study, or no = have not returned to work or study.

Data Analysis

Data analyses were conducted across two stages using SPSS Version 23. First, a derived variable of children’s emotional-behavioral resilience was constructed to determine children displaying emotional-behavioral resilience following IPV exposure. To do this, “at-risk” and “clinical” cut-off scores on the SDQ-Parent reports were cross-classified with cut-off scores on the CAS indicative of experiencing IPV. This resulted in four distinct groups of children being identified: (1) the “competent” group comprising children whose mothers did not report experiences of IPV at 12 months postpartum (low-risk) and displayed few emotional-behavioral difficulties at four years postpartum; (2) the “struggling” group comprising children whose mothers did not report IPV at 12 months postpartum (low-risk) and displayed emotional-behavioral difficulties; (3) the “vulnerable” group comprising children whose mother reported IPV at 12 months postpartum (high-risk) and displayed emotional-behavioral difficulties at four years and; (4) the “resilient” group comprising children whose mother reported experiences of IPV at 12 months postpartum (high-risk) and displayed positive emotional-behavioral outcomes at four years postpartum. Chi-square tests and an analysis of variance (ANOVA) were conducted to determine whether the four groups statistically differed across the variables of interests. Chi-square statistics and a one-way ANOVA were used for categorical and continuous variables, respectively

Second, to identify protective and resource factors, multivariate logistical regressions models were estimated. These models evaluated the association between the psychosocial factors and positive emotional-behavioral outcomes for children exposed to IPV at 12 months postpartum (model 1) and children not exposed to IPV at 12 months postpartum (model 2). Two separate models were estimated to assess similarities and differences in the psychosocial factors of significance in high-risk versus low-risk contexts. The reported associations between factors and positive outcomes are expressed as odds ratios with a 95% confidence interval. An odds ratio of <1.0 indicates a reduced likelihood of a positive outcome in relation to the reference group (i.e., children identified with poor functioning within the high and low adversity contexts), whereas an odds ratio of >1.0 indicates an increased likelihood of a positive outcome relative to the reference group.

Missing data were replaced in the analyses using multiple imputation. Ten complete datasets incorporating all analysis variables were imputed. Maternal involvement in home activities and maternal physical health were included in the imputation model as continuous variables, whilst all other variables were imputed as binary variables. All estimates were obtained by averaging results across the ten imputed datasets. Only participants who had completed the main outcomes measures of the SDQ-Parent report and the CAS were included in the analysis. Therefore no missing data were observed for these measures.

The statistical approach used in the current study was based on a person-centered approach to operationalizing resilience. However, as variable-centered approaches are also commonly used, a sensitivity analysis was conducted. In this analysis, a hierarchical multiple regression was used to determine if any of the potential predictor variables moderated the relationship between children’s exposure to IPV at 12 months postpartum and their scores of the SDQ. A predictor was identified as protective if its interaction with exposure to IPV at 12 months postpartum resulted in significantly reduced scores on the SDQ.

Results

Descriptive Statistics

The mean proportion of missing data across the variables of interest was 1.1%. Maternal income during pregnancy had the highest amount of missing data (6.1%) with the remaining missing data within variables ranging from 0% to 4%. Missing data were replaced in the analyses using multiple imputation as described above. Descriptive statistics for the study variables are presented in Table 2.

Table 2.

Study Variables Descriptive across Total Sample and Child Emotional and Behavioral Group.

	Total sample n = 1060 (%)	Struggling n = 431(%)	Competent n = 469 (%)	Vulnerable n = 100 (%)	Resilient n = 60 (%)	X² / F-value	η² or Cramer’s V
Emotional-behavioral resilience, M (range 0–40)	7.98	10.71	4.77	12.03	4.72	315.56***	0.47
Demographic Variables
Maternal age at enrolment, M (range 18–49)	30.85	30.55	31.20	29.94	31.81	3.67*	.01
Child gender						16.67**	.13
Female	501 (47.3)	180 (41.8)	251 (53.5)	38 (38)	32 (53.3)
Male	544 (51.3)	247 (54.8)	211 (45)	58 (58)	28 (46.7)
missing	15 (1.4)	4 (.9)	7 (1.5)	4 (4)	0
Maternal education during pregnancy						14.62**	.12
Tertiary	795 (75)	319 (74)	372 (79.3)	63 (63)	41 (68.3)
Year 12 or less	260 (24.5)	109 (25.3)	95 (20.3)	37 (37)	19 (31.7)
Missing	5 (.5)	3 (.7)	2 (.4)	0	0
Maternal income						15.23*	.09
≤40,000	490 (46.2)	204 (47.4)	204 (43.5)	58 (58)	24 (40)
40,000-60,000	352 (33.2)	139 (32.3)	149 (36)	20 (20)	24 (40.0)
≥60,000	153 (14.4)	57 (13.2)	77 (16.4)	9 (9)	10 (16.7)
missing	65 (61)	31 (7.2)	19 (4.1)	13 (13)	2 (3.3)
12-month postpartum variables
Return to work/study						8.50*	.09
Yes	700 (66)	277 (64.3)	322 (68.7)	56 (56)	45 (75)
No	358 (33.8)	152 (35.3)	147 (31.1)	44 (44)	15 (25)
missing	2 (.2)	2 (.5)	0	0	0
Confidence in looking after baby						14.07**	.12
Yes	821 (77.5)	324 (75.2)	384 (81.9)	55 (66)	47 (78.3)
No	239 (22.5)	107 (24.8)	85 (18.1)	36 (34)	13 (21.7)
Missing	0	0	0	0	0
4-year postpartum variables
Intimate Partner Violence						223.24**	.45
No abuse	884 (83.4)	384 (89.1)	430 (91.7)	35 (35)	35 (58.3)
Emotional and/or physical abuse	168 (15.8)	44 (10.2)	38 (8.1)	61 (61)	25 (41.7)
Missing	8 (.8)	3 (.7)	1 (.2)	4 (4)	0
Maternal Mental Health						31.17**	.171
Below cut-off	934 (88.1)	377 (87.5)	433 (92.3)	73 (73)	51 (85)
Above cut-off	124 (11.7)	54 (12.5)	35 (7.5)	27 (27)	8 (13.3)
Missing	2 (.2)	0	1 (.2)	0	1 (1.7)
Stressful life events, M (range – 0–20)	1.84	1.75	1.48	3.19	2.97	33.63**	.09
Maternal Physical Health						3.28*	.01
Physical Component Score, M (range 0–100)	52.66	52.4	52.96	50.95	55.08
Maternal involvement in home activities, M (range 5–20)	13.63	13.20	14.11	12.99	14.08	9.28**	.03

Note. *p < .05, **p < .01, ***p < .001.

Prevalence of IPV and Children’s Emotional-Behavioral Functioning

IPV was reported by 15.1% of women in the 12 months following the birth of their first child. More specifically, 8.7% reported experiencing emotional violence alone, 2% reported physical violence alone, and 4.4% reported experiencing both physical and emotional violence. Compared to children not exposed to IPV in the first year of life, children exposed to IPV had increased odds of emotional or behavioral difficulties at four years postpartum (OR = 1.67 95% CI 1.21– 2.30, p = .002).

Children’s Emotional-Behavioral Resilience Status

The classification of children’s emotional-behavioral status at age 4 demonstrated that out of the entire sample, 44.2% of children were not exposed to IPV at 12 months postpartum and displayed good emotional-behavioral functioning (“competent outcomes”), whereas 40.7% were not exposed to IPV and displayed emotional-behavioral difficulties (“struggling outcomes”). Conversely, 9.4% of children were exposed to IPV and displayed emotional-behavioral difficulties (“vulnerable outcomes”) and 5.7% of children were exposed to IPV at 12 months postpartum and displayed good emotional-behavioral functioning (“resilient outcomes”). Of those children who were exposed to IPV with 12 months postpartum, 37.5% were categorized as resilient, displaying good emotional-behavioral functioning, whereas 62.5% displayed emotional-behavioral difficulties. Descriptive statistics for the variables of interest across each emotional-behavioral resilience category are displayed in Table 2. Chi-square X² values and ANOVA F-values indicating whether the four groups statistically differed across the predictor variables are also presented in Table 2.

Predictors of Child Emotional-Behavioral Status

Table 3 presents the multivariable logistic regression models predicting the likelihood of emotional-behavioral resilience within the context of exposure to IPV during the first 12 months of life (Model 1) and no exposure to IPV in early life (Model 2).

Table 3.

The Multivariate Logistic Regression Results Predicting “Resilient” Children in the IPV Exposed Group and “Competent” Children in the Non-Exposed Group.

Psychosocial variables	Model 1: Exposure to Intimate Partner Violence at 12 months postpartuma n = 160	Model 2: No exposure to Intimate Partner Violence at 12 months postpartumb n = 900
Psychosocial variables	OR (95% CI)	OR (95% CI)
Maternal age at enrolment-
≥25 years	1.07 (.30–3.80)	1.88* (1.13–3.13)
Child gender – female	2.16 (.97–4.79)	1.67*** (1.27–2.20)
Maternal education during pregnancy – tertiary	1.26 (.51–3.11)	1.16 (.83–1.63)
Maternal income
<40,000	Ref.	Ref.
40,000–60,000	1.68 (.49–5.78)	1.08 (.71–1.66)
≥60,000	2.36 (.93–6.01)	1.01 (.73–1.39)
12-month postpartum variables
Return to work/study – yes	2.38* (1.04-5.48)	1.15 (.85–1.56)
Confident in looking after baby
– yes	1.59 (.66–3.81)	1.59** (1.14–2.23)
4-year postpartum variables
Intimate Partner Violence – No abuse	2.31* (1.03– 5.19)	1.05 (.64–1.73)
Maternal Mental Health – below cut-off	2.54 (.90–7.14)	1.54 (.95–2.49)
Stressful life events
None	.62 (.19–2.02)	1.35 (.91–2.01)
One to two events	1.16 (.47–2.86)	1.02 (.72–1.46)
Three or more events	Ref.	Ref.
Maternal Physical Health	1.07** (1.02–1.13)	1.01 (.99–1.03)
Maternal Involvement	1.10 (.97–.26)	1.11*** (1.06–1.16)

Note. *p < .05, **p < .01, ***p < .001.

^aReference category for Model 1 is the Vulnerable group.

^bReference category of Model 2 is the Struggling group.

Model 1 exposure to IPV during the first 12 months of life-modeling likelihood of emotional-behavioral resilience versus vulnerability

Results of the model revealed that mothers’ returned to work or study at 12 months postpartum; maternal physical well-being at four years postpartum; and no exposure to IPV between three and four years postpartum were significantly associated with children’s emotional-behavioral resilience at 4 years of age.

Due to the well-established link between the experience of IPV and depressive symptoms (Devries et al., 2013), the analysis was also conducted with exposure to IPV between 3 to 4 years of age removed from the model. In this model, low maternal psychological distress at four years postpartum predicted children’s emotional-behavioral resilience in those exposed to IPV, with an odds ratio of 2.81 (95% CI 1.03–7.71, p = .04).

Model 2 no exposure to IPV during the first 12 months of life-modeling the likelihood of emotional-behavioral competence versus struggling

Table 3 also displays the multivariate logistic regression model predicting emotional-behavioral competence. Within this model, the following factors were significantly associated with children’s emotional-behavioral functioning: being a female; having a mother who was 25 years or older at enrolment; having a mother who felt confident looking after their child; and mothers who reported higher levels of involvement with their child.

Factors Uniquely Associated with Emotional-Behavioral Resilience in Children Exposed to IPV in Their First Year of Life

Models 1 and 2 were compared to identify protective factors associated with emotional-behavioral functioning in children who were exposed to IPV at 12 months postpartum but not in children who had not been exposed to IPV at this time. Three factors were identified as being uniquely associated with emotional-behavioral resilience for children exposed to IPV: (1) mother’s return to work/study at 12 months postpartum; (2) no exposure to IPV between ages 3 and 4 years; and (3) having a mother with higher levels of physical wellbeing.

Factors Associated with Positive Emotional-Behavior Outcomes in Both Children Exposed and Not-Exposed to IPV during the First Year of Life

Model 1 and 2 were also compared to identify resource factors which were associated with positive emotional-behavioral functioning both for the group of children exposed to IPV and for those children not exposed. As no variables were significant for both model 1 and model 2, no resource factors were identified.

Sensitivity Analysis – Variable-Centered Approach

In acknowledgment of traditional variable-centered approaches, we conducted a series of hierarchical multiple regression models predicting children’s emotional-behavioral difficulties on the SDQ. The results of the multiple regression analyses are presented in Table 4. In step 1, child gender, maternal age at pregnancy and maternal education were entered as covariates. The model explained 4% of the variance in SDQ scores (R² = .04, F(3,1056) = 14.01, p < .001), with all covariates significantly contributing to the model.

Table 4.

Multiple Regression Predicting Child Emotional-Behavioral Resilience at 4 Years (n = 1060).

Predictor variables	B	95% CI	β	t	R²	ΔR²	F
Step one					.04	.04	14.01**
Child gender	1.30	.77–1.9	.14	4.73***
Maternal age at pregnancy	−.10	−.16– −.04	−.10	−3.42***
Maternal education at pregnancy	.95	.33–1.58	.09	2.97***
Step two					.13	.12	12.30**
Child gender	1.27	.75–1.79	.14	4.74***
Maternal age at pregnancy	−.09	−.16– −.04	−.09	−3.05**
Maternal education at pregnancy	.73	.10–1.35	.07	2.28*
Exposure to intimate partner violence at 12 months	.59	−.23–1.41	.05	1.42
Maternal return to work or study at 12 months	.60	.04–1.15	.06	2.10*
Maternal confidence looking after baby at 12 months	.84	.21–1.47	.08	2.60**
4 year Maternal Mental Health	1.87	1.03–2.72	.13	4.34***
Maternal physical health at 4 years	−.03	−.06–.01	−.05	−1.62
Maternal involvement in home activities at 4 years	−.17	−.25– −.08	−.11	−3.67***
Exposure to IPV at 4 years	.87	.05–1.70	.07	2.08*
Maternal stressful life events at 4 years	.55	−.08–1.18	.05	1.71
Step three					.12	.01	8.15**
Child gender	1.29	.76–1.81	.14	4.78***
Maternal age at pregnancy	−.09	−.15– −.03	−.09	−3.08**
Maternal education at pregnancy	.74	.12–1.37	.07	2.32*
Exposure to intimate partner violence at 12 months	.71	−.68–2.09	.06	1.00
Maternal return to work or study at 12 months	.47	−.14–1.07	.05	1.52
Maternal confidence looking after baby at 12 months	.90	.21–1.59	.08	2.54*
Maternal mental health at 4 years	2.20	1.22–3.18	.16	4.41***
Maternal physical health at 4 years	−.01	−.04–.03	−.02	−.46
Maternal involvement in home activities at 4 years	−.17	−.26– −.07	−.11	−3.44***
Exposure to intimate partner violence at 4 years	.74	−.27–1.76	.06	1.44
Maternal stressful life events at 4 years	.65	−.05–1.34	.07	1.82
Exposure to intimate partner violence at 12 months X Maternal return to work or study 12 months	.96	−.56–2.49	.05	1.24
Exposure to intimate partner violence at 12 months X Maternal confidence looking after baby at 12 months	−.56	−2.20–1.09	−.03	−.66
Exposure to intimate partner violence at 12 months X maternal mental health at 4 years	−1.18	−3.13–.77	−.04	−1.18
Exposure to intimate partner violence at 12 months X Maternal physical health at 4 years	−.10	.01– −.18	−.08	−2.51*
Exposure to intimate partner violence at 12 months X maternal involvement in home activities at 4 years	.06	−.20–.31	.01	.42
Exposure to intimate partner violence 12 months X Exposure to intimate partner violence at 4 years	.63	−1.13–2.39	.04	.70
Exposure to intimate partner violence at 12 months X Maternal stressful life events at 4 years	−.75	−2.52–.84	−.05	−.93

Note. *p < .05, **p < .01, ***p < .001.

In step 2, the main effects of all predictor variables (i.e., child gender, maternal age, maternal education exposure to IPV at 12 months, maternal return to work or study maternal confidence looking after baby, maternal mental health, maternal physical health, maternal involvement in home activities exposure to IPV at 4 years, and maternal stressful life events) were entered into the model. The main effects model accounted for an additional 12% of variance in SDQ scores (R² = .13, F(8,1048) = 12.30, p < .001). The following main effects were significantly associated with SDQ scores: IPV exposure at 12 months postpartum, maternal mental health, maternal involvement in home activities and maternal stressful life events.

In step 3, interaction terms between exposure to IPV at 12 months and each of the predictor variables were also entered into the model. The addition of the interaction effects accounted for an additional 1% of the variance (R² = .12, F(7,1041) = 8.15, p < .01). The interactions between exposure to IPV at 12 months postpartum maternal physical health at four years postpartum was significantly associated with children’s SDQ scores (β = −.10, p = .012). Figure 1 displays the interaction effect between exposure to IPV at 12 months postpartum and maternal physical health. It reveals that for children exposed to IPV at 12 months postpartum, higher scores on maternal physical health (better physical health) was associated with lower SDQ scores.

Figure 1.

The Interaction Effect between IPV Exposure in the First 12 Months Postpartum and Maternal Physical Health at 4 Years Postpartum on Predicting Children’s Emotional-Behavioral Functioning at 4 Years of Age (n = 1060).

Discussion

Drawing on data from a large population-based sample of first-time mothers and their children, this study aimed to identify psychosocial factors associated with emotional-behavioral resilience in 4-year-old children exposed and not exposed to IPV within the first year of life. The results showed children who were exposed to IPV were more likely to experience emotional-behavioral difficulties than those children not exposed; however, 38% of these children displayed emotional-behavioral resilience at 4 years. This finding highlighted that despite exposure to IPV in early life when children’s brain and biological systems are highly vulnerable to “toxic stress,” many children overcome this adversity and maintain expected health and wellbeing.

Factors Associated with Resilience in Children Exposed to IPV

This study sought to identify psychosocial factors, uniquely associated with resilience in children exposed to IPV in early life. In this study no longer being exposed to IPV between 3 and 4 years was associated with emotional-behavioral resilience. For a child not to have ongoing exposure, a shift may have occurred within the family which is protective. This shift might represent the mother leaving the IPV relationship, the partner seeking support and/or access to other services. This finding reinforces the importance of early intervention to support families to make changes resulting in cessation or decreased exposure to IPV for women and their children.

Maternal physical wellbeing at 4 years was found to be protective for children exposed to IPV. This is a new contribution to the field. It is likely mothers who are physically well and have higher energy levels are more able to provide a consistent and nurturing home environment, and be more available to provide support to their child despite exposure to IPV.

Mothers’ return to work or study at 12 months postpartum was also associated with children’s emotional-behavioral resilience. Previous research helps clarify the protective role of women’s return to work. In women who have experienced IPV, employment has been linked with leaving violent relationships (Wilson, Baglioni, & Downing, 1989) and in some instances, reduced exposure for some children. However, it is important to note that leaving violent relationships can lead to an escalation of violence for many families (Campbell et al., 2003). Employment is also linked to increased self-esteem in women experiencing IPV (Lynch & Graham-Bermann, 2004). Consequently, mothers’ return to work or study might contribute to an increased sense of control over their lives and contribute positively to interactions with their child.

It is worth noting that in contrast to prior studies (Howell et al., 2010; Martinez-Torteya et al., 2009) the association between maternal mental health and children’s emotional-behavioral resilience did not reach statistical significance despite a large effect size. Given the common co-occurrence of IPV and psychological ill-health (Devries et al., 2013), it is likely adjusting for both factors in our analysis resulted in an over adjusted model. When we removed further exposure to IPV from the model, maternal mental was significant in predicting emotional-behavior resilience in children exposed to IPV at 12 months. This finding highlighted that although the impact of maternal depression on children’s development is well established (Goodman & Gotlib, 1999), the mechanisms by which good maternal health promotes children’s resilience in the face of IPV is less clear, and worthy of further investigation.

Factors Unique to Children Not Exposed to IPV Within the First Year Postpartum

This study identified several factors associated with positive outcomes for children not exposed to IPV including: maternal age in early pregnancy (≥ 25 years) and confidence looking after baby at 12 months postpartum. This suggests that although these factors are protective against emotional-behavioral difficulties in the broader population, exposure to IPV may disrupt this protective process.

Study Strengths

To our knowledge, this is the largest study to explore resilience in children exposed to emotional and physical IPV within a community sample. We have addressed gaps in the literature, which has focused primarily on families residing in domestic violence shelters and exposed to physical violence. The findings provide valuable insight into what contributes to resilience within the context of IPV exposure.

Another strength of our study is the acknowledgment of variable-centered approaches often taken in resilience research. We conducted a sensitivity analysis using multiple hierarchical regressions. In both approaches, maternal physical health was identified as protective factors for children exposed to IPV at 12 months postpartum. The person-centered approach also identified mothers’ return to work or study at 12 months postpartum and no further exposure to IPV between 3 and 4 years as protective, however these factors were not significant in our variable-centered approach. It is likely that this is due to our sample comprising a large number of children not exposed to IPV. Consequently, it is likely our model was driven by the strength of the associations of the larger group of children not exposed to IPV, rather than the smaller number of children exposed to IPV at each time point. Despite this difference, the results of the sensitivity analyses strengthens our confidence in the role of maternal physical health as being protective for emotional-behavioral functioning in children exposed to IPV. This sensitivity analysis highlights the need to further investigate the differences between person-centered and variable-centered approached to resilience research.

Study Limitations and Future Direction

It is important to note several limitations of this study. The definition of resilience was limited to emotional-behavioral resilience. We recognize resilience covers multiple domains of functioning and can change over time. The findings of this study reflect only resilience to emotional-behavioral difficulties. Future research investigating resilience across multiple domains throughout childhood would provide a more comprehensive insight into the development of resilience in children exposed to IPV.

Selective attrition was observed in the MHS. Consequently, the findings of this study may not be generalizable to families of lower socio-economic backgrounds or those experiencing a higher severity of violence.

Given the MHS was designed to investigate women’s health after childbirth and not specifically designed to examine factors associated with child resilience, we were limited in the factors available for the present analyses. Some items were brief and relied on parent-report which can be biased by the psychological functioning of mothers. There are also other factors such as child temperament, parenting behavior, school engagement, and access to services that are likely to be important in promoting resilience in children exposed to IPV that were not included in the study.

The current study represents an initial step in identifying psychosocial protective factors for children exposed to IPV within a large Australian cohort study. Complex models which incorporate mediation and moderating effects, as well as protective processes over time, would further advance the understanding of what assists children to maintain emotional-behavioral functioning when exposed to IPV.

Implications and Conclusion

This study provides a valuable contribution to the understanding of what assists young children to overcome adversity when exposed to IPV in early life. Our findings highlight many children who are exposed to IPV maintain positive emotional-behavioral wellbeing and that certain factors appear to assist with this process. Assisting mothers who have, or are experiencing IPV to access employment or further education may have long-term benefits for the wellbeing of their family. In addition, encouraging these mothers to prioritize their own health so they are in the best position to respond and interact with their children is an important part of promoting emotional-behavioral resilience in children exposed to IPV. Finally, our results highlighted the importance of early intervention efforts in our health care services that identify and support women to explore the options available to them. This may facilitate turning points within intimate partner relationships that eliminates or reduces IPV and its impact on children.

Footnotes

Authors’ note

All authors of this research have reviewed and significantly contributed to this paper.

This research has not been previously published nor is it under consideration for publication elsewhere.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The work was supported by the National Health and Medical Research Council (NHMRC), VicHealth, and the Victorian Government’s Operational Infrastructure Support Program. The corresponding author, Alison Fogarty, received support through an Australian Government Research Training Program Scholarship. Rebecca Giallo was supported by a NHMRC Career Development Fellowship and Stephanie Brown by a NHMRC Research Fellowship.

References

Australian Bureau of Statistics. (2007). Basic community profile: Victoria (Statistical Division). Retirieved from http://quickstats.censusdata.abs.gov.au/census_services/getproduct/census/2006/communityprofile/2?%20opendocument

Australian Domestic and Family Violence Clearinghouse. (2011). The Impact of Domestic Violence on Children: A Literature Review. Kensington, Australia: University of New South Wales. Retrieved from http://earlytraumagrief.anu.edu.au/files/ImpactofDVonChildren.pdf

Bowen

(2017). Conduct disorder symptoms in pre-school children exposed to intimate partner violence: Gender differences in risk and resilience. Journal of Child & Adolescent Trauma, 10, 97–107. doi:10.1007/s40653-017-0148-x

Bronfenbrenner

(1986). Ecology of the family as a context for human development: Research perspectives. Developmental Psychology, 22, 723–742. doi:10.1037/0012-1649.22.6.723

Brown

Lumley

McDonald

Krastev

(2006). Maternal health study: A prospective cohort study of nulliparous women recruited in early pregnancy. BMC Pregnancy and Childbirth, 6, 1–12. doi:10.1186/1471-2393-6-12

Campbell

J. C.

Webster

Koziol-McLain

Block

Campbell

Curry

M. A.

… Laughon

. (2003). Risk factors for femicide in abusive relationships: Results from a multisite case control study. American Journal of Public Health, 93, 1089–1097.

Chemtob

C. M.

Carlson

J. G.

(2004). Psychological effects of domestic violence on children and their mothers. International Journal of Stress Management, 11, 209–226. doi:10.1037/1072-5245.11.3.209

Cicchetti

(2013). Annual research review: Resilient functioning in maltreated children–past, present, and future perspectives. Journal of Child Psychology and Psychiatry, 54, 402–422.

Cohodes

Chen

Lieberman

A. L.

(2017). Maternal meta-emotion philosophy moderates effect of maternal symptomology on preschoolers exposed to domestic violence. Journal of Child Family Studies, 26, 1813–1843. doi:10.1007/s10826-017-0699-3

10.

Cox

J. L.

Chapman

Murray

Jones

(1996). Validation of the Edinburgh Postnatal Depression Scale (EPDS) in non-postnatal women. Journal of Affective Disorders, 39, 185–189. doi:10.1016/0165-0327(96)00008-0

11.

Cox

J. L.

Holden

J. M.

Sagovsky

(1987). Detection of postnatal depression. Development of the 10-item edinburgh postnatal depression scale. The British Journal of Psychiatry, 150, 782–786.

12.

Devries

K. M.

Mak

J. Y.

Bacchus

L. J.

Child

J. C.

Falder

Petzold

… Watts

C. H.

(2013). Intimate partner violence and incident depressive symptoms and suicide attempts: A systematic review of longitudinal studies. PLoS Med, 10, e1001439. doi:10.1371/journal.pmed.1001439

13.

Fergusson

D. M.

Horwood

L. J.

Ridder

E. M.

(2005). Show me the child at secen: The consequences of conduct problems in childhood for psychosocial functioning in adulhood. Journal of Child Psychology and Psychiatry, 46, 837–849. doi:10.1111/j.1469-7610.2004.00387.x

14.

Flach

Leese

Heron

Evans

Feder

Sharp

Howard

L. M.

(2011). Antenatal domestic violence, maternal mental health and subsequent child behaviour: A cohort study. BJOG: An International Journal of Obstetrics & Gynaecology, 118, 1383–1391. doi:10.1111/j.1471-0528.2011.03040.x

15.

Giallo

Gartland

Woolhouse

Mensah

Westrupp

Nicholson

Brown

(2018). Emotional-behavioural resilience among children of first-time mothers with and without depression across the early childhood period. International Journal of Behavioural Development, 42, 214–224. doi:10.1177/0165025416687413

16.

Gilbert

B. C.

Shulman

H. B.

Fischer

L. A.

Rogers

M. M

. (1999). The pregnancy risk assessment monitoring system (PRAMS): methods and 1996 response rates from 11 states. Maternal and Child Health Journal, 3, 199–209. doi:10.1023/A:1022325421844

17.

Goodman

(1997). The strengths and difficulties questionnaire: A research note. Journal of Child Psychology and Psychiatry, 38, 581–586. doi:10.1111/j.1469-7610.1997.tb01545.x

18.

Goodman

S. H.

Gotlib

I. H.

(1999). Risk for psychopathology in the children of depressed mothers: A developmental model for understanding mechanisms of transmission. Psychological Review, 106, 458–490. doi:10.1037/0033-295X.106.3.458

19.

Graham-Bermann

S. A.

Gruber

Howell

K. H.

Girz

(2009). Factors discriminating among profiles of resilience and psychopathology in children exposed to intimate partner violence (IPV). Child Abuse & Neglect, 33, 648–660. doi:10.1016/j.chiabu.2009.01.002

20.

Hawes

D. J.

Dadds

M. R

. (2004). Australian data and psychometric properties of the Strengths and Difficulties Questionnaire. Australian and New Zealand Journal of Psychiatry, 38, 644–651. Retreived from: http://web.b.ebscohost.com.ezproxy.lib.swin.edu.au/ehost/pdfviewer/pdfviewer?%20vid%20=%201&sid%20=%20c2800746-1bc6-4703-a4bb-43e49afc7966%40sessionmgr103

21.

Hegarty

Bush

Sheehan

. (2005). The composite abuse scale: Further development and assessment of reliability and validity of a multidimensional partner abuse measure in clinical settings. Violence and Victims, 20, 529–547. Retreived from: https://searchproquestcom.ezproxy.lib.swin.edu.au/docview/208530876?%20OpenUrlRefId%20=%20info:%20xri/sid:%20primo&accountid%20=%2014205

22.

Hegarty

Sheehan

Schonfeld

. (1999). A multidimensional definition of partner abuse: Development and preliminary validation of the Composite Abuse Scale. Journal of Family Violence, 14, 399–415. doi:10.1023/A:1022834215681

23.

Holmes

M. R.

(2013). The sleeper effect of intimate partner violence exposure: Long-term consequences on young children’s aggressive behavior. Journal of Child Psychology and Psychiatry, 54, 986–995. doi:10.1111/jcpp.12071

24.

Holt

Buckley

Whelan

(2008). The impact of exposure to domestic violence on children and young people: A review of the literature. Child Abuse & Neglect, 32, 797–810. doi:10.1016/j.chiabu.2008.02.004

25.

Hopkins

K. D.

Zubrick

S. R.

Taylor

C. L.

(2014). Resilience amongst australian aboriginal youth: An ecological analysis of factors associated with psychosocial functioning in high and low family risk contexts. PLoS One, 9, 22. doi:e102820

26.

Howell

K. H.

(2011). Resilience and psychopathology in children exposed to family violence. Aggression and Violent Behavior, 16, 562–569. doi:10.1016/j.avb.2011.09.001

27.

Howell

K. H.

Graham-Bermann

S. A.

Czyz

Lilly

(2010). Assessing resilience in preschool children exposed to intimate partner violence. Violence & Victims, 25, 150–164. doi:10.1891/0886-6708.25.2.150

28.

Kitzmann

K. M.

Gaylord

N. K.

Holt

A. R.

Kenny

E. D.

(2003). Child witnesses to domestic violence: A meta-analytic review. Journal of Consulting and Clinical Psychology, 71, 339. doi:10.1037/0022-006X.71.2.339

29.

Lupien

S. J.

McEwen

B. S.

Gunnar

M. R.

Heim

(2009). Effects of stress throughout the lifespan on the brain, behaviour and cognition. Nature Reviews Neuroscience, 10, 434–445. doi:10.1038/nrn2639

30.

Luthar

S. S.

Cicchetti

Becker

(2000). The construct of resilience: A critical evaluation and guidelines for future work. Child Development, 71, 543–562. doi:10.1111/1467-8624.00164

31.

Lynch

S. M.

Graham-Bermann

S. A.

(2004). Exploring the relationship between positive work experiences and women’s sense of self in the context of partner abuse. Psychology of Women, 28, 159–167. doi:10.1111/j.1471-6402.2004.00132.x

32.

Martinez-Torteya

Bogat

G. A.

von Eye

Levendosky

A. A.

(2009). Resilience among children exposed to domestic violence: The role of risk and protective factors. Child Development, 80, 562–577. doi:10.1111/j.1467-8624.2009.01279.x

33.

Masten

A. S.

Obradović

(2006). Competence and resilience in development. Annals of the New York Academy of Sciences, 1094, 13–27. doi:10.1196/annals.1376.003

34.

Miller

L. E.

VanZomeren-Dohm

Howell

K. H.

Hunter

E. C.

Graham-Bermann

S. A.

(2014). In-home social networks and positive adjustment in children witnessing intimate partner violence. Journal of Family Issues, 35, 462–480. doi:10.1177/0192513X13478597

35.

Miller-Lewis

L. R.

Searle

A. K.

Sawyer

M. G.

Baghurst

P. A.

Hedley

(2013). Resourse factors for mental health resilience in early childhood: An anlysis with multiple methologies. Child and Adolescent Psychiatry and Mental Health, 7. doi:10.1186/1753-2000-7-6.

36.

Murray

Cox

J. L.

(1990). Screening for depression during pregnancy with the Edinburgh Depression Scale (EDDS). Journal of Reproductive and Infant Psychology, 8, 99–107.

37.

National Center for Education Statistics. (2002). Early Childhood Longitudinal study (ECLS-K) Psychometric report for kindergarten through first grade (NCES 2002-5). Washington DC: U.S Department of Education.

38.

Pico-Alfonso

M. A.

Garcia-Linares

M. I.

Celda-Navarro

Blasco-Ros

Echeburúa

Martinez

(2006). The impact of physical, psychological, and sexual intimate male partner violence on women’s mental health: depressive symptoms, posttraumatic stress disorder, state anxiety, and suicide. Journal of Women’s Health, 15, 599–611. doi:10.5222/HEAD.2016.267

39.

Reiley

Davey

M. A.

King

(2005). Births in Victoria 2003-2004. Melbourne, Australia: Victorian Perinatal Data Collection Unit, Victorian Government Department of Human Services.

40.

Swartout

A. G.

Swartout

K. M.

White

J. W.

(2011). What your data didn’t tell you the first time around: Advanced analytic approaches to longitudinal analyses. Violence Against Women, 17, 309–321.

41.

Ware

J. E.

Gandek

(1998). Overview of the SF-36 health survey and the international quality of life assessment (IQOLA) project. Journal of Clinical Epidemiology, 51, 903–912. doi:10.1016/S0895-4356(98)00081-X

42.

Ware

Snow

Kosinski

Gandek

(1993). SF-36 General Health Survey: Manual and Interpretation Guide. Boston, MA: The Health Institute.

43.

Wilson

M. N.

Baglioni

A. J.

Jr Downing

(1989). Analyzing factors influencing readmission to a battered women’s shelter. Journal of Family Violence, 4, 275–284.