Prevalence,Patterns,and Determinants of Intimate Partner Violence Experienced by Women Who Are Pregnant in Sanma Province,Vanuatu

Abstract

Violence perpetrated by an intimate partner (IPV) is an important human rights and public health problem worldwide and when experienced during pregnancy is of special concern due to the harmful impact on maternal and child health. Women living in Vanuatu, and especially Sanma Province, experience high rates of IPV, however little is known about their experiences of violence when pregnant. The aim was to describe the prevalence, patterns and determinants of IPV among women who are pregnant in Sanma Province, Vanuatu. A cross-sectional survey was used. All healthy adult women attending Northern Provincial Hospital antenatal clinic from late May to late July 2019 were eligible and invited to participate. Psychological, physical and sexual IPV and controlling behaviours were assessed with a modified version of the World Health Organization Violence Against Women Instrument administered as an individual interview. Descriptive statistics were used to analyse prevalence and patterns of IPV and logistic regression models to identify determinants. Of 214 women who expressed interest in participating, 192 women contributed data. Overall 64.2% of women had experienced any IPV during their lifetime and 42.2% had experienced IPV during their current pregnancy. Experience of co-occurrence of violence types was common, and it was more common for IPV to continue than to cease during pregnancy. Factors which increased likelihood of experiencing IPV included being employed, occupying a lower socioeconomic position, having a partner who was unemployed or used alcohol or illicit substances at least once a week. IPV, in all its forms, is a common problem faced by women who are pregnant and living in Vanuatu.

Keywords

battered women domestic violence cultural contexts predicting domestic violence

Women face violent transgressions of human rights at all stages of life, including physical, sexual, and psychological violence perpetrated by an intimate partner during pregnancy. Besides constituting a fundamental form of oppression of women’s ability to enjoy life and participate in society, intimate partner violence (IPV) experienced by women who are pregnant is a special concern given the extensive associations with adverse health outcomes for both the mother and her fetus (Ellsberg et al., 2008; Halim et al., 2018; Hill et al., 2016).

The prevalence of IPV varies within and between countries (World Health Organization [WHO] et al., 2013), reflecting differences in complex cultural, societal, and community factors that enable or limit the perpetration of violence (Lawson, 2012). Similarly, there are differences in the prevalence of IPV experienced by women who are pregnant worldwide. The landmark World Health Organization Multi-Country Study on Women’s Health and Domestic Violence Against Women found that the prevalence of experiencing physical violence during pregnancy varied from 1% in Japan to 28% in Peru, in most countries the range was between 4% and 12% (WHO, 2005). Taillieu and Brownridge’s (2010) more recent narrative synthesis of studies from low-, middle-, and high-income countries also found that prevalence of psychological (1.5–36.0%), sexual (1.0–3.9%), and physical (3.0–10.9%) violence during pregnancy varied among nations.

Besides true epidemiological variation, heterogeneity in research methods may influence differences in prevalence of IPV experienced during pregnancy when comparing study findings internationally. Taillieu and Brownridge’s (2010) review identified substantial variation in data collection instruments and drew attention to some limitations of the most commonly used instrument, the Abuse Assessment Screen (AAS). The AAS is a five-question tool for use in healthcare settings (McFarlane et al., 1992). It is less sensitive than other instruments (Reichenheim & Moraes, 2004) and does not disaggregate psychological violence from physical or sexual violence. In contrast, the longer Violence Against Women Instrument (VAWI) developed for the WHO Multi-Country Study (WHO, 2005) measures psychological, physical, and sexual IPV independently and is widely used in low- and middle-income countries (Costa & Barros, 2016). However, this tool was not specifically developed to assess IPV experienced during pregnancy and there are variations in its use among this study population. Some assess experiences of IPV within the previous 12 months (Bernstein et al., 2016; Dunkle et al., 2004), whereas other studies modified the instrument to measure IPV during pregnancy specifically (Bikinesi et al., 2017; Islam et al., 2021). The timeframe in which experiences of violence are measured is especially important in order to establish whether it intensifies, diminishes, occurs for the first time or does not occur during pregnancy (Ballard et al., 1998).

In Taillieu and Brownridge’s (2010) review, 31.3–69.2% of women who had ever experienced violence reported that it continued, 30.8–68.7% that it ceased, and for 3.8–40.0% of women violence was experienced for the first time during pregnancy. Studies that included psychological violence were more likely to find that violence continues during pregnancy rather than ceasing. Not measuring psychological and sexual violence estimates of exposure to IPV during pregnancy may be underestimated. For example, Silva et al. (2011) found that while the prevalence of physical violence decreased during pregnancy, the prevalence of psychological violence increased.

Experiencing violence prior to pregnancy is a significant risk factor for violence during pregnancy (Brownridge et al., 2011; James et al., 2013). In a meta-analysis of determinants of IPV experienced during pregnancy, James et al. (2013) found that pre-pregnancy violence was the only strong predictor among women living in “developed” countries. Whereas for women in “developing” countries, lower education, lower income, partner alcohol abuse, and an unwanted pregnancy also predicted IPV during pregnancy (James et al., 2013). Brownridge et al. (2011) found IPV during pregnancy was associated with younger age, being a member of an ethnic minority group, unemployment, being unmarried or separated, and living arrangement instability.

Despite advances in our global understanding of the prevalence, patterns, and determinants of IPV experienced during pregnancy, this knowledge has limited applicability to local communities and regions. Vanuatu, the setting for this study, is a lower-middle income country in Melanesia, encompassing 83 islands with a population of 234,023 people in 2009 (Vanuatu National Statistics Office [VNSO], 2009). The majority of the population live rurally (74.0%), with 20% living in the capital, Port Vila, and 6% living in Luganville, the sole other urban center located in Sanma Province (VNSO, 2010). In 2010, 12.7% of the ni-Vanuatu population lived below the Basic Needs Poverty Line, increasing to 23.6% in Luganville and one in eight households were headed by women (VNSO, 2010). The total fertility rate in 2009 was 4.2 children per woman (VNSO, 2009).

Using the WHO Multi-Country Study methods and questionnaire, the Vanuatu Women’s Centre (VWC) and National Statistics Office conducted a population-level survey on women’s experiences of IPV (Vanuatu Women’s Centre, 2011). They found that 60% had experienced physical or sexual IPV and 68% had experienced psychological IPV in their lifetime. The survey illustrates that women living in Sanma Province experience violence at one of the highest known rates in the world; 74% of women have ever experienced physical and/or sexual IPV, and 78% have experienced psychological IPV. Of women who had ever been pregnant in Vanuatu, 15% reported having experienced physical violence during a pregnancy (VWC, 2011), 14% in Luganville. There are no data on the prevalence of other forms of IPV experienced by ni-Vanuatu women who are pregnant, or on the patterns or determinants of IPV during pregnancy in Vanuatu, nor in any other low- or-middle income Pacific Island Country. This knowledge is needed to inform evidence-based local policy, programs, and public services to reduce preventable maternal morbidity and mortality caused by violence. Globally, data collected using instruments that measure disaggregated forms of violence using pregnancy specific timeframes are needed to contribute to our understanding of the relationship between IPV and pregnancy. This study aimed to ascertain the prevalence, patterns, and determinants of psychological, physical, and sexual violence experienced by women who are pregnant and living in Sanma, Vanuatu.

Methods

Design

A cross-sectional survey.

Setting

The study was conducted at Northern Provincial Hospital (NPH) in the township of Luganville. NPH is the smaller of two tertiary health services in Vanuatu and serves people living in Sanma Province. The midwife-led antenatal clinic operates four days a week, providing care to between 60 and 100 women per week. While community antenatal health services are established, women travel from across Sanma to access more specialized services at the hospital clinic.

Participants and Recruitment

Sample size calculations were performed to estimate a proportion using Cochrane’s formula (Cochrane, 1977) based on the previously known prevalence of physical violence (14% in Luganville), 95% confidence interval and 5% margin of error. These indicated a sample size of 185 women was needed. The inclusion criteria were to be a pregnant adult woman attending NPH for antenatal care. Exclusion criteria were to be younger than 18 years or unable to give or communicate informed consent because of being physically or intellectually differently abled. All people who attended NPH antenatal clinic for care during the data collection period were invited to participate in the study. The head of antenatal care informed patients of the study during a group health talk at the beginning of clinic. Brief information about the study, described as being about “women’s health and relationships during pregnancy” in order to ensure safety and confidentiality, was also given during individual appointments. Women who expressed interest to the researcher, interviewers or antenatal care staff were accompanied to a private interview room, where study interviewers read aloud an explanatory statement, which further elucidated the purpose of the study and the nature of the questions in the questionnaire, and the woman’s eligibility was confirmed. The woman’s verbal consent to participate was witnessed by a researcher and interviewer, following which the researcher left the interview room.

Data Sources

Outcomes. IPV was defined and measured using the operational definitions and data collection instrument developed for the WHO Multi-Country Study on Women’s Health and Domestic Violence Against Women (WHO, 2005). In addition to the standard WHO-defined variables measured by the VAWI, two pregnancy-specific variables were added for physical and sexual violence. These variables were based on questions from the pregnancy section of the full WHO Women’s Health and Life Experiences survey. The psychometric properties of the VAWI have not been tested in Vanuatu.

Psychological violence was defined as experience of one or more of the following: being insulted or humiliated, belittled in front of family, or others, scared or intimated on purpose, threatened with harm to her or her family. Physical violence was defined as one or more of the following: being slapped, pushed, hit, kicked, choked or burnt, threatened with a weapon, harmed with a weapon, or hit, kicked or punched in the pregnant abdomen. Sexual violence included experience of one or more of the following: being forced to have sex, having sex because she was frightened of what he might do, being forced to do something sexual she found humiliating or being vaginally penetrated to harm the baby. Controlling behaviors included trying to restrict her contact with friends or family, insisting on knowing her whereabouts, treating her indifferently, becoming angry when she spoke to another man, being suspicious that she is unfaithful, and expecting her to ask permission before doing anything or before seeking healthcare.

We modified the VAWI (WHO, 2005) to assess experiences of violence perpetrated by an intimate partner during the lifetime, past pregnancies, and the current pregnancy. We measured frequency and severity of violence perpetrated by the woman’s current partner by asking if the behaviors had happened once, a few times or many times before and during pregnancy, and whether they had worsened, improved or stayed the same. We used the WHO Women’s Health and Life Experiences questionnaire to measure economic violence, asking if a woman’s partner had ever taken her earnings or belongings without her permission, had refused to give her money for household expenses or had forced her to quit or refuse paid work.

Explanatory Factors. The sociodemographic characteristics and reproductive history factors of participants that were collected included age, literacy status, educational attainment, employment status, rurality of origin and current living location, number of times partnered, household composition, marital status, gravidity, and gestation. Participants were asked their partner’s age, literacy status, educational attainment, employment status and frequency of kava, alcohol, and illicit substance use. Participants were asked about (joint or sole) ownership of vehicles, property, valuables, and savings, and various household objects. The classification used in the Vanuatu National Study (VWC, 2011) was applied to categorize a woman’s socioeconomic status based on their ownership of these items.

Procedure

Data were collected by trained, female, ni-Vanuatu interviewers using consecutively numbered paper questionnaires. The interviewers were fluent in Bislama, the most commonly spoken national language, and had completed secondary education. Before data collection began they were given three days of training in gender-based violence and interview techniques. The study questionnaire was developed in English, checked for cultural sensitivity and competency, and translated by an independent, professional ni-Vanuatu woman. It was back translated and checked for accuracy by a bilingual interviewer. Two pilot interviews were conducted and minor inaccuracies corrected. Data were collected between late May and late July 2019. Women who participated were given a small gift, a scented bar of soap, as a token of appreciation of their time.

Data Management and Analyses

Data entry was completed manually and, to check for errors, 10% of the questionnaires were entered twice. SPSS Version 25 was used for data cleaning and analyses. Missing data were excluded in a pairwise fashion, with the exception of computing the binary categorical variables of any IPV experience and any controlling behavior experience. Descriptive analyses were used to determine the prevalence and patterns of IPV. Parametric and nonparametric tests, including chi squared, t-test for independence, Mann-Whitney U, and Fisher’s exact tests, were used for initial bivariate analysis. This was followed by backwards stepwise logistic regression models performed for multivariate analysis of independent associations with a p-value of < 0.2. Regression models were performed for outcome variables of any IPV (excluding controlling behavior) during the lifetime, any IPV (excluding controlling behavior) during the current pregnancy, as well as individual models for psychological, physical, and sexual violence both during the lifetime and the current pregnancy. The factors (independent variables) included in the models were age, secondary school completion, parity, gestation, income level, rurality, partnered more than once, co-habitation with her partner, partner’s family and her family, employment, income greater than her partner’s, partner unemployment, partner alcohol, drug and kava use, partner’s education level and whether the pregnancy was planned. Odds ratios were used to measure associations and the 95% confidence intervals were obtained. P-values of < 0.05 were considered significant.

Ethical Considerations

The study followed the World Health Organization ethical and safety guidelines for researching violence against women (WHO, 2001). All interviews were conducted in a private room, with children older than the age of two years supervised outside the interview room. All women who participated in the study, regardless of whether they had disclosed experiences of violence, were given verbal and written information about local support services. Ethics approval for the study was provided by Monash University Human Research Ethics Committee (approval number 18348) and the Vanuatu Ministry of Health (reference MOH/DG 02/6-GKT/ms).

Results

Participants

On days when data were being collected, a total of 537 patients attended NPH antenatal clinic. Of these, 214 expressed interest in participating and were read the study explanatory statement. Two women were excluded because they did not meet inclusion criteria and 19 declined to participate. One woman withdrew from the study midway through the interview, leaving 192 women who provided complete data included in the analyses. The characteristics of the group of women who participated are shown in Table 1. Participants were more likely to have completed secondary education than the general population of females aged between 20 and 24 years in Vanuatu. Similarly, participants were less likely to be in the low household income category compared to all women in Vanuatu aged 20 to 24 (Vanuatu Ministry of Health et al., 2014). Most had been partnered only once and were living with their partner, although only a quarter had been legally or formally married. The mean gestation at assessment for this study was 27 weeks and participants had been pregnant between one and seven times, 37.89% were nulliparous.

Table 1.

Sociodemographic Characteristics, Antenatal and Reproductive History of Participants and Comparison With National Data Where Available.

Variable	N	Mean	Std Dev	% of female population of Vanuatu
Age	191	25.7	5.4
gestation	188	27.3	8.8
	N	n	%
Completed education None Primary Secondary Tertiary	191	3 56 112 20	1.57 29.31 58.64 10.47	25.3† 19.2† 43.0† 11.0†
Socioeconomic status Low Middle High	191	4 130 57	2.09 68.06 29.84	20.0† 60.0† 20.0†
Employed before pregnancy	190	89	46.84	55.0‡
Continued employment during pregnancy	190	63	33.16	–
Current living location Luganville Rural Sanma Remote Sanma Other province	192	103 73 14 2	53.65 38.02 7.29 1.04	Urban 12.7† Rural 87.3†
Number of times partnered Never partnered 1 partner 2 or more partners	191	1 151 39	0.52 79.06 20.42	– – –
Living with current partner	186	170	91.40	–
Living with other family Her family Partner’s family Both Neither	191	40 94 29 28	20.94 49.21 15.18 14.66	– – – –
Married to current partner	191	47	24.61	42.0‡
Parity Nulliparous Parous	190	72 118	37.89 62.11	37.8† 62.2†

Source of data for Vanuatu population: 2013 Demographic and Health Survey.

Note. †Of women aged between 20 and 24 years.

‡Of the total adult female population.

Prevalence of Violence

The prevalence of psychological, physical, and sexual violence during the lifetime and current pregnancy are shown in Table 2. Two women completed the psychological and physical violence sections but were not willing to complete the sexual violence section. They were included in the analyses of any IPV experience. Two women were not willing to provide data for one and two variables out of a total of eight variables in the controlling behavior section and they were included in the any controlling behaviour experience variable. Almost 65% had experienced psychological, physical or sexual violence perpetrated by an intimate partner during their lifetime, almost 43% had experienced psychological, physical or sexual violence during the current pregnancy. Including women’s accounts of previous pregnancies, 103 out of 187 had experienced violence during any pregnancy (55.08%). Psychological violence was the most common form, followed by physical and then sexual violence. However, experience of sexual violence was not uncommon with over 12% of the sample reporting it during the current pregnancy. A total of 8 in 10 women (out of 185 women) were currently experiencing at least one controlling behavior by their partner. The most commonly experienced controlling behaviors were being expected to ask for permission before doing anything (68.48% of 184 women), specifically before seeking healthcare (42.16% of 185 women) and her partner insisting on knowing where she is at all times (44.86% of 185 women). A total of 12% of women (out of 189 women) had experienced a partner taking their money without consent during their lifetime and 21.05% (out of 190 women) said their partner had refused to give them money for household products even though the partner had money for his own personal use. Experience of co-occurrence of different forms of IPV was common both during pregnancy and over the lifetime. A total of 7% of women had experienced psychological, physical, and sexual violence during their current pregnancy and 14.44% had experienced all three forms during their lifetime (see Figure 1). Experience of multiple forms of violence over the lifetime may include different forms of violence perpetrated by different partners as well as perpetration of all three forms by a single partner. It is therefore not directly comparable to the data on co-occurrence during the current pregnancy, which measures violence perpetrated by a single partner.

Figure 1.

Note. IPV = Intimate Partner Violence.

Table 2.

Prevalence of Women’s Experience of Psychological, Physical, and Sexual Intimate Partner Violence During Lifetime and Current Pregnancy.

	N	Lifetime	Current Pregnancy
		n (%)	n (%)
Psychological violence	187	108 (57.75)	63 (33.69)
Physical violence	187	74 (39.57)	41 (21.93)
Sexual violence	185	36 (19.46)	23 (12.43)
Any intimate partner violence	187	120 (64.17)	79 (42.25)

Patterns of Violence

The patterns of violence experienced by women before and during pregnancy are shown in Table 3. The most frequently observed pattern was violence experienced before pregnancy by a current partner, which continued during pregnancy. For example, of 108 women who had experienced psychological violence in their lifetime, 55 said that at least one of the behaviors comprising psychological violence had begun before and continued during pregnancy. The next most common pattern was having experienced violence by the current partner before the current pregnancy, which ceased during pregnancy. This was followed by lifetime experience of violence perpetrated by a former partner. Finally, the least frequently observed pattern was experiencing violence perpetrated by the current partner for the first time in the current pregnancy. Around 13% of all reports of violent behaviors were experienced for the first time during pregnancy. Of those who had experienced violence both before and during the current pregnancy, for all forms of violence, most women reported a reduction in severity. At most, 2% of women reported a violent behavior worsening during the current pregnancy compared to before.

Table 3.

Patterns of Psychological, Physical, And Sexual Intimate Partner Violence Before and During Pregnancy.

	Psychological Violence		Physical Violence		Sexual Violence
	n	%	n	%	n	%
Experienced violence by current partner before current pregnancy but not during	35	32.41†	30	40.54†	11	30.56†
Experienced violence by current partner during current pregnancy but not before	15	13.89†	10	13.51†	5	13.89†
Experienced violence by current partner both before and during current pregnancy	55	50.93†	34	45.95†	19	52.78†
Experienced violence in lifetime but not by current partner	28	25.93†	21	28.38†	9	25.00†
Total (Experienced violence by any partner during lifetime)	108	100	74	100	36	100

Note. †In this table “n” is the number of women who reported at least one violent behaviour within each violence type following each pattern. Many women experienced different patterns for different violent behaviours within the same type and so the same woman may be represented in more than one row within the same column. For example, a woman may have reported that her partner continued to slap her during pregnancy, but ceased kicking her. Therefore the patterns of violence presented here are not mutually exclusive and the percentages of each column do not sum to 100.

Determinants of Violence

The regression models of variables associated with experience of IPV during the lifetime and IPV during the current pregnancy are shown in Table 4. The likelihood of a woman experiencing IPV during her lifetime was increased if she was employed or had been partnered more than once, or if her partner was unemployed or he used alcohol at least once a week. Factors that protected against lifetime experience of IPV were living in Luganville rather than other areas of Sanma and living with her family of origin or her partner’s family. In addition to the factors associated with lifetime risk of violence, occupying a lower socioeconomic position, and partner’s use of illicit substances at least once a week were associated with an increased risk of IPV during the current pregnancy. Living in Luganville, having completed secondary education, living with her family of origin or living with her partner were protective against violence during pregnancy.

Table 4.

Associations With Intimate Partner Violence Experienced During the Lifetime and Current Pregnancy.

	OR	95% CI
Any intimate partner violence experience during the lifetime
Employed	2.76	1.16 – 6.60
Partnered more than once	3.57	1.13 – 11.24
Partner unemployed	2.98	1.06 – 8.33
Partner uses alcohol at least once a week	2.72	1.01 – 7.31
Living in Luganville	0.23	0.09 – 0.55
Living with birth family	0.07	0.01 – 0.36
Living with partner’s family	0.09	0.02 – 0.46
Partner completed secondary school	0.50	0.19 – 1.27
Any intimate partner violence experience during the current pregnancy
Employed	2.54	1.11 – 5.81
Partner unemployed	2.33	0.93 – 5.81
Partner uses illicit drug at least once a week	5.38	1.33 – 21.74
Partner uses alcohol at least once a week	3.54	1.45 – 8.66
Low or middle income	2.60	1.08 – 6.24
Living with birth family	0.39	0.14 – 1.07
Living in Luganville	0.40	0.18 – 0.92
Completed secondary school	0.39	0.16 – 0.92
Living with partner	0.14	0.02 – 0.94

Note. OR = odds ratio; CI = confidence interval.

Discussion

In this study, we describe for the first time, the prevalence, patterns, and some determinants of psychological, physical, and sexual IPV against women who are pregnant and receiving antenatal care in Luganville. These are the most comprehensive data to date on this major risk to women’s health in any low- and middle-income Pacific Island Country.

Prevalence of Violence

The first main finding is the high prevalence of violence perpetrated against women who are pregnant. Overall, 42.25% of participants experienced psychological, physical or sexual IPV during their current pregnancy. This is higher than the rates reported in most studies in the review by Taillieu and Brownridge (2010) and the meta-analysis by James et al. (2013). It is within ranges found in regional reviews of prevalence studies in Asia (Kashif et al., 2010) and Africa (Shamu et al., 2011), demonstrating that the risk of experiencing IPV during pregnancy in Vanuatu is similar to that in some other low- and middle-income countries in the Global South and is higher than most places in the world.

We found the prevalence of IPV during any pregnancy was much higher (55.08%) than the previously determined estimate of 15% in Vanuatu found in the Vanuatu National Survey using the WHO Multi-Country Study methods and questionnaire (VWC, 2011). As per the WHO Multi-Country Study, the prevalence found in the Vanuatu National Survey considers only physical violence. It is therefore not surprising that, by using a more inclusive definition and measures of IPV during pregnancy, we found a greater percentage of women had experienced violence when pregnant. Even comparing the prevalence of physical violence in our study to that in the Vanuatu National Survey we found a considerably greater prevalence. We found 33.69% of women in our sample had experienced physical violence during any pregnancy in contrast to the 15% in the Vanuatu National Survey (VWC, 2011). This phenomenon has also been found in other IPV studies specific to pregnancy when comparing results to population-level household surveys using the WHO Multi-Country Study methods and the complete Women’s Health and Life Experiences Questionnaire. Using a modified version of the violence tool within the full WHO questionnaire, the VAWI, Islam et al. (2021) found a much higher prevalence of physical violence during pregnancy in Bangladesh compared to the original WHO Multi-Country Study. In the Multi-Country Study the prevalence of physical violence was found to be 12.4% in provincial areas and 10.2% in urban areas in Bangladesh (WHO, 2005). However, in an urban setting, Islam et al. (2021) found that the prevalence of all forms of IPV during pregnancy was 66.4% and the prevalence of physical violence during pregnancy was 35.2%. This is comparable to the prevalence found in our study, and the modified VAWI used in Islam et al.’s study was very similar to the one used in this study. The differences found here may be explained by the use of comprehensive and multidimensional behavioral-specific measures of violence and defining the pregnancy period precisely. Our study was able to measure violence occurring during pregnancy much more comprehensively than the original Multi-Country Study and population-based surveys using its methods, given the target sample of all women aged between 18 and 49, not solely women who are pregnant, in these larger studies.

Patterns of Violence

The second main finding is that IPV during pregnancy largely represents a continuation of violence and that for most women in this setting pregnancy is not protective against experiencing violence. Of women who had ever experienced violence in our study, more than half also experienced it during pregnancy. The wide variation in patterns of continuing violence in Taillieu and Brownridge’s (2010) review may indicate that there are cultural differences in the effect of pregnancy on men’s attitudes and use of violence against their partners. In a systematic review of global prevalence studies, Campbell et al. (2004) concluded that cultural differences regarding pregnancy and violence have not been well studied. In our study, although most women reported that violence decreased in severity, the high rate of continuation of all forms of violence during pregnancy demonstrates that pregnancy itself is not protective against violence for women in Sanma, Vanuatu. This may illustrate social norms in which violence against women who are pregnant are seen as acceptable in Vanuatu, especially given that over 97% of women reported that their partner welcomed the pregnancy.

Whether or not pregnancy itself is a risk factor for violence has been debated in the literature. In a nationally representative sample of men living in Hong Kong, Chan et al. (2011) found that the prevalence of violence perpetration against partners who were pregnant was higher than against partners who were not pregnant, but in the multivariable analyses the significant relationship between pregnancy and violence disappeared. In our study, despite the fact that violence continued during pregnancy for most women, the majority also reported that violence improved in severity during this time, with some reporting it stayed the same and only a very small number of women reporting that violent behaviors worsened. However, only women who had experienced violence before pregnancy perpetrated by their current partner were asked about changes in severity, and we found that nearly 14% of women experienced each type of violence for the first time during pregnancy by their current partner. This may indicate that for a substantial minority of women, pregnancy does incur an increased risk of violence by their male partners. However, a partner beginning perpetration of violence during pregnancy may not be related to the pregnancy itself.

Determinants of Violence

The third main finding is that women’s employment and partner unemployment and substance use were significantly associated with higher likelihood of women experiencing IPV. Although also found in the Vanuatu National Survey (VWC, 2011), employed women having a higher likelihood of IPV experience is a finding that diverges from most global evidence which is that unemployment is associated with experiences of IPV (Brownridge et al., 2011). Like this study, in a cross-sectional investigation in Peru, Perales et al. (2009) found employment associated with increased risk of IPV. The authors concluded that in patriarchal societies, employment of women increases risk as men may use violence to reinforce their dominance over financially independent women. Thus, the finding in our study may demonstrate the low societal status of women in Vanuatu, with women who do not adhere to strict traditional gender roles being placed at increased risk of violence. Few studies have demonstrated an association between male unemployment and women’s experience of IPV, however in the two studies that investigated this included in the review by Brownridge et al. (2011) both found the association to be significant. In our study, partner unemployment was significantly associated with overall IPV and most strongly with psychological violence during the current pregnancy.

We found that partner use of alcohol and illicit drugs was associated with experiencing IPV. In most of the regression models, partner’s use of alcohol at least once a week contributed independently to the likelihood of experience of IPV. Partners’ substance use was also associated with women’s lifetime experience of IPV in the Vanuatu National Survey (VWC, 2011). Reviews by James et al. (2013) and Brownridge et al. (2011) affirm a significant relationship between partner substance use and perpetration of IPV. In the United Nations Multi-Country Study on Men’s Use of Violence in Asia-Pacific, alcohol use was significantly associated with increased likelihood of IPV perpetration (Fulu et al., 2013). However, the authors emphasized that individual factors do not necessarily operate in isolation and are interconnected with community and societal factors. It is likely that the relationship between alcohol and illicit drug use and IPV perpetration in Vanuatu is multifaceted and there are many potential confounding or mediating factors that have not been investigated in this study, such as partner’s mental health and experiences of violence. Nevertheless, these data indicate that men who use alcohol or illicit drugs are more likely to abuse their partners.

We found that having completed rather than not completed secondary school, living in Luganville rather than in a remote area, and living with family were protective against IPV. Higher education as a protective factor is consistent with other studies (Brownridge et al., 2011). Rurality and household composition and associations with IPV during pregnancy have not been explored in any reviews of risk factors. These findings may be specific to family and social organization in Vanuatu, in which most people live in rural areas with or in close proximity to extended family members.

Strengths and Limitations

The strengths of our study include the careful modification of the gold-standard VAWI to ensure women’s experiences before and during pregnancy were captured and attention to cultural sensitivity and inclusiveness was paid. Local health professionals and experts in field of violence against women and girls, many of them women, were involved in the conceptualization, design and implementation of the study. The topic was of relevance and importance to the local area, as recognized by local people. The study procedures were appropriate and culturally sensitive and drew upon lessons from the Vanuatu National Survey on Women’s Health and Family Relationships (VWC, 2011).

Nevertheless we acknowledge some limitations. While we have established the prevalence of women’s experiences of psychological, physical, and sexual violence during pregnancy, we were unable to document economic violence in as much detail. This is primarily because there are no data collection instruments which measure economic violence alongside psychological, physical, and sexual violence (Costa & Barros, 2016). Further, fewer than half the participants were engaged in paid work before pregnancy and among those who were employed it was unclear how employment was affected by pregnancy and how many sources of income the women usually relied upon. This may reflect the dynamic earning behaviors of women in Sanma; it is likely that women have multiple and transient sources of income which were difficult to capture in our questionnaire.

Further, our study was cross-sectional and so we cannot conclude that relationships between sociodemographic or reproductive factors and experience of IPV are causal. Women at any gestation were included, meaning that it was not possible to examine whether risk was higher in one trimester of pregnancy than another. It is possible that some women who reported they had not experienced violence during the index pregnancy actually experienced it later in pregnancy or that the frequency or severity of violence changed later in the pregnancy. However, the gestational ages of women in our sample ranged from the beginning to the end of pregnancy, with the mean gestation at the end of the second trimester. Therefore, the experiences of women at all stages of pregnancy are represented in our study, and we also ascertained the prevalence of violence during previous pregnancies.

We comprehensively describe experiences of IPV during pregnancy among women attending antenatal care in a tertiary hospital setting. However, these findings might not pertain to women in Sanma who do not attend hospital antenatal care and women living in other provinces in Vanuatu. At the conclusion of the Millennium Development Goals period, it was estimated that 85% of women in Vanuatu and 61% of women in the province of Sanma, attend at least one antenatal clinic visit during pregnancy (United Nations Population Fund, 2015). In their systematic review, Metheny and Stephenson (2017) found that experiencing IPV reduces the likelihood of adequate antenatal care in some low- and middle-income countries. This evidence indicates that the prevalence of IPV found in this study may be an underestimate of the prevalence of IPV experienced by those who do not attend for antenatal care.

Our study was conducted at a single site, in Luganville, and therefore may have limited generalizability to other provinces. Nearly 38% of the women in our sample were originally from another province, although the proportion of partners who had migrated to Sanma from another province was unknown. The sociodemographic characteristics of the women in our sample were mostly comparable to women in the general population. However, women in our study were more likely to have completed any schooling, have high socioeconomic status, and be from an urban area. Considering the results from this study concerning the determinants of IPV experience, the prevalence of violence reported is likely an underestimate of the true magnitude of IPV against women who are pregnant in the general population of Vanuatu.

Implications and Conclusions

These data reveal that IPV, in all its forms, is a significant public health problem affecting women who are pregnant in Vanuatu. They provide relevant evidence for policies and programs in especially, but not limited to, the health and social development sectors in Vanuatu and possibly other Pacific Island Countries.

Given that women’s health care during pregnancy needs to address all relevant risk factors, it is recommended that health services and healthcare providers should play an active role in the detection and management of IPV (WHO, 2013). The WHO recommends that IPV screening programs require a protocol or standard operating procedure, training for healthcare providers who conduct screening, an environment in which privacy and confidentiality can be ensured, and an appropriate referral system to support services in place (WHO, 2013). Given the high prevalence demonstrated here, we believe that in Vanuatu, antenatal care presents a unique opportunity for intervention, both for women who have experienced violence for a long time and those for whom violence has begun during the pregnancy.

However, it is not only the health sector that must engage with this problem. Cross-sectoral action, at local and national levels, is needed to both prevent and manage IPV during pregnancy; a crucially important time in a woman’s life, which influences not only her own future but also that of her child. Our results suggest that this action must be targeted at men as well as women, and norms which reinforce gender inequalities, which in turn lead to the widespread perpetration of IPV, must be addressed. Further research into culturally appropriate and effective interventions, which focus on respectful, patient-centered care with attention to mental health and promote women’s autonomy are needed.

Footnotes

Acknowledgments

The authors would like to acknowledge Dr Andy Ilo, Sala Nial and Marie-Michelle Manwo from Northern Provincial Hospital for their logistical and administrative assistance. We thank the research interviewers, Airine Manwo, Jefline Talo, Ellina Abel, and Caroline Garae for assistance with translation of study materials. We acknowledge the Sanma Counselling Centre (VWC) for their contribution to interviewer training. We thank Hau Nguyen for assistance with data-analyses and Ruby Stocker for comments on this manuscript. We thank Soroptimist International Gippsland for their donation. We acknowledge Northern Provincial Hospital and Monash University for supporting the study.

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: JF is supported by the Finkel Professorial Fellowship, which is funded by the Finkel Family Foundation; TT is supported by an Australian National Health and Medical Research Council Early Career Research Fellowship and a Monash Strategic Bridging Fellowship. Monash University provided a student research grant which was used to remunerate the interviewers and cover field research costs. Soroptimist International Gippsland provided a grant which was used to fund small gifts for the women who participated by contributing data.

ORCID iD

Stephanie McKelvie

Author Biographies

Stephanie McKelvie, BMedSc/MD BMedSc (Hons) MPH Student, is a medical and honors research student at Monash University. She has undertaken student medical and surgical placements in Australia and the Pacific, including Vanuatu. She is interested in global sexual and reproductive health and rights.

Basil Leodoro, MD, general and paediatric surgeon, past president of the Vanuatu Medical and Dentist Association and Chief Medical Officer for Santo, is the head of surgery at Northern Provincial Hospital.

Thomas Sala, MD, obstetrician and gynaecologist, is head of obstetrics and gynaecology at Northern Provincial Hospital.

Thach Tran, PhD, BEc, MIRB, MSc, is a research fellow at the Global and Women’s Health, Monash University. He has more than 20 years’ experience of public health research in Vietnam and other low- and lower-middle-income countries. His research focuses on social determinants of women’s mental health, child health, nutrition, and development in early childhood in low- and middle-income countries, and measurement in health sciences.

Jane Fisher, AO, PhD, an academic clinical and health psychologist is Finkel Professor of global health and director of Global and Women’s Health at Monash University. She has longstanding research interests in the social and structural determinants, including interpersonal violence, of mental health problems experienced by women in low- and high-income countries.

References

Ballard

T. J.

, Saltzman

L. E.

, Gazmararian

J. A.

, Spitz

A. M.

, Lazorick

, & Marks

J. S.

(1998). Violence during pregnancy: Measurement issues. American Journal of Public Health , 88(2), 274–276. https://doi.org/10.2105/ajph.88.2.274

Bernstein

, Phillips

, Zerbe

, McIntyre

J. A.

, Brittain

, Petro

, Abrams

E. J.

, & Myer

(2016). Intimate partner violence experienced by HIV-infected pregnant women in South Africa: A cross-sectional study. BMJ Open , 6(8), https://dx-doi-org.web.bisu.edu.cn/10.1136/bmjopen-2016-011999

Bikinesi

L. T.

, Mash

, & Joyner

(2017). Prevalence of intimate partner violence and associated factors amongst women attending antenatal care at Outapi clinic, Namibia: A descriptive survey. African Journal of Primary Health Care and Family Medicine, 9(1), 1512. https://doi.org/10.4102/phcfm.v9i1.1512

Brownridge

D. A.

, Taillieu

T. L.

, Tyler

K. A.

, Tiwari

, Ling

C. Santos S.

(2011). Pregnancy and intimate partner violence: Risk factors, severity, and health effects. Violence Against Women , 17(7), 858–881. https://doi.org/10.1177/1077801211412547

Campbell

, García-Moreno

, & Sharps

(2004). Abuse during pregnancy in industrialized and developing countries. Violence Against Women , 10(7), 770–789. https://doi.org/10.1177/1077801204265551

Chan

K. L.

, Brownridge

D. A.

, Tiwari

, Fong

D. Y.

, Leung

W. C.

, & Ho

P. C.

(2011). Associating pregnancy with partner violence against Chinese women. Journal of Interpersonal Violence , 26(7), 1478–1500. https://doi.org/10.1177/0886260510369134

Cochrane

W. G.

(1977). Sampling techniques (3rd ed.). John Wiley & Sons.

Costa

, & Barros

(2016). Instruments to assess intimate partner violence: A scoping review of the literature. Violence and Victims , 31(4), 591–621. https://doi.org/10.1891/0886-6708.VV-D-14-00122

Dunkle

K. L.

, Jewkes

R. K.

, Brown

H. C.

, Yoshihama

, Gray

G. E.

, McIntyre

J. A.

, & Harlow

S. D.

(2004). Prevalence and patterns of gender-based violence and revictimization among women attending antenatal clinics in Soweto, South Africa. American Journal of Epidemiology , 160(3), 230–239. https://doi.org/10.1093/aje/kwh194

10.

Ellsberg

, Jansen

H. A.

, Heise

, Watts

C. H.

, & Garcia-Moreno

(2008). Intimate partner violence and women’s physical and mental health in the WHO multi-country study on women’s health and domestic violence: An observational study. Lancet , 371(9619), 1165–1172. https://doi.org/10.1016/S0140-6736(08)60522-X

11.

Fulu

, Warner

, Miedema

, Jewkes

R. K.

, Roselli

, & Lang

(2013). Why do some men use violence against women and how can we prevent it? Quantitative findings from the United Nations multi-country study on men and violence in Asia and the Pacific . UNDP, UFPA, UN Women and UNV. https://www.undp.org/content/dam/rbap/docs/Research%20&%20Publications/womens_empowerment/RBAP-Gender-2013-P4P-VAW-Report.pdf

12.

Halim

, Beard

, Mesic

, Patel

, Henderson

, & Hibberd

(2018). Intimate partner violence during pregnancy and perinatal mental disorders in low and lower middle income countries: A systematic review of literature, 1990-2017. Clinical Psychology Review , 66, 117–135. https://doi.org/10.1016/j.cpr.2017.11.004

13.

Hill

, Pallitto

, McCleary-Sills

, & Garcia-Moreno

(2016). A systematic review and meta-analysis of intimate partner violence during pregnancy and selected birth outcomes. International Journal of Obstetrics and Gynaecology , 133(3), 269–276. https://doi.org./10.1016/j.ijgo.2015.10.023

14.

Islam

M. J.

, Broidy

, Mazerolle

, Baird

, & Mazumder

(2021). Exploring intimate partner violence before, during, and after pregnancy in Bangladesh. Journal of Interpersonal Violence , 36(7-8), 3584–3612. https://doi.org/10.1177/0886260518775753

15.

James

, Brody

, & Hamilton

(2013). Risk factors for domestic violence during pregnancy: A meta-analytic review. Violence & Victims , 28(3), 359–380. https://doi.org/10.1891/0886-6708.vv-d-12-00034

16.

Kashif

, Murtaza

, & Kirkman

(2010). Violence against women during pregnancy in some Asian countries: A review of the literature. Italian Journal of Public Health , 7(2), 6–11. https://doi.org/10.2427/5721

17.

Lawson

(2012). Sociological theories of intimate partner violence. Journal of Human Behavior in the Social Environment , 22(5), 572–590. https://doi.org./10.1080/10911359.2011.598748

18.

McFarlane

, Parker

, & Soeken

(1992). Assessing for abuse during pregnancy: Severity and frequency of injuries and associated entry into prenatal care. Journal of the American Medical Association , 267(23), 3176–3178. https://doi.org/10.1001/jama.1992.03480230068030

19.

Metheny

, & Stephenson

(2017). Intimate partner violence and uptake of antenatal care: A scoping review of low- and middle-income country studies. International Perspectives on Sexual and Reproductive Health , 43(4), 163–171. https://doi.org/10.1363/43e4917

20.

Perales

M. T.

, Cripe

S. M.

, Lam

, Sanchez

S. E.

, Sanchez

, & Williams

M. A.

(2009). Prevalence, types, and pattern of intimate partner violence among pregnant women in Lima, Peru. Violence Against Women , 15(2), 224–250. https://doi.org/10.1177/1077801208329387

21.

Reichenheim

M. E.

, & Moraes

C. L.

(2004). Comparison between the abuse assessment screen and the revised conflict tactics scales for measuring physical violence during pregnancy. Journal of Epidemiology and Community Health , 58(6), 523–527. https://doi.org/10.1136/jech.2003.011742

22.

Shamu

, Abrahams

, Temmerman

, Musekiwa

, & Zarowsky

(2011). A systematic review of African studies on intimate partner violence against pregnant women: Prevalence and risk factors. PLOS One , 6(3), Article e17591. https://doi.org/10.1371/journal.pone.0017591.

23.

Silva

E. P.

, Ludermir

A. B.

, Araujo

T. V.

, & Valongueiro

S. A.

(2011). Frequency and pattern of intimate partner violence before, during and after pregnancy. Revista de Saude Publica , 45(6), 1044–1053. https://doi.org/10.1590/s0034-89102011005000074

24.

Taillieu

T. L.

, & Brownridge

D. A.

(2010). Violence against pregnant women: Prevalence, patterns, risk factors, theories, and directions for future research. Aggression and Violent Behavior , 15(1), 14–35. https://doi.org/10.1016/j.avb.2009.07.013

25.

United Nations Population Fund. (2015). Vanuatu sexual and reproductive health rights needs assessment . Author. https://pacific.unfpa.org/sites/default/files/pub-pdf/4.VanuatuSexualandReproductiveHealthRightsNeedsAssessmentReportLRv1.pdf

26.

Health

Vanuatu Ministry of

, Office

Vanuatu National Statistics

, & Community.

Secretariat of the Pacific

(2014). Vanuatu demographic and health survey 2013 . Secretariat of the Pacific Community. https://vnso.gov.vu/images/Special_Report/Vanuatu_Demographic_and_Health_Survey_Report/2013_VDHS_FINAL_Report.pdf

27.

(VNSO).

Vanuatu National Statistics Office.

(2009). 2009 National housing and population census . VNSO. https://vnso.gov.vu/index.php/census-and-surveys/census/censuses

28.

VNSO. (2010). Vanuatu hardship and poverty report: Analysis of the 2010 household income and expenditure survey . VNSO and UNDP Pacific Centre Suva. https://vnso.gov.vu/images/Special_Report/Vanuatu_Hardship_and_Poverty/2010_Vanuatu_Hardship_and_Poverty_Report.pdf

29.

Vanuatu Women’s Centre. (VWC). (2011). Vanuatu national survey on women’s lives and family relationships . Author. https://www.dfat.gov.au/sites/default/files/womens-centre-survey-womens-lives.pdf

30.

World Health Organization. (WHO). (2001). Putting women first: Ethical and safety recommendations for research on domestic violence against women . Author. https://www.who.int/gender/violence/womenfirtseng.pdf

31.

WHO. (2005). WHO multi-country study on women’s health and domestic violence against women: initial results on prevalence, health outcomes and women’s responses . Author. https://www.who.int/reproductivehealth/publications/violence/24159358X/en/

32.

WHO. (2013). Responding to intimate partner violence and sexual violence against women: WHO clinical and policy guidelines . Author. https://apps.who.int/iris/bitstream/handle/10665/85240/9789241548595_eng.pdf?sequence=1

33.

World Health Organization, Department of Reproductive Health and Research, London School of Hygiene and Tropical Medicine, & South African Medical Research Council. (2013). Global and regional estimates of violence against women: prevalence and health effects of intimate partner violence and non-partner sexual violence . WHO. https://apps.who.int/iris/bitstream/handle/10665/85239/9789241564625_eng.pdf?sequence=1