Trends in Hospital Admissions for Intimate Partner Violence in Australian Mothers With Children Born From 1990 to 2009

Abstract

This study aims to determine the prevalence, and trends over time, of Western Australian (WA) mothers who were victims of intimate partner violence (IPV) requiring hospital admission. The study investigated the prevalence of all mothers and the specific prevalence of Aboriginal and non-Aboriginal mothers. A population-based cohort study using de-identified linked health data of mothers of children born from 1990 to 2009 in WA was carried out. The prevalence of hospitalizations for IPV in mothers of children born in the period 1990-2009 (per 1,000 births) was calculated. Results indicate that the overall prevalence of hospital admissions for mothers assaulted 12 months prior to their child’s birth month increased in the period 1990-2009, from 2.7 to 7.7 per 1,000 births. There was also an increase in the overall prevalence of hospital admissions of mothers who were assaulted 12 months prior to the birth month and 36 months after the birth month, from 8.9 per 1,000 births in 1990 to 19.4 per 1,000 births in 2009. In addition, being Aboriginal, having a mother <30 years of age, and being of low SES significantly increased the odds of having a mother with an IPV admission. This study highlights that while there has been an increase in the prevalence of IPV admissions for mothers of children born from 1990 to 2009 in WA, the level of prevalence has remained persistent for the last decade for the whole population. However, non-Aboriginal mothers have seen an increase in prevalence in the last decade. This increase is associated with the introduction of the Z63.0 code in International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Australian Modification (ICD-10-AM). This study highlights the importance of prioritizing groups for targeted early intervention and prevention as well as the need for culturally appropriate strategies to reduce the burden of interpersonal violence.

Keywords

intimate partner violence family and domestic violence hospitalization Aboriginal linked population data

Introduction

Violence against women is a global social and public health issue (World Health Organization [WHO], 2002) of epidemic proportion (WHO, 2013). In Australia, as seen in many other countries, the most common perpetrator of assault against women is a current or previous spouse or domestic partner (Australian Institute of Health and Welfare [AIHW], 2018; WHO, 2013); this violence is termed as intimate partner violence (IPV). The WHO (2012) defines IPV as any behavior within an intimate relationship that causes physical, psychological, or sexual harm to those in the relationship. From a West Australian (WA) perspective, IPV has been defined as physical assault, sexual assault, verbal abuse, emotional/psychological abuse, social isolation, financial abuse, and spiritual abuse, or the threat of these acts, carried out by a current or former partner (Department of Child Protection and Family Support, 2015). While IPV can occur in same-sex relationships and be perpetrated by women against men, it is acknowledged as a gendered issue that is disproportionately perpetrated by men against women (Morgan & Chadwick, 2009; WHO, 2014). IPV occurs in all cultural, religious, and socioeconomic groups (WHO 2012), with approximately one in four Australian women experiencing IPV in their lifetime (Australian Bureau of Statistics [ABS], 2017).

The experience of IPV is not exclusively concurrent, and the impact may last a long time after the violence. For instance, it is acknowledged that women experiencing IPV have higher rates of adverse health outcomes. The women have an increased risk of physical health conditions such as chronic pain (Dillon, Hussain, Loxton, & Rahman, 2013), cardiovascular problems, and respiratory problems (Loxton, Schofield, Hussain, & Mishra, 2006). IPV also affects psychological health status, with women at increased risk of chronic episodic conditions such as depression (Vos et al., 2006), anxiety (Dillon et al., 2013), and substance misuse (Bacchus, Ranganathan, Watts, & Devries, 2018). An association between IPV and an increased risk of suicidal behaviors has also been acknowledged (Devries et al., 2011). Globally, lifetime exposure to violence is the second most common risk factor for disability-adjusted life years in young women (Mokdad et al., 2016).

Previous research has reported greater use of health services by women who are victims of violence (Ansara & Hindin, 2010; Black, 2011; Bonomi, Anderson, Rivara, & Thompson, 2009). In WA, population-based data have been used to investigate hospitalizations due to violence against women (Meuleners, Hendrie, & Lee, 2008; Meuleners, Lee, Hendrie, & Fraser, 2010; Meuleners, Lee, Janssen, & Fraser, 2011). Meuleners et al. (2008) determined that women of childbearing age were more likely to be hospitalized for assault than women of non-childbearing age. Despite a greater use of health services, women are reluctant to disclose they are victims of IPV to health professionals, for a multitude of reasons such as stigma, fear, and lack of a safe place to disclose (Garcia-Moreno, Jansen, Ellsberg, Heise, & Watts, 2005; Keeling & Fisher, 2015).

While researchers disagree about the prevalence of IPV during pregnancy, past research has shown that this time period is associated with an increased risk of IPV for some women (Brownridge et al., 2011). In Australia, nearly half (48%) of women who reported violence in a relationship had experienced violence during pregnancy (ABS, 2017). IPV in pregnancy is a particular concern as the woman is at risk of both fatal and nonfatal adverse health outcomes. IPV exposure has been associated with delayed entry to antenatal care (Islam, Broidy, Baird, & Mazerolle, 2017) and increased risk of adverse birth outcomes (Coker, Sanderson, & Dong, 2004; Heaman, 2005). The direct trauma of violence to the pregnant woman’s body through physical assault to the abdomen or sexual trauma can increase the risk of preterm birth and low birthweight (Campbell, 2002), spontaneous abortion, or neonatal death (Coker et al., 2004). Adverse birth outcomes may also be compounded by the behavior of the mother in response to violence such as poor nutrition or missing prenatal health care appointments (Coker et al., 2004).

The 2016 Australian Personal Safety Survey (ABS, 2017) reported that of women who were victims of assault, in 65% of cases the most recent incident was most likely to be in their home. Furthermore, 49% of women assaulted had children in their care.

Gartland et al.’s (2014) study of Australian first-time mothers found that 20% had been victims of IPV in the first postpartum year, rising to 29% by the time the child was 4 years old. Due to the time they spend in the family home, younger children are particularly vulnerable and disproportionately exposed to IPV (Fantuzzo & Fusco, 2007; Gjelsvik, Verhoek-Oftedahl, & Pearlman, 2003). Substantial research has found that when mothers are victims of IPV, their children can be adversely affected (Holt, Buckley, & Whelan, 2008). Children exposed to IPV have a higher risk of poorer outcomes in areas of physical health, psychological, emotional, behavioral, social, and academic achievements (Artz et al., 2014; Howell, Barnes, Miller, & Graham-Bermann, 2016).

Aboriginal and Torres Strait Islander (hereafter referred to as Aboriginal) women have been identified in previous research as having a higher risk of being a victim of violence than non-Aboriginal women (Meuleners et al., 2010; Meuleners et al., 2011). Aboriginal women are 32 times more likely to be hospitalized due to violence than non-Aboriginal women (Steering Committee for the Review of Government Service Provision, 2016). Higher levels of violence in Aboriginal communities than in the general population are acknowledged as both a cause and an effect of social disadvantage and intergenerational trauma (AIHW, 2018), attributed to the legacy of colonization. Furthermore, demographic groups who have been identified at higher risk of IPV are socioeconomically disadvantaged women (Khalifeh, Hargreaves, Howard, & Birdthistle, 2013; WHO, 2004) and younger women (ABS, 2017).

In conjunction with the community, the Australian Commonwealth, State and Territory Governments developed a 12-year “National Plan to Reduce Violence against Women and their Children 2010-2022” (the National Plan; Australian Government, 2010). The National Plan is now in its third stage. The plan stresses the need for research to fill gaps and increase an understanding of health issues experienced by women and children exposed to IPV.

Given that women of childbearing age are more likely to be admitted to hospital due to assault, an understanding of the prevalence of assault hospitalizations of mothers is important. Therefore, this study used administrative data to gain an understanding of the prevalence of IPV in both Aboriginal and non-Aboriginal mothers, and patterns over time. As pregnancy has been identified as a period of heightened risk of IPV (Devries et al., 2010) and the potential impact of IPV on young children, we focused on the key stages of 12 months prior to birth and up to 36 months of age.

Method

Study Design

A retrospective population-based cohort study was undertaken that included all mothers of children born in WA in the period 1990-2009. We used de-identified linked data from the WA Data Linkage System. Data were extracted from the Midwives Notification System (MNS), Birth Registry, and Hospital Morbidity Data Collection (HMDC), all statutory data collections. Best-practice probabilistic matching (Kelman, Bass, & Holman, 2002) was used for linkage across the data sets. The MNS contains data of all babies born >400 g in weight or of at least 20 weeks’ gestation. This study used MNS and the Birth Registry from 1990 to 2009. The HMDC includes all separations from public and private hospitals. To capture admissions 12 months prior to birth and 36 months after birth, we used HMDC data from 1989 to 2012. The HMDC contains diagnosis and external cause codes according to the International Statistical Classification of Diseases and Related Health Problems, Ninth Revision (ICD-9: 1979 to June 1999; International Statistical Classification of Diseases and Related Health Problems, Tenth Revision [ICD-10]: July 1999 to present).

IPV Categories

Mothers who had an IPV hospitalization were principally identified through ICD codes (Table 1) used in previous studies, which capture the external cause of admissions due to assault and abuse (Meuleners et al., 2010; Meuleners et al., 2011). While assault-related ICD codes do not specifically indicate the relationship to the perpetrator, there are specific Z codes related to partner relationships. These codes represent factors that influence a person’s health state, and we used the Z code relating to partner conflict that affects health status: Z63.0 and its predecessor V61.1. The Z63.0 code captures victimization of multiple forms of violence, which are important to the understanding of hospitalizations due to IPV (Olive, 2018). We focused on hospitalizations of mothers in two time periods: (a) those occurring 12 months prior to birth, and (b) those occurring 12 months prior to birth and up to 36 months after birth. For both groups, classifications were grouped into the following:

Assault (Assault) included homicide, purposeful inflicted injury, and physical, emotional, psychological, and sexual abuse (ICD-9: E960 to E969, 995.81, V71.5; International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Australian Modification [ICD-10-AM]: T74.1, T74.2, T74.3, and Z04.4, X85 to Y09).

Partner Conflict Impacting Health Status (Conflict) included counseling for marital and partner problems and problems in relationship with spouse or partner that influence health status (ICD-9: V61.1; ICD-10-AM: Z63.0—Discord between partners resulting in severe or prolonged loss of control, in generalization of hostile or critical feelings, or in a persisting atmosphere of severe interpersonal violence [hitting or striking]). Z codes are acknowledged as a conceptually discrete category (Olive, 2018; WHO, 2016); therefore, it was important to acknowledge them in their own specific category.

All IPV (All) included all codes for Assault and Conflict and additional codes for external cause of injury and poisoning, including undetermined intent (ICD-9: E980 to E989 and ICD-10-AM: Y10 to Y34).

Table 1.

Categories With ICD Code Definitions.

ICD Code	Definition	ICD
Assault
E960-E969	Homicide and injury purposely inflicted by other persons (E960-E969)	ICD-9
995.81	Adult physical abuse	ICD-9
V71.5	Observation following alleged rape or seduction Examination of victim or culprit	ICD-9
T74.1	Physical abuse	ICD-10-AM
T74.2	Sexual abuse	ICD-10-AM
T74.3	Psychological abuse	ICD-10-AM
Z04.4	Examination and observation following alleged rape or seduction Examination of victim or culprit following alleged rape or seduction	ICD-10-AM
X85-Y09	Assault	ICD-10-AM
Partner Conflict Impacting Health Status (Conflict)
V61.1	Counseling for marital and partner problems	ICD-9
Z63.0	Problems in relationship with spouse or partner (Discord between partners resulting in severe or prolonged loss of control, in generalization of hostile or critical feelings, or in a persisting atmosphere of severe interpersonal violence [hitting or striking])	ICD-10-AM
All IPV (All)
E980-E989	Injury undetermined whether accidentally or purposely inflicted	ICD-9
Y10-Y34	Event of undetermined intent	ICD-10-AM

Note. ICD-9 = International Statistical Classification of Diseases and Related Health Problems, Ninth Revision; ICD-10-AM = International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Australian Modification; IPV = intimate partner violence.

For inclusion in the cohort, the mothers were required to have one or more of the ICD codes in one of the three categories. Each mother with a hospitalization for IPV was counted once in each category, even if they had more than one hospitalization. The mothers could be in one or more of the classification groups: Assault, Conflict, and All.

Neighborhood-level socioeconomic status (SES) was determined by the Index of Relative Social Disadvantage from the ABS (2013b) using the Birth and Midwives data. Five levels of disadvantage were assigned to census collection districts (∼200 households), ranging from 1 (high disadvantage) to 5 (low disadvantage). Aboriginal mothers were identified using Birth and Midwives data. Aboriginal Australians have higher levels of socioeconomic disadvantage compared with non-Aboriginal Australians (Australian Government, 2014). Therefore, we stratified our analysis by Aboriginality.

Ethics

Ethics approval for this study was obtained from the WA Department of Health Human Research Ethics Committee (#2012/37), the WA Aboriginal Health Ethics Committee, and the University of Western Australia Human Research Ethics Committee.

Statistical Analysis

The prevalence of hospitalizations for IPV in mothers of children born in the period 1990-2009 (per 1,000 births) was calculated for each birth year using the number of children with mothers who had a hospital admission for each birth year as the numerator and total births as the denominator. As rates of violence differ between Aboriginal and non-Aboriginal women, analyses were stratified by ethnicity. For analysis by Aboriginality, we calculated the prevalence using Aboriginal mothers who had a hospital admission for each birth year as the numerator and the total population of mothers with Aboriginal children as the denominator. This was replicated for non-Aboriginal mothers. Separate analysis was conducted for the three categories of Assault, Conflict, and All. Time trends in IPV admissions were evaluated using Poisson regression (Kuhn, Davidson, & Durkin, 1994). All trends are presented as effect sizes with 95% confidence intervals (CIs) and associated p values (statistically significant at p < .05 level). As a cautionary note, some of the ICD codes are rarely used and there may be an effect that is undetected due to low sample size (O’Donnell et al., 2017). Therefore, consideration of effect sizes and CI of trends should be given. Due to the significant increase in admissions in 1999, corresponding with the implementation of ICD-10-AM coding, we analyzed the data with a break in series, grouped into two periods 1990-1998 and 2000-2009 to determine whether there was an upward or downward trend in each series. Odds ratios and 95% CIs for maternal hospitalizations were calculated for Aboriginality, maternal age, and SES using logistic regression. Analyses were stratified by birth year group (1990-1994, 1995-1999, 2000-2004, and 2005-2009). All analyses were conducted using SAS^® statistical software (SAS Institute Inc., 2013).

Results

Of the 524,534 live births in WA during 1990-2009, 3,065 (0.6%) children were born to mothers admitted to hospital 12 months prior to their child’s birth in one of the three classification groups: Assault, Conflict, and All. This number increased to 8,368 (1.6%) when admissions 12 months before birth to 36 months after birth were included.

12 Months Prior to Birth

The overall prevalence of hospital admissions increased from 1990 to 2009 for all three categories (Figure 1). All increased from 2.7 to 7.7 per 1,000 births from 1990 to 2009, respectively, whereas Assault increased from 2.7 to 3.8 per 1,000 births over this period. No admissions were recorded for Conflict prior to 1999, but from 1999 there was an increase from 3.7 to 4.2 per 1,000 births. Hospital admissions, for the All category in the 12 months prior to birth, for births from 1990 to 1998 increased by 6.4% per year (95% CI: [3.4%, 9.4%], p < .001). However, the trend did not continue in the second series (2000-2009) where there was a nonsignificant decrease (incidence rate ratios [IRR]: –0.5%, 95% CI: [−2.0%, 1.1%], p = .53). The Assault category also had significant increases in 1990-1998 by 5.9% per year (95% CI: [2.7%, 9.2%], p < .001). Like the All category, this trend did not continue into the second series, where there was a significant decrease of –2.9% per year (95% CI: [−4.5%, −1.2%], p < .001). The Conflict trend data are only available from 2000 onward where there was a nonsignificant increase per year (IRR: 1.8%, 95% CI: [−1.3%, 4.9%], p = .26).

Figure 1.

Prevalence of children having a mother admitted to hospital 12 months prior to birth (all births, WA, 1990-2009).

12 Months Prior to Birth and 36 Months After Birth

When the analysis was expanded to include admissions from 12 months prior to 36 months after birth, the overall prevalence of hospital admissions increased between 1990 and 2009 for all three categories (Figure 2). The All category increased from 8.9 per 1,000 births in 1990 to 19.4 per 1,000 births in 2009. Assault increased from 8.7 per 1,000 births to 10.8 per 1,000 births in 2009. There were no admissions for Conflict until 1996, after which an increase from 0.4 to 9.5 per 1,000 births in 2009 was observed. Within the Conflict category, the Z63.0 code, introduced in ICD-10-AM, was associated with psychological admissions, 16.9% had an acute stress reaction at admission, 13.9% had a depressive episode, and 8.2% had poisoning by benzodiazepines, antidepressants, or non-opioid analgesics.

Figure 2.

Prevalence of children having a mother admitted to hospital 12 months prior to birth and 36 months after birth (all births, WA, 1990-2009).

The All category had a significant increase in admissions of 7.1% (95% CI: [4.8%, 9.4%], p < .001) per year from 1990 to 1998. This trend did not continue in the second series with a nonsignificant decrease (IRR = −0.1%, 95% CI: [−0.6%, 0.5%], p = .75). Similarly, the Assault category had a significant increase in the first series (1990-1998) of 2.8% (95% CI: [1.7%, 4.0%], p < .001) per year but did not continue in the second series with a significant decrease of −1.9% (95% CI: [−3.2%, −0.6%], p = .006) per year. The Conflict category only commenced being used from 1996 and had a substantial significant increase from 1996 to 1998 of 251% (95% CI: [156%, 480%], p < .001) per birth year. The second series had a nonsignificant increase (IRR = 1.6%, 95% CI: [−0.1%, 3.3%], p = .07).

Factors Associated With a Child Born to a Mother With Admission

The results of multiple logistic regression analysis for factors associated with a child born to a mother with an admission between 12 months prior to birth and first 36 months after birth are shown in Table 2. Being Aboriginal, <30 years of age, and of low SES significantly increased the odds of having a mother with an admission. Importantly, the risk of Aboriginal children being born to a mother with a hospital admission (Table 2) dropped considerably from 39.6% (95% CI: [33.1, 45.0]) in the period 1990-2004 to 8.7% (95% CI: [7.9, 9.6]) in the period 2005-2009.

Table 2.

Factors Associated With a Child Born to a Mother With a Hospital Admission 12 Months Prior to Birth and First 36 Months of Life (All Births, WA, 1990-2009).

	Birth Year (OR, 95% CI)
	1990-1994	1995-1999	2000-2004	2005-2009
Aboriginality
Aboriginal	38.6 [33.1, 45.0]	20.4 [18.2, 23.0]	10.8 [9.8, 11.9]	8.7 [7.9, 9.6]
Non-Aboriginal	Reference	Reference	Reference	Reference
Maternal age (years)
<20	2.1 [1.7, 2.6]	1.7 [1.5, 2.0]	2.0 [1.7, 2.3]	2.2 [1.9, 2.5]
20-29	1.6 [1.3, 1.9]	1.5 [1.1, 1.7]	1.8 [1.6, 1.9]	1.7 [1.5, 1.8]
30-39	Reference	Reference	Reference	Reference
>39	0.7 [0.3, 1.9]	0.8 [0.4, 1.4]	1.0 [0.7, 1.4]	1.0 [0.8, 1.4]
Socioeconomic level
1—high disadvantage	2.2 [1.8, 2.7]	1.8 [1.6, 2.2]	2.0 [1.7, 2.2]	1.9 [1.7, 2.1]
2	1.3 [1.0, 1.6]	1.4 [1.2, 1.6]	1.5 [1.3, 1.8]	1.5 [1.3, 1.7]
3	1.5 [1.2, 1.9]	1.1 [1.0, 1.4]	1.3 [1.1, 1.5]	1.3 [1.2, 1.5]
4	1.3 [1.1, 1.6]	1.0 [0.8, 1.2]	1.1 [0.9, 1.2]	1.1 [0.9, 1.2]
5—low disadvantage	Reference	Reference	Reference	Reference

Note. WA = Western Australia; OR = odds ratio; CI = confidence interval.

Aboriginality

When we analyzed the data by Aboriginality, the prevalence in each of the three categories was substantially higher in Aboriginal mothers than non-Aboriginal mothers. This disparity continued through the 20-year observation period (Figure 3a and 3b). In 1990, the prevalence of All in Aboriginal mothers was 112.56 per 1,000 births compared with 2.26 per 1,000 births of non-Aboriginal mothers. By 2009, the prevalence had increased in both Aboriginal and non-Aboriginal to 152.20 per 1,000 births and 11.73 per 1,000, respectively. Focusing on the endpoints of the two series, we see that the prevalence rates of Conflict in Aboriginal mothers climbed from 22.27 per 1,000 births in 1998 to 35.08 per 1,000 births in 2009, a 35% increase. However, the prevalence for Conflict in non-Aboriginal mothers climbed 100% from 3.98 per 1,000 births in 1998 to 8.01 per 1,000 births in 2009. Again, focusing on the endpoints, the Assault category in Aboriginal mothers was 131.19 per 1,000 births in 1998 and reduced marginally to 129.01 per 1,000 births in 2009. However, the non-Aboriginal Assault category increased from 2.88 per 1,000 births in 1998 to 3.99 per 1,000 births in 2009.

Figure 3.

Prevalence of (a) Aboriginal and (b) non-Aboriginal children having a mother admitted to hospital 12 months prior to birth and 36 months after birth (all Aboriginal births, WA, 1990-2009).

Discussion

This is the first time that linked population data have been used to determine the prevalence of hospital admissions for IPV, against mothers of children born in WA. Importantly, analysis of the data has given us a new understanding of IPV trends over time for both Aboriginal and non-Aboriginal mothers.

Our findings suggest that, overall, there has been an increase in the prevalence of IPV admissions for mothers of children born from 1990 to 2009 in WA, with the increased level of prevalence remaining persistent for the last decade. These findings are in contrast to the 2012 Australian Personal Safety Survey (ABS, 2013a) which reported a decrease in women experiencing violence from 1996 to 2012. A possible explanation for the disparity is that our study focused specifically on mothers and not women. Furthermore, our cases are women who have been hospitalized and do not necessarily represent victims of IPV who are not hospitalized.

The overall trends highlight a significant increase in All admissions in 2000. A possible explanation, in part, for the increase could be the introduction of the decade-long “Freedom from Fear” campaign that launched in 1998. This campaign raised awareness of IPV through television, radio, press, and outdoor advertising. The raised awareness in the general public is likely to flow through to health care professionals, leading to better recording of IPV.

A further possible explanation for the increase in admissions in 2000 is the implementation of the ICD-10-AM codes. The implementation had a significant impact on the Conflict category, specifically with the introduction of the Z63.0 code. To accommodate this increase, we analyzed trends pre-1999 and post-1999, finding a marked difference. In the 12 months prior to birth admissions in the All category, there was a significant increase pre-1999 and some volatility post-1999, resulting in a nonsignificant decrease. The dissimilarity was also reflected in 12 months prior to birth and 36 months after birth All admissions. These findings were unexpected but could be a reflection that the practice of coders may have changed with the implementation of ICD-10. We therefore recommend that trends continue to be monitored to inform service providers and clinicians and to support implementation of evidence-based interventions. The increase in prevalence in the Conflict category appears to be directly linked to the introduction of the Z63.0 code. The previous related ICD-9 code was not widely used, and as it was developed in the 1970s, it has become less relevant (Lalonde & Taylor, 1997) with new understandings of IPV. While Z codes have been described as general and overly vague (Whisman & Beach, 2013), the use of Z63.0 provided the ability to analyze the psychological harm of IPV that is difficult to capture, such as stress and depressive disorders. Our use of separate categories for Assault and Conflict demonstrated similar trends in both these categories post-1999 (Figures 1 and 2). Therefore, we recommend that future research on IPV utilizes Z codes to capture psychological harm and also investigate the reliability of these codes.

Characteristics that were associated with mothers being hospitalized included being <30 years of age, low SES, and Aboriginal. Our findings that maternal age <30 years was associated with increased risk of being hospitalized are consistent with the Australian Personal Safety Study (ABS, 2013a) where young women (18-24 years) reported higher rates of violence than other age groups. Therefore, mothers <30 years of age continue to be an important target group for violence prevention and intervention strategies.

Our study found that being from a low socioeconomic background was associated with an increased risk of a mother being hospitalized. This differs from an Australian population-based study (Meuleners et al., 2008) which found hospitalizations for violence were evenly distributed in all SES groups, although differences in readmission rates within SES groups were noted. However, the Meuleners et al. study included males and females, and given differences in gender experiences of victimization, this could explain inconsistencies between respective results. Due to the multidimensional nature of SES, the association between low socioeconomic household status and increased risk of interpersonal violence, including IPV, has not been reflected within all literature (Taillieu & Brownridge, 2010) and requires further research.

We found that Aboriginal mothers are at higher risk of hospital admissions, which is consistent with previous research that Aboriginal people are disproportionally affected by IPV (ABS, 2016). Reducing violence against Aboriginal women is a priority in the National Plan (Australian Government, 2010).

An unanticipated finding was the increasing prevalence of the Conflict category in non-Aboriginal mothers compared with Aboriginal mothers. This could be partially explained by the Z63.0 code. The code focuses on intimate partner relationships (spouse/partner), which does not take account of other familiar relationships that are included in Aboriginal understandings of “family and domestic violence” and included as part of the formal response to the issue in WA. As such, behaviors included in Z63.0 code may be too narrow to reflect the broader experiences of Aboriginal people (Olsen & Lovett, 2016). In addition, it has been acknowledged that Aboriginal women are less likely to disclose partner violence than non-Aboriginal women (Willis, 2011). The women’s decision not to disclose partner conflict could be attributed to the lack of cultural safety, for example, lack of Aboriginal health workers or culturally trained non-Aboriginal staff, and lack of trust due the absence of a relationship with their health care provider (Spangaro et al., 2016). There may also be distrust of authority figures due to past negative experiences, including racism and forced removal of children (Willis, 2011).

Our findings highlight an association with Z63.0 and psychological admissions. Therefore, it is important that mental health–related admissions for Aboriginal mothers take into account whether family violence is a factor which would benefit clinicians and service providers in devising appropriate intervention and support programs. Furthermore, the Z63.0 admissions may provide an opportunity to identify and support all women who are not necessarily physically assaulted but have had their health status affected by their partners behavior.

When we analyzed hospitalizations by Aboriginality, we found, for children born in 2009, that Aboriginal mothers were more than 32 times more likely to be admitted for an Assault category compared with non-Aboriginal mothers. These findings are in keeping with the existing literature that highlights Aboriginal women are 32 times more likely to be hospitalized due to violence than non-Aboriginal women (Steering Committee for the Review of Government Service Provision, 2016). While there is a disproportionate burden of violence against Aboriginal mothers, it is important to acknowledge that violence is not customary in Aboriginal culture (Olsen & Lovett, 2016). Higher rates have been attributed to intergenerational trauma and the social disadvantage caused by the legacy of colonization (AIHW, 2018). The National Plan (Australian Government, 2010) acknowledges that Aboriginal women are more likely to experience violence than non-Aboriginal women and highlights the need for interventions to be place-based and developed in partnership with communities. In highlighting the prevalence differences between Aboriginal and non-Aboriginal mothers, our research extends the knowledge of violence experienced by Aboriginal mothers and can contribute to collaborative interventions that strengthen Aboriginal communities and prevent violence.

Although we have addressed them where possible, there are some limitations with the study. First, while most violence against women is perpetrated by a male intimate partner, we acknowledge that some of the cases may have been perpetrated by someone other than a male intimate partner. Second, data are reliant on clinical coding of hospital morbidity records, and an undetermined intent code is utilized where information is insufficient to confirm assault, self-harm, or accidental injury. Third, HMDC data were only available for women who were admitted to hospital; therefore, the data are an underestimation of IPV in mothers and not generalizable to women who did not require hospitalization or report their assault. Future research should utilize linked data sets from Emergency Departments and/or Police records to improve ascertainment of IPV.

The National Plan (Australian Government, 2010) recognized that health professionals are often an early point of contact for women who are victims of IPV. Therefore, the hospitalization of mothers offers an opportunity for health professionals to provide targeted interventions. It is important that the complexities of overrepresentation of Aboriginal women are acknowledged and interventions are developed with partnerships between health services and Aboriginal communities. As the first response is pivotal to women’s safety and support, it is essential that health care professionals are educated on violence against women and the consequences of such violence on not only the women but also their children. To support holistic care, health professionals need to be aware of appropriate services to which they can refer mothers and families. It is important to acknowledge that due to the hidden nature of IPV, mothers may not have disclosed that they are victims of IPV. Therefore, targeted screening using an appropriate tool for women who are hospitalized for assault and partner discord/conflict should be utilized to ensure that mothers and children have appropriate care and support. Despite the limitations, the use of population-level data reduces the difficulties associated with research methods where mothers may be reluctant to disclose that they are a victim of IPV due to stigma (Murray, Crowe, & Akers, 2016). These findings should be used to highlight priority groups for targeted early intervention and prevention as well as the need for culturally appropriate strategies to reduce the burden of IPV. While this study has described the prevalence and trends in IPV admissions in mothers, further research is required to investigate the impact this has on their children’s health and well-being.

Footnotes

Acknowledgements

We acknowledge the partnership of the Western Australian Government Departments of Health, Communities, Education, and Justice which provided data for this project. We thank the Data Linkage Branch of the Department of Health for linking the data. The views expressed are not necessarily those of the Government Departments that have provided data for this project, and any errors of omission or commission are the responsibility of the researchers.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Melissa O’Donnell was supported by a National Health and Medical Research Council Early Career Fellowship (1012439). This research was also supported by an Australian Research Council Linkage Project Grant (LP100200507) and an Australian Research Council Discovery Grant (DP110100967). Carol Orr was supported with a Research Training Program (RTP) Stipend Scholarship, The University of Western Australia Safety Net Top-Up Scholarship, and Research Training Program (RTP) Fees Offset Scholarship.

ORCID iD

Carol Orr

Author Biographies

Carol Orr has an MSc in public health and is a PhD candidate at The University of Western Australia. She is also a teaching academic at the School of Nursing & Midwifery at The University of Notre Dame, Australia. Her research focuses on the outcomes of children exposed to interpersonal and intimate partner violence.

David Preen has a PhD and is the chair of public health at the School of Population and Global Health, The University of Western Australia. He has been involved with conducting health services research using population-based linked data for over 15 years to study areas, including (a) cancer service delivery, (b) pharmacoepidemiology, (c) chronic disease management, (d) morbidity of marginalized populations, and (e) methodological advances using medical record linkage.

Colleen Fisher has a PhD, and she is the head of school in the School of Population and Global Health, The University of Western Australia. Her research expertise is in the areas of women’s health, inequality, and family and domestic violence. Her research includes intervention and program evaluations as well as research across the life span and cross-culturally—including in refugee and international contexts.

Scott Sims has a master’s degree in biostatistics and works as a biostatistician at the Telethon Kids Institute. His current research focuses on spatial epidemiology and alcohol-related harm.

Melissa O’Donnell has a PhD and is a National Health and Medical Research Council (NHMRC) research fellow at the Telethon Kids Institute, The University of Western Australia. Her research focuses on the area of child abuse and neglect and utilizes linked government data.

References

Ansara

Hindin

(2010). Formal and informal help-seeking associated with women’s and men’s experiences of intimate partner violence in Canada. Social Science & Medicine, 70, 1011-1018. doi:10.1016/j.socscimed.2009.12.009

Artz

Jackson

Rossiter

Nijdam-Jones

Géczy

Porteous

(2014). A comprehensive review of the literature on the impact of exposure to intimate partner violence for children and youth. International Journal of Child, Youth & Family Studies, 5, 493-587. doi:10.18357/ijcyfs54201413274

Australian Bureau of Statistics. (2013a). Personal Safety Survey Australia, 2012 (ABS Cat. No. 4906.0). Available from http://www.abs.gov.au

Australian Bureau of Statistics. (2013b). Socio-Economic Indexes For Areas. Retrieved from http://www.abs.gov.au/websitedbs/censushome.nsf/home/seifa

Australian Bureau of Statistics. (2016). Recorded crime—Victims, Australia, 2015 (ABS Cat. No. 4510.0). Available from http://www.abs.gov.au

Australian Bureau of Statistics. (2017). Personal safety, Australia, 2016 (ABS Cat. No 4906.0). Available from http://www.abs.gov.au

Australian Government. (2010). The national plan to reduce violence against women and their children 2010-2022. Available from https://www.dss.gov.au

Australian Government. (2014). Aboriginal and Torres Strait Islander health performance framework 2014 report. Available from https://www.pmc.gov.au

Australian Institute of Health and Welfare. (2018). Family, domestic and sexual violence in Australia (Cat. No. FDV 2). Canberra, Australia: Author.

10.

Bacchus

Ranganathan

Watts

Devries

(2018). Recent intimate partner violence against women and health: A systematic review and meta-analysis of cohort studies. BMJ Open, 8(7), e019995. doi:10.1136/bmjopen-2017-019995

11.

Black

(2011). Intimate partner violence and adverse health consequences. American Journal of Lifestyle Medicine, 5, 428-439. doi:10.1177/1559827611410265

12.

Bonomi

Anderson

Rivara

Thompson

(2009). Health care utilization and costs associated with physical and nonphysical-only intimate partner violence. Health Services Research, 44, 1052-1067. doi:10.1111/j.1475-6773.2009.00955.x

13.

Brownridge

Taillieu

Tyler

Tiwari

Chan

Santos

S. C.

(2011). Pregnancy and intimate partner violence: Risk factors, severity, and health effects. Violence Against Women, 17, 858-881. doi:10.1177/1077801211412547

14.

Campbell

(2002). Health consequences of intimate partner violence. The Lancet, 359, 1331-1336. doi:10.1016/S0140-6736(02)08336-8

15.

Coker

Sanderson

Dong

(2004). Partner violence during pregnancy and risk of adverse pregnancy outcomes. Paediatric and Perinatal Epidemiology, 18, 260-269. doi:10.1111/j.1365-3016.2004.00569.x

16.

Department for Child Protection and Family Support. (2015). What is family and domestic violence. Perth, Australia: Author.

17.

Devries

Kishor

Johnson

Stöckl

Bacchus

Garcia-Moreno

Watts

(2010). Intimate partner violence during pregnancy: Analysis of prevalence data from 19 countries. Reproductive Health Matters, 18(36), 158-170. doi:10.1016/S0968-8080(10)36533-5

18.

Devries

Watts

Yoshihama

Kiss

Schraiber

Deyessa

. . . Garcia-Moreno

(2011). Violence against women is strongly associated with suicide attempts: Evidence from the WHO multi-country study on women’s health and domestic violence against women. Social Science & Medicine, 73, 79-86. doi:10.1016/j.socscimed.2011.05.006

19.

Dillon

Hussain

Loxton

Rahman

(2013). Mental and physical health and intimate partner violence against women: A review of the literature. International Journal of Family Medicine, 2013, 313909. doi:10.1155/2013/313909

20.

Fantuzzo

Fusco

(2007). Children’s direct exposure to types of domestic violence crime: A population-based investigation. Journal of Family Violence, 22, 543-552. doi:10.1007/s10896-007-9105-z

21.

Garcia-Moreno

Jansen

H. A.

Ellsberg

Heise

Watts

(2005). WHO multi-country study on women’s health and domestic violence against women: Initial results on prevalence, health outcomes and women’s responses. Geneva, Switzerland: World Health Organization.

22.

Gartland

Woolhouse

Mensah

Hegarty

Hiscock

Brown

(2014). The case for early intervention to reduce the impact of intimate partner abuse on child outcomes: Results of an Australian cohort of first-time mothers. Birth, 41, 374-383. doi:10.1111/birt.12123

23.

Gjelsvik

Verhoek-Oftedahl

Pearlman

(2003). Domestic violence incidents with children witnesses: Findings from Rhode Island surveillance data. Women’s Health Issues, 13, 68-73. doi:10.1016/S1049-3867(02)00197-4

24.

Heaman

(2005). Relationships between physical abuse during pregnancy and risk factors for preterm birth among women in Manitoba. Journal of Obstetric, Gynecologic & Neonatal Nursing, 34, 721-731. doi:10.1177/0884217505281906

25.

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

26.

Howell

Barnes

Miller

Graham-Bermann

(2016). Developmental variations in the impact of intimate partner violence exposure during childhood. Journal of Injury & Violence Research, 8, 43-57. doi:10.5249/jivr.v8i1.663

27.

Islam

M. J.

Broidy

Baird

Mazerolle

(2017). Exploring the associations between intimate partner violence victimization during pregnancy and delayed entry into prenatal care: Evidence from a population-based study in Bangladesh. Midwifery, 47, 43-52. doi:10.1016/j.midw.2017.02.002

28.

Keeling

Fisher

(2015). Health professionals’ responses to women’s disclosure of domestic violence. Journal of Interpersonal Violence, 30, 2363-2378.

29.

Kelman

Bass

Holman

(2002). Research use of linked health data—A best practice protocol. Australian and New Zealand Journal of Public Health, 26, 251-255. doi:10.1111/j.1467-842X.2002.tb00682.x

30.

Khalifeh

Hargreaves

Howard

Birdthistle

(2013). Intimate partner violence and socioeconomic deprivation in England: Findings from a national cross-sectional survey. American Journal of Public Health, 103, 462-472. doi:10.2105/AJPH.2012.300723

31.

Kuhn

Davidson

Durkin

(1994). Use of Poisson regression and time series analysis for detecting changes over time in rates of child injury following a prevention program. American Journal of Epidemiology, 140, 943-955. doi:10.1093/oxfordjournals.aje.a117183

32.

Lalonde

Taylor

(1997). Medical classification systems in Canada: Moving toward the year 2000. Canadian Medical Association Journal, 157, 1561-1565.

33.

Loxton

Schofield

Hussain

Mishra

(2006). History of domestic violence and physical health in midlife. Violence Against Women, 12, 715-731. doi:10.1177/1077801206291483

34.

Meuleners

Hendrie

Lee

(2008). Hospitalisations due to interpersonal violence: A population-based study in Western Australia. The Medical Journal of Australia, 188, 572-575.

35.

Meuleners

Lee

Hendrie

Fraser

(2010). A population study on Indigenous hospitalisations for interpersonal violence. Australian Health Review, 34, 123-126. doi:10.1071/AH09666

36.

Meuleners

Lee

Janssen

Fraser

(2011). Maternal and foetal outcomes among pregnant women hospitalised due to interpersonal violence: A population based study in Western Australia, 2002-2008. BMC Pregnancy and Childbirth, 11, Article 70. doi:10.1186/1471-2393-11-70

37.

Mokdad

A. H.

Forouzanfar

M. H.

Daoud

Mokdad

A. A.

El Bcheraoui

Moradi-Lakeh

. . . Murray

J. L.

(2016). Global burden of diseases, injuries, and risk factors for young people’s health during 1990-2013: A systematic analysis for the global burden of disease study 2013. The Lancet, 387, 2383-2401. doi:10.1016/S0140-6736(16)00648-6

38.

Morgan

Chadwick

(2009). Key issues in domestic violence (Research in Practice No. 7). Canberra: Australian Institute of Criminology.

39.

Murray

Crowe

Akers

(2016). How can we end the stigma surrounding domestic and sexual violence? A modified Delphi study with national advocacy leaders. Journal of Family Violence, 31, 271-287. doi:10.1007/s10896-015-9768-9

40.

O’Donnell

Sims

Maclean

Gonzalez-Izquierdo

Gilbert

Stanley

(2017). Trends in alcohol-related injury admissions in adolescents in Western Australia and England: Population-based cohort study. BMJ Open, 7(5), e014913. doi:10.1136/bmjopen-2016-014913

41.

Olive

(2018). Intimate partner violence and clinical coding: Issues with the use of the International Classification of Disease (ICD-10) in England. Journal of Health Services Research & Policy, 23, 212-221. doi:10.1177/1355819618781413

42.

Olsen

Lovett

(2016). Existing knowledge, practice and responses to violence against women in Australian Indigenous communities: State of knowledge paper. Alexandria: Australia’s National Research Organisation for Women’s Safety.

43.

SAS Institute Inc. (2013). SAS/ACCESS® 9.4 interface to ADABAS. Cary, NC: Author.

44.

Spangaro

Herring

Koziol-Mclain

Rutherford

Frail

Zwi

A. B.

(2016). “They aren’t really black fellas but they are easy to talk to”: Factors which influence Australian aboriginal women’s decision to disclose intimate partner violence during pregnancy. Midwifery, 41, 79-88. doi:10.1016/j.midw.2016.08.004

45.

Steering Committee for the Review of Government Service Provision. (2016). Overcoming Indigenous disadvantage: Key indicators 2016. Canberra, Australia: Productivity Commission.

46.

Taillieu

Brownridge

(2010). Violence against pregnant women: Prevalence, patterns, risk factors, theories, and directions for future research. Aggression and Violent Behavior, 15, 14-35. doi:10.1016/j.avb.2009.07.013

47.

Vos

Astbury

Piers

Magnus

Heenan

Stanley

… Webster

(2006). Measuring the impact of intimate partner violence on the health of women in Victoria. Australia Bulletin of the World Health Organization, 84(9), 739-744. doi:10.2471/BLT.06.030411

48.

Whisman

Beach

(2013). Relationship distress: Assessment definition and implications for mental health diagnosis. In Foran

Beach

Smith

Heyman

Wamboldt

(Eds.), Problems and family violence (pp. 71-78). New York, NY: Springer.

49.

Willis

(2011). Non-disclosure of violence in Australian Indigenous communities (Trends and Issues in Crime and Criminal Justice No. 405). Canberra: Australian Institute of Criminology.

50.

World Health Organization. (2002). World report on violence and health: Summary. Geneva, Switzerland: Author.

51.

World Health Organization. (2004). The economic dimensions of interpersonal violence. Geneva, Switzerland: Author.

52.

World Health Organization. (2012). Understanding and addressing violence against women. Geneva, Switzerland: Author.

53.

World Health Organization. (2013). Global and regional estimates of violence against women: Prevalence and health effects of intimate partner violence and non-partner sexual violence. Geneva, Switzerland: Author.

54.

World Health Organization. (2014). Violence against women: Intimate partner and sexual violence against women (Fact Sheet No. 239). Geneva, Switzerland: Author.

55.

World Health Organization. (2016). International classification of diseases. Geneva, Switzerland: Author.