Social Science & Medicine 130 (2015) 284e291

Contents lists available at ScienceDirect

Social Science & Medicine journal homepage: www.elsevier.com/locate/socscimed

Testing a cycle of family violence model in conflict-affected, lowincome countries: A qualitative study from Timor-Leste Susan Rees a, *, Rosamund Thorpe b, Wietse Tol c, Mira Fonseca d, Derrick Silove a a

University of New South Wales, Australia James Cook University, Australia c Johns Hopkins University, USA d Alola Foundation, Timor-Leste b

a r t i c l e i n f o

a b s t r a c t

Article history: Available online 12 February 2015

The present study examines key aspects of an emerging cycle of violence model as applied to conflictaffected countries. We focus specifically on the roles of intimate partner violence (IPV), consequent experiences of explosive anger amongst women, and associated patterns of harsh parenting. Between 2010 and 2011, we conducted a women-centred and culturally sensitive qualitative inquiry with 77 mothers drawn consecutively from a data-base of all adults residing in two villages in Timor-Leste. We over-sampled women who in the preceding whole of household survey met criteria for Intermittent Explosive Disorder (IED). Our methodology included in-depth qualitative interviews followed by a focus group with a comprehensive array of service providers. We used the NVivo software package to manage and analyse data. Our findings provide support for a link between IPV and experiences of explosive anger amongst Timorese mothers. Furthermore, women commonly reported that experiences of explosive anger were accompanied by harsh parenting directed at their children. Women identified the role of patriarchy in legitimizing and perpetuating IPV. Our findings suggest that empowering women to address IPV and poverty may allow them to overcome or manage feelings of anger in a manner that will reduce risk of associated harsh parenting. A fuller examination of the cycle of violence model will need to take into account wider contributing factors at the macro-level (historical, conflict-related, political), the meso-level (community-wide adherence to patriarchal norms affecting the rights and roles of women), and the micro-level (family interactions and gendered role expectations, individual psychological responses, and parenting). Longitudinal studies in post-conflict settings are needed to examine whether the sequence of male violence against women, mothers experience of explosive anger, and consequent harsh parenting contributes to risk of aggression and mental disorder in offspring, both in childhood and adulthood. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Armed-conflict Explosive anger Women Parenting Intimate partner violence

1. Background Epidemiological research into the psychosocial and mental health consequences of armed conflict has focused primarily on the individual as the unit of analysis (Panter-Brick, 2010). In the earlier literature, emphasis was given to the impact of traumatic events (TEs) on the prevalence of disorders such as post-traumatic stress disorder (PTSD) and depression (Mollica et al., 2001; Cardozo et al., 2003; Hinton et al., 2003; Momartin et al., 2003). In recent times,

* Corresponding author. Psychiatry Research and Teaching Unit, Level 2, Mental Health Centre, Liverpool Hospital, New South Wales 2170, Australia. E-mail address: [email protected] (S. Rees). http://dx.doi.org/10.1016/j.socscimed.2015.02.013 0277-9536/© 2015 Elsevier Ltd. All rights reserved.

however, greater attention has been given to the contributions of ongoing stressors, such as poverty and associated hardships, to adverse mental health outcomes (Miller and Rasmussen, 2010a, 2010b). There is a comparative dearth of research, however, into the effects of mass conflict on family functioning and the capacity of adults to parent their children, the focus of the present study.

1.1. The cycle of violence model A cycle of violence model offers a framework for examining the impact of mass violence on family functioning and parenting. The term “cycle of violence” was first applied in civilian settings to describe a repetitive pattern in which male perpetrators of intimate

S. Rees et al. / Social Science & Medicine 130 (2015) 284e291

partner violence (IPV) alternated between bouts of aggression and periods of remorse in which they made attempts at reconciliation (Schrager, 2011). The cycle of violence model has been expanded to take into account transgenerational effects, drawing on the assumption that children exposed to violence in the family are at heightened risk of enacting aggression in their future relationships (Kim, 2009; Thornberry et al., 2012). The model has been further adapted to apply to countries exposed to mass conflict, the notion being that adult males exposed to a range of human rights violations such as torture and related abuses are at increased risk of enacting IPV when they return to their families, initiating the sequence of adverse psychosocial effects that impact on their wives and children. The hypothesis that war exposure is a factor in the genesis of IPV has received most attention in the research literature involving male military personnel returning to high income countries such as the USA (Orcutt et al., 2003). The focus has been on manifestations of irritability, anger and agitation amongst veterans experiencing combat-related PTSD. A constellation of symptoms, however, is not limited to PTSD but commonly is associated with a range of other emotional disorders (Bell and Orcutt, 2009; Taft et al., 2012). Examining an emotional-behavioural pattern characterized primarily by explosive anger and aggression may assist in defining more clearly the pathway leading from TE exposure to violence within the family. Here we apply the DSM-IV criteria for Intermittent Explosive Disorder (IED), a category defined by repetitive episodes of uncontrollable anger in which aggression is directed at people or property (Coccaro, 2000). Such a focus is particularly relevant to the post-conflict field given growing evidence that IED is associated with exposure to traumatic events (TEs) including those related to human rights violations (Silove et al., 2009; AlHamzawi et al., 2012). Caution needs to be exercised, however, in assuming equivalence in the ontology and meaning of patterns of anger and aggression when studied across cultures. In particular, there is a risk of labelling a reaction pattern as pathological when the behavioural response is regarded as normative within the culture. We therefore apply the neutral term “explosive anger” to characterize women experiencing this response. For pragmatic reasons and to allow comparisons with other studies, we use the DSM-IV criteria for IED to define our index of the pattern of episodic anger and aggression. 1.2. Extant evidence supporting the post-conflict cycle of violence model There is accruing evidence supporting core elements of the post-conflict cycle of violence model. For example, men exposed to torture and other human rights abuses are at heightened risk of enacting IPV when they return to their families (Byrne and Riggs, 1996; Orcutt et al., 2003). The civilian literature is consistent in showing that women exposed to IPV are at higher risk of a wide range of mental disorders (Rees et al., 2011; Trevillion et al., 2012). In addition, the occurrence of IPV in families has been shown to have an adverse psychosocial impact on children (Holden and Ritchie, 1991; Levendosky and Graham-Bermann, 2000; 2001). Importantly, IPV has a deleterious effect on the mother's parenting capacity, the risk being greater when the woman is experiencing mental disorder (often a consequence of IPV), has a history of personal abuse in childhood, and is living in ongoing conditions of adversity with inadequate family and social support (Belsky, 1984; Levendosky and Graham-Bermann, 2000). These exacerbating factors are commonly encountered by women residing in lowincome, post-conflict countries. Yet the link between mothers' exposure to IPV and adverse

285

mental health outcomes amongst children in conflict-affected countries has only attracted research attention in recent times (Catani et al., 2008a; Catani, 2010; Catani et al., 2008b). In a longitudinal study in Afghanistan, family violence was a key factor in predicting the course of mental health symptoms amongst children (Panter-Brick et al., 2011). Mothers exposed to IPV in war-affected Uganda showed harsher parenting in their interactions with their children than mothers living in non-violent conjugal relationships (Saile et al., 2013). A key question that remains to be elucidated, however, is the nature of the mother's psychological response to IPV that putatively leads to harsh parenting behaviours directed at her children. One possibility is that cumulative exposure to abuse and deprivation increases the risk that the mother will respond with feelings of explosive anger, the accompanying acts of uncontrollable aggression being directed at children. That pathway has received no prior research attention. The only broadly relevant study, undertaken in a refugee clinic in the USA, found that Cambodian survivors of the Pol Pot genocide who exhibited high rates of anger and aggression, experienced disturbed relationships with significant others, particularly their children (Hinton et al., 2009). 1.3. Explosive anger in Timor-Leste The present study focuses on women in post-conflict Timor-Leste. During the prolonged armed resistance against the Indonesian occupation (1975e1999), the Timorese population was subjected to politically motivated human rights violations, including extrajudicial killings, forced imprisonment, torture, murder of family members, and amongst women in particular, rape, forced marriages and compulsory chemical sterilization (Silove, 1999; Winters, 1999; Dunn, 2012). Further abuses and deaths occurred on a large scale during the humanitarian emergency of 1999 in which Indonesiansupported militia destroyed 80% of the infrastructure of the country and temporarily displaced the majority of the population. In 2006e07, there was a further period of internal conflict resulting in deaths, violence, destruction of property and large-scale displacement (Scambary, 2009). Many families continue to live under conditions of extreme poverty in Timor-Leste. Since gaining independence in 2002, the country has remained one of the poorest in the world, with 23% of the population of 1.2 million being undernourished (National Statistics Directorate (NSD) [Timor-Leste], 2010). In our 2004 total adult household survey of 1544 Timorese adults (82% response) residing in a rural and an urban village, 38% of respondents reached the defined threshold of one attack of explosive anger a month (Fava et al., 1993; Silove et al., 2009; Rees et al., 2013). The quantum of exposure to TEs of mass conflict was the strongest predictor of explosive anger, although ongoing hardships and deprivations added to the risk. Importantly, rates of explosive anger were higher amongst women than men (Silove et al., 2009). In a six-year mixed methods follow-up study we applied a more extensive culturally adapted measure to assess a stricter definition of explosive anger based on DSM-IV IED criteria which specifies that anger is accompanied by overt acts of aggression. Our community measure exhibited sound convergence with a gold standard clinical interview for IED (Liddell et al., 2013). Importantly, women had twice the rate (12%) of explosive anger compared to men (8%) (Rees et al., 2013). Predictors of explosive anger amongst women included exposure to conflict-related TEs and ongoing hardships related to poverty. In qualitative interviews, women commonly linked their explosive anger episodes to ongoing exposure to IPV in the home (Rees et al., 2013). IPV was attributed in part to men's exposure to torture and other human rights violations during the war, and a consequent increase in male alcohol abuse (Rees et al., 2013).

286

S. Rees et al. / Social Science & Medicine 130 (2015) 284e291

1.4. The socio-cultural context Cultural factors may play an important role in determining both the rates of IPV and parenting styles (Jewkes et al., 2005; Sunar et al., 2013). In post-conflict countries such as Timor-Leste, it is common for women to be the principle caregivers of children, a responsibility that has been intensified by extensive loss of family and social supports as a consequence of recurrent periods of mass conflict. Poor access to, and ongoing taboos against family planning, and consequent high fertility rates, add to the pressures mothers experience in relation to child-bearing and child-rearing (Rees et al., 2013). In addition, women carry a substantial responsibility to meet the material needs of the family, often remaining engaged in work and income-generating activities during the perinatal and early child-rearing periods (Rees et al., 2013). A further consideration is that traditional norms sanction to some extent the use of physical violence as a disciplinary method in managing children (Bornstein, 2012; Rees et al., 2013). 1.5. Caveats The risk of stigmatizing women by focusing on responses of anger and aggression needs to be balanced against the cost of ignoring a key reaction pattern that is a major source of women's distress and disability. In studying the topic, it is important to acknowledge that in general, anger and aggression are normative responses that can be adaptive, particularly amongst women exposed to conditions of abuse and deprivation. We attempt to avoid any risk of blame being attributed to women experiencing anger and aggression by making explicit that our study is firmly grounded in a public health and feminist framework. This perspective locates women's experiences within a broader historical context of mass conflict, patriarchal values influencing gender roles and normative behaviours, and ongoing social disadvantage. Our overall objective in studying this topic is to provide data that will inform programs of prevention and intervention that will advance the status, empowerment and mental health of women (Jewkes, 2002). 1.6. Aims The aims of this qualitative study are to examine core components of the post-conflict cycle of violence model. We investigate the broader social and economic factors as well as conditions in the family, particularly the occurrence of IPV, in relation to experiences of explosive anger amongst Timorese women. We also inquire into women's experiences of explosive anger and their reports concerning the effects of that reaction on the parenting of their children. 2. Methods 2.1. Method and sampling The qualitative study was undertaken according to the Consolidated Criteria for Reporting Qualitative Research (COREQ) checklist (Tong et al., 2007). We drew a subsample from the database of an epidemiological study in which all adults living in households in a rural and an urban village of Timor-Leste were eligible for inclusion (Silove et al., 2014a). The catchment area consists of two administrative villages (sucos), one an urban settlement in the capital, Dili, the other a rural site situated approximately 1 h's drive from the urban area (Rees et al., 2013). Global positioning system (GPS) coordinates and maps generated by the Timor-Leste National Directorate of Statistics allowed us to identify all dwellings in the two

villages. The full sample consisted of 2964 persons of whom 1513 were women (response rate for women ¼ 92.6%). A subsample (77) was drawn from the full database, using a sequential selection method, structuring the final sample to ensure an approximate 4:1 oversampling of women who had met criteria for IED in the preceding survey (n ¼ 64) (Rees et al., 2013). The inclusion of a minority of women without explosive anger allowed us to gather observer perspectives from participants not personally affected by that reaction pattern. We conducted interviews in participants' homes between May, 2010 and November, 2011, each lasting up to one hour. We applied strict provisions of confidentiality and privacy based on World Health Organization guidelines for settings where IPV may be present (Garcia-Moreno et al., 2006). We sought to gather data of sufficient breadth and depth to increase the possibility that our findings may be generalized to other populations of women affected by armed conflict (Shuval et al., 2011; Rees et al., 2013). Interviews and data analysis were undertaken iteratively, with sampling ceasing when further interviews produced a high level of informational redundancy (i.e., data saturation) (Sandelowski, 1995; Rees et al., 2013). At that point, we deemed that we had obtained sufficient, rich, case-oriented data to answer deductively our research questions. 2.2. Sample characteristics Of the 77 women, 62.3% were rural residents, and the majority (68.9%) were married, the remaineder being single, widowed or living in “other conjugal-like relationships.” Approximately half (55.3%) had received no or only some primary education. The number of children per household ranged from 1 to 10. Full-time employment (48%) largely involved subsistence farming supplemented by tending a small shop or stall selling produce; 27.9% of women were subsistence farmers with no personal source of cash income and 24.1% were unemployed. There was a wide range of ages with most women aged between 25 and 54 years. 2.3. Personnel Four Timorese women with prior qualitative research experience conducted the interviews under the regular supervision of an Australian Project Director and a Timorese in-country manager. These field workers were drawn from the Maternal and Child Health Unit of the Alola Foundation, the peak non-government organization for women's health and well-being in Timor-Leste. The interviewers received two weeks' training followed by field testing of interviews. 2.4. Procedure Phase 1 of the qualitative study involved in-depth interviews with 19 Timorese women. We explored the social, economic and psychological status of women, manifestations of explosive anger, and the factors associated with that emotional reaction. In Phase 2, we applied a semi-structured interview based on Phase 1 data, designed to confirm the social, cultural and family-related factors associated with the experience of explosive anger. This phase also allowed examination of deviant or negative cases, that is, data that did not support emerging propositions (Pope et al., 2000). Phase 3 involved a professional sample focus group of workers drawn from government and non-government organizations in Timor-Leste working specifically with women in the IPV and mental health sectors. Data generated from the field interviews were discussed, clarified and checked by way of feedback from this professional sample.

S. Rees et al. / Social Science & Medicine 130 (2015) 284e291

Phase 4 involved a secondary analysis of the combined data from phases 1, 2 and 3 to identify confirmatory data (the most often described or compellingly depicted associations and manifestations) that supported our findings in relation to IPV, anger and parenting, examining for evidence of dissonant and ambiguous findings. 2.5. Analysis We used QSR Nvivo10 to assist data management and analysis. Two raters, one chief investigator and one staff member not directly involved in data gathering, independently coded the raw data using open coding and then axial coding, where the significant text related to our study questions were identified and coded/labelled, followed by a hierarchical analysis of the codes and their relationships (including setting, context, causes, and examples) (Corbin and Strauss, 2008). Dominant as well as dissonant themes and their inter-relationships were first identified by numerical prevalence as well as their expressed or verbalised predominance (Sandelowski, 2000; Tong et al., 2007). Minor differences were reconciled by consultation. Triangulation was achieved by comparing data from phases 1, 2 and 3. We have not included the ages of the women who are quoted to further protect their identities. All quotations presented in this paper are from different participants. 2.6. Partnerships, ethics and funding The study was undertaken in partnership with the Alola Foundation. The Human Ethics Committee of the University of New South Wales and the Timor-Leste Ministry of Health approved the study. The information sheet regarding the study objectives and the involvement of participants was read verbally to each potential participant because of low rates of literacy, and possible anxiety associated with signing of forms. Participants provide verbal consent to participate in the presence of a respected Timorese person (acting as a witness) and the interview staff who counter-signed the document. We followed the World Health Organization protocol for research involving questions relating to gender-based violence (Garcia-Moreno et al., 2006). To avoid the risk of retaliatory violence, the interview was presented as a study on family harmony (Harmonia Iha Familia). Because some questions referred to experiences of difficult emotions, the health professionals in our team were trained in providing immediate support if participants showed signs of distress. If required or requested, prompt referrals were made to women's support agencies and/or community mental health services with which the team had close working relationships. The study was funded by a grant from the Australian Research Council (DP0987803). 3. Results Our findings illustrate the dominant themes that emerged in relation to key components of the conflict-related cycle of violence model (Diagram 1). 3.1. Traditional patriarchal values Women made repeated reference to the patriarchal culture and its association with IPV. Women were expected to meet the needs of men without qualification, including providing timely and sufficient quantities of food under conditions of extreme poverty. A participant said:

287

“Men are sometimes angry because before or after work there is not enough food or drinks like coffee prepared for them. When he comes back home from work he feels tired and sees a mess at home and this makes him angry: talking a lot and treating their wives badly”. The following participant described the consequences for women of entrenched gender inequality: “We have to understand their needs or to be patient because men always want us to follow everything that they want. If we don't, they will say that we are not good women. Some men want women to be hard workers but some want women to be able to do everything, like collect fire wood, collect water, wash clothes, and cook and so on.”

3.2. War-related emotional problems amongst men, alcohol and IPV Most women reported that men's consumption of alcohol had increased since their exposure to the resistance war, exacerbating risk of enactment of IPV in the home. Intoxication increased the tendency for men to feel entitled to punish or belittle women. When intoxicated, men would undermine, accuse and in some cases physically abuse their female partners. Men often would compare their partners unfavourably with other women. A participant said: “Another problem is that my husband is always drunk (since the war) and says that I'm not a good wife. He always compares me with other women. It makes me very sad.” When intoxicated, men tended to accuse their partners of infidelity. A woman said: “My husband is always drunk and always suspects me with other men. I want him to sleep if he gets drunk (so that I can avoid his violence).”

3.3. IPV, explosive anger and harsh parenting Some women drew connections between the low status of women, risk of IPV and the anger they experienced, recognizing that this sequence resulted in the enactment of harsh parenting in relation to their children. A woman said: “Women feel angry because they do not understand why harm is being done to them by men. I then feel angry if the children are naughty and sometimes my sister-in-law always snitches to my brother (about her harsh treatment of her children) so that makes us angry with each other.” Another woman said: “I think women are angrier than men because we deal with children for the whole day, washing and cooking. Men only do one thing, they only deal with one job outside the home but when they come home they want everything to be ready. This makes women very angry. When I feel angry I normally threaten or say bad words, and more than that, I hit the children.”

288

S. Rees et al. / Social Science & Medicine 130 (2015) 284e291

Diagram 1. Conflict-related cycle of violence model1. 1The boxes that have an outline shadow are the areas of central focus of this paper.

A further participant reported: “Women normally express their anger by talking but some act violently like hitting their children or throwing things.”

3.4. Attitudes of women to anger and harsh parenting Women were inconsistent in their perspectives concerning acts of harsh parenting, alternating between expressions of regret and remorse, and, at other times, justifying the punishment of children as being important to teaching young people to behave correctly. Several women attributed their anger and propensity towards harsh parenting to the pressures of caring for many children in conditions of extreme hardship. In several instances, women acknowledged that the violence directed at children was of a serious nature, resulting in physical injury. A participant reported that her treatment of her children was sufficiently harsh to evince admonishment by neighbours, leading to a dispute that escalated into a communal conflict. The same participant insisted, however, that her behaviour was important to teaching the children lessons, a position that was commonly articulated: “Some community people ask why this lady is very angry. Some families ask, why are you angry at the children? If the children are lazy then we have to teach them, if not they will get used to it until they become adult.” “When the children are naughty, I feel very angry. If it happens, I always hit them.” Acting out physically towards the children also had a cathartic effect: “For myself, if the children naughty I have to hit them then I can feel calm. Other women maybe they tell bad words, hit or treat the children badly.” The burden of supporting large numbers of children was experienced as a major stressor by women. A woman said: I have twelve children. I don't have enough money to support them to go to school. Some of them are attending school but some are not. … School is very expensive. I feel angry because the children do not listen and do not follow me when I ask them to do some work.

Personal coping with anger. Recognising the harm associated with their acts of aggression, some women with explosive anger articulated personal strategies to avoid or mitigate episodes of anger, for example, by taking very long walks, cleaning the house and even neighbours' houses, or drinking alcohol in an effort to fall asleep. 3.5. Health care Women commonly involved health-care agencies in the problem. A woman stated: “I took my husband (because of IPV) to a health worker once when we were in the mountains. The health workers (nuns) said that he has “bad blood” and that's what makes his brain not work well. Some people said that he has “fulan lotuk” (a psychotic condition attributed to the moon) because during the crescent moon he always talks a lot and makes problems.” A number of women had sought medical advice specifically for their explosive anger but most reported that community health centre doctors did not understand the nature of the problem. A woman reported: “I was so angry that I was sent to a doctor who told me that the anger could kill me because of hypertension … but the medicine did not work.” Other women reported that difficulties in accessing health care prevented them from seeking assistance for their explosive anger. A woman said: “I get angry. My anger is also about the kids running around and having to care for them. If my neighbours hit my children for being naughty I get really mad and have to hit their children in return. None of our family will come to stay with us because of my anger. I did things like throw all their clothes out. I need some drugs to stop me being angry. There is no doctor in Hera and I cannot go to the clinic because I have the kids and can't get there.”

3.6. Women's attempts to seek redress for IPV Several women indicated that they no longer accepted the

S. Rees et al. / Social Science & Medicine 130 (2015) 284e291

tradition of patriarchy that allowed men to act violently against their partners. A participant described her attempts to obtain assistance from the police to stop her husband's violence: “I went to the police two times and to the chefe aldea (the village chief) three times to solve the problem.” In her lengthy description of the process that followed, she detailed how her initiative had proved fruitless because of the intercession of local community leaders who supported the husband and insisted that he had an obligation to remain in the household. The consequence, according to the respondent, was that the status quo remained because: “he (the husband) is forced by local leaders (to remain in the home) but not from his own decision.”

4. Discussion Our findings build on evidence from our past studies in TimorLeste that showed a general association between exposure to conflict-related violence and explosive anger, by showing that specific conditions in the family, particularly exposure to IPV, contribute to this reaction pattern amongst women. In addition, our qualitative data suggest that mothers' experience of explosive anger may be associated with harsh parenting directed at their children. 4.1. Strengths and limitations The study was conducted in accordance with COREQ guidelines for qualitative research (Tong et al., 2007). The sample was relatively large for a qualitative study of this kind and participants were selected serially from a comprehensive database of adults residing in urban and rural communities typical of those exposed to prolonged conflict in Timor-Leste. We note, however, that we weighted our sample to include a higher proportion of women with explosive anger drawn from the 12% of those meeting IED criteria in the preceding household survey. As such, it is important not to generalize the findings to Timorese families as a whole. Moreover, restriction of the sample to two villages underscores the importance of replicating the study in other localities in Timor-Leste and further afield. Phase 3 involved a focus group of Timorese professionals with extensive experience working with women and families. Crosschecking with this group formed part of a process of triangulation, a method that allows confirmation of qualitative findings of this type (Creswell et al., 2003). We did not study men directly, basing our assessments on the reports of mothers alone. In a prior quantitative study, however, we found strong links between the experience of conflict-related traumatic events and explosive anger amongst men (Silove et al., 2009). We did not study directly the impact of maternal parenting on children's mental health, development or behaviour, a vital next step in examining the transgenerational component of the cycle of violence model. 4.2. The salience of IPV in Timor-Leste IPV appears to be a widespread problem in Timor-Leste, with a United Nations Population Fund (UNFPA) study reporting that 43% of women had experienced that form of abuse in the previous year (UNFPA, 2012). In our past studies, Timorese women attributed IPV

289

to several factors, including men's exposure to human rights abuses during the war, their consequent increase in alcohol consumption, and, more broadly, the prevailing patriarchal attitudes that sanction violence against women (Rees et al., 2013). Social change in TimorLeste is being accompanied by a growing rejection of patriarchy and the sense of legitimacy men have claimed in the past to enact IPV in the home. Not unexpectedly, however, resistance to change persists at the community level. As indicated by our data, even representations to the police by women can be over-ridden by local leaders who ensure that the status quo in the family is maintained. 4.3. The cultural significance of explosive anger We avoided assigning a psychiatric label (IED) to women by applying the non-diagnostic term of explosive anger. We did, however, use the criteria of IED to distinguish women with and without explosive anger, allowing at least broad benchmarking of our findings with studies applying that definition in other settings (Silove et al., 2014b). It is evident that much further work is needed to define the universal and culture-specific characteristics of explosive anger as a reaction pattern across contexts. Available evidence suggests that there are major cross-country differences in the prevalence and demography of IED (Kessler et al., 2006). For example, in countries such as the USA, men have higher rates of IED than women, and the peak age of onset is in adolescence (Coccaro, 2000). An important question that remains is the extent to which explosive anger experienced by Timorese women represents a normative and adaptive response as opposed to one that may require mental health service intervention. In a previous report, we found that women regarded symptoms of explosive anger as a form of sickness, commonly leading them to seek medical care. As shown by the data reported here, women had difficulty accessing health care or received ineffective assistance for their core symptoms when attending community health clinics (Rees et al., 2013). Viewed from a broader perspective, the constellation of explosive anger and aggression we identified appears to be firmly grounded in the social and family matrix in which women live in Timor-Leste. As such, the construct of explosive anger we measured may be best regarded as a reaction pattern that bridges the domains of mental health and the social and cultural worlds, rather than as solely reflecting an expression of psychopathology. In particular, exposure to past mass conflict, conditions of extreme poverty, and the occurrence of intra-familial violence all emerged as key to understanding the sources of anger amongst mothers. 4.4. Social interventions and the status of women Our observations suggest that a suite of related social interventions aimed at empowering women is essential to overcoming the conditions that lead to anger amongst Timorese mothers. Policies and practices are needed that challenge patriarchal gender values, promote women's control over their fertility, and create an environment of opportunity for women so that they can participate fully in socio-economic development, thereby assuming an active and independent role in overcoming conditions of poverty in which many live. Importantly, challenging mores that legitimize IPV and ensuring that there is a systematic approach to curtailing the problem is essential to reducing a key pathway leading to feelings of frustration and anger amongst women (Moser, 2002; Niner, 2012). Promoting respectful relationships and conditions of safety in the family are important components. The impetus for change towards reducing gender inequalities and improving the status of women in Timor-Leste has received substantial support by government policies and legislation in recent

290

S. Rees et al. / Social Science & Medicine 130 (2015) 284e291

years. In particular, the objective of overcoming violence against women has been given a high national priority (Bye, 2005; UNFPA, 2012). 4.5. Implications for parenting interventions Although overcoming gender inequality and promoting the empowerment of women are fundamental to progress in countries such as Timor-Leste, parenting programs at the family level may make a contribution to reducing the immediate trans-generational effects of the cycle of violence. There is growing evidence that the involvement of fathers in parenting programs may enhance their effectiveness, providing a platform for engendering mutual understanding of the harm associated with men's violence in the family (Pease, 2012; Sunar et al., 2013; Panter-Brick et al., 2014). Parenting programs need to be informed by and adapted to each social and cultural setting. For example, in Timor-Leste, it is vital to take into account the pressures on women that arise from conditions of severe poverty, and the duress caused by having too many children to care for under conditions of poor family and social support (Rees et al., 2013). In addition, our findings suggest that in societies with strong traditional roots, the approach to parenting programs needs be sensitive to the extant mores related to child discipline. Our findings indicated that although Timorese women recognized that their experience of explosive anger was at times accompanied by excessively harsh approaches to parenting, several respondents also maintained traditional beliefs about the legitimacy of using physical punishment to teach children appropriate behaviour (Bornstein, 2012). Consequently, the substance and focus of programs should be guided by the complex mix of factors that shape parenting practices in each setting, the ultimate aim being to facilitate the capacity of parents to establish a joint approach to creating a family environment of respect and safety that promotes the well-being and optimal psychosocial development of children. 4.6. Clinical interventions Given the severe limitations in mental health resources in countries such as Timor-Leste, the capacity to provide direct clinical interventions for women experiencing explosive anger is highly constrained. Individual interventions will need to be selective, focusing on mothers experiencing severe and disabling forms of explosive anger, particularly where there is a high risk to the woman and/or her children. As yet, there is no evidence for the efficacy of specific treatments for explosive forms of anger in transcultural settings of this type. Our data suggest, however, the potential value of shifting the focus of mental health interventions in low-income post-conflict environments from the conventional approach of treating individuals, towards identification of and intervention for at-risk family units (Panter-Brick, 2010). 4.7. Developing an empirical foundation for the cycle of violence model In high income countries, there is evidence that children exposed to parental maltreatment have an increased risk of developing aggressive, deviant and criminal behaviour during adolescence and adulthood (Widom and Maxfield, 2001). Clearly, there are a range of direct and indirect pathways within families that may build either resiliency or generate vulnerability to mental or behavioural disturbances in offspring. Our findings point towards one important pathway involving IPV, mothers' experience of explosive anger, and consequent harsh parenting in generating a potential cycle of violence. Other contributing factors need to be considered. It is important to take into account macro-level forces

such as economic (lack of employment opportunities) and political instability and conflict that impinge directly and indirectly on the family. Women in Timor-Leste were directly exposed to the traumatic events of war, experiences that may add to the effects of IPV to increase their feelings of anger. Children may have been directly exposed to TEs during the period of conflict and are also subject to the direct effects of social and communal adversity in the postconflict environment. Traumatic stress reactions in children may be associated with oppositional tendencies that interact with the mother's predisposition to respond with anger in escalating conditions of parent-child conflict. Finally, from a biological perspective, it is important to consider the possible epigenetic effect of maternal exposure to violence on the neurodevelopment of children, a potential contributor to the multiple pathways creating risk of violent behavioural patterns emerging in later life (Shonkoff et al., 2012). The evidence base testing all these contributing factors will be strengthened by the initiation of longitudinal studies, ideally commencing in the perinatal period in low-income, postconflict societies. Ultimately, the goal of this direction in research is to demonstrate that a constellation of factors in high risk families increases the propensity to violence amongst offspring, potentially contributing to conditions of longer-term social instability and risk of future communal violence commonly observed in fragile states exposed to prior conflict. It is noteworthy that Timor-Leste experienced a further period of conflict in 2006e07, in this instance, involving factions within the country (Scambary, 2009; Silove et al., 2009; Silove et al., 2014b). Youth and young adults were prominent in the ensuing street violence. Whether the cycle of violence model offers a partial explanation for this pattern of recurrent violence remains to be established. In pursuing the study of the cycle of violence model, it is important to avoid a deterministic framework, particularly when evaluating the role of family relationships in the genesis of reactions such as anger and aggression. Interpersonal relationships are interactional, requiring a close examination of the dynamic within spousal and parent-offspring dyads in understanding emotional reactions and parenting behaviours. In addition, mothers and children may be capable of exhibiting high levels of resiliency, surviving and adapting to conditions of past and ongoing adversity in ways that allow them to function effectively, without manifesting mental disorder or associated aggressive tendencies. Understanding the factors that promote resilience within families in a manner that contributes to peace-building at a community level is as important as tracing pathways that link past adversity to future mental disturbances (Eggerman and Panter-Brick, 2010). Funding Funding was provided by the Australian Research Council (DP0987803). Acknowledgements Alola Women's Foundation Timor-Leste. Ms Elisa Savio, Alola Women's Foundation and University of New South Wales, TimorLeste. References Al-Hamzawi, A., Al-Diwan, J.K., Al-Hasnawi, S.M., Taib, N.I., Chatterji, S., Hwang, I., Kessler, R.C., McLaughlin, K.A., 2012. The prevalence and correlates of intermittent explosive disorder in Iraq. Acta Psychiatr. Scand. 126 (3), 219e228. Bell, K.M., Orcutt, H.K., 2009. Posttraumatic stress disorder and male-perpetrated intimate partner violence. JAMA J. Am. Med. Assoc. 302 (5), 562e564. Belsky, J., 1984. The determinants of parenting: a process model. Child Dev. 55 (1),

S. Rees et al. / Social Science & Medicine 130 (2015) 284e291 83e96. Bornstein, M.H., 2012. Cultural approaches to parenting. Parenting 12 (2e3), 212e221. Bye, H.H., 2005. The Fight against Domestic Violence in East Timor. Forgetting the Perpetrators. In: Masters in Peace and Conflict Transformation. Faculty of Social Sciences, University of Tromso. Byrne, C.A., Riggs, D.S., 1996. The cycle of trauma: relationship aggression in male Vietnam veterans with symptoms of posttraumatic stress disorder. Violence Vict. 11 (3), 213e225. Cardozo, B.L., Kaiser, R., Gotway, C.A., Agani, F., 2003. Mental health, social functioning, and feelings of hatred and revenge of Kosovar Albanians one year after the war in Kosovo. J. Trauma. Stress 16 (4), 351e360. Catani, C., 2010. War at homeea review of the relationship between war trauma and family violence. Verhaltenstherapie 20, 19e27. Catani, C., Jacob, N., Schauer, E., Kohila, M., Neuner, F., 2008a. Family violence, war, and natural disasters: a study of the effect of extreme stress on children's mental health in Sri Lanka. BMC Psychiatry 8 (1), 33. Catani, C., Schauer, E., Neuner, F., 2008b. Beyond individual war trauma: domestic violence against children in Afghanistan and Sri Lanka. J. Marital Fam. Ther. 34 (2), 165e176. Coccaro, E.F., 2000. Intermittent explosive disorder. Curr. Psychiatry Rep. 2 (1), 67e71. Corbin, J., Strauss, A., 2008. Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. Sage, CA. Creswell, J., Clark, V., Gutmann, M., Hanson, W., 2003. Advanced Mixed Methods Research Designs. In: Handbook of Mixed Methods in Social and Behavioural Research. Sage, CA, pp. 209e240. Dunn, J., 2012. Genocide in East Timor. In: Totten, Samuel, Parsons, William S. (Eds.), Centuries of Genocide: Essays and Eyewitness Accounts. Routledge, New York, pp. 279e315. Eggerman, M., Panter-Brick, C., 2010. Suffering, hope, and entrapment: resilience and cultural values in Afghanistan. Soc. Sci. Med. 71 (1), 71e83. Fava, M., Rosenbaum, J.F., Pava, J.A., McCarthy, M.K., Steingard, R.J., B. E, 1993. Anger attacks in unipolar depression, part 1: clinical correlates and response to fluoxetine treatment. Am. J. Psychiatry 150, 1158e1163. Garcia-Moreno, C., Jansen, H.A.F.M., Ellsberg, M., Heise, L., Watts, C.H., 2006. Prevalence of intimate partner violence: findings from the WHO multi-country study on women's health and domestic violence. Lancet 368 (9543), 1260e1269. Hinton, D., Hsia, C., Um, K., Otto, M.W., 2003. Anger-associated panic attacks in Cambodian refugees with PTSD; a multiple baseline examination of clinical data. Behav. Res. Ther. 41 (6), 647e654. Hinton, D.E., Rasmussen, A., Nou, L., Pollack, M.H., Good, M.-J., 2009. Anger, PTSD, and the nuclear family: a study of Cambodian refugees. Soc. Sci. Med. 69 (9), 1387e1394. Holden, G.W., Ritchie, K.L., 1991. Linking extreme marital discord, child rearing, and child behavior problems: evidence from battered women. Child Dev. 62 (2), 311e327. Jewkes, R., 2002. Intimate partner violence: causes and prevention. Lancet 359 (9315), 1423e1429. Jewkes, R., Penn-Kekana, L., Rose-Junius, H., 2005. “If they rape me, i can't blame them”: reflections on gender in the social context of child rape in South Africa and Namibia. Soc. Sci. Med. 61 (8), 1809e1820. Kessler, R.C., Coccaro, E.F., Fava, M., Jaeger, S., Jin, R., Walters, E., 2006. The prevalence and correlates of DSM-IV intermittent explosive disorder in the National Comorbidity Survey Replication. Arch. Gen. Psychiatry 63 (6), 669e678. Kim, J., 2009. Type-specific intergenerational transmission of neglectful and physically abusive parenting behaviors among young parents. Child. Youth Serv. Rev. 31 (7), 761e767. Levendosky, A.A., Graham-Bermann, S.A., 2000. Trauma and parenting in battered women. J. Aggress. Maltreatment Trauma 3 (1), 25e35. Levendosky, A.A., Graham-Bermann, S.A., 2001. Parenting in battered women: the effects of domestic violence on women and their children. J. Fam. Violence 16 (2), 171e192. Liddell, B., Silove, D., Tay, K., Tam, N., Nickerson, A., Brooks, R., Rees, S., B Zwi, A., Steel, Z., 2013. Achieving convergence between a community-based measure of explosive anger and a clinical interview for intermittent explosive disorder in Timor-Leste. J. Affect. Disord. 150 (3), 1242e1246. http://dx.doi.org/10.1016/ j.jad.2013.06.006. Miller, K.E., Rasmussen, A., 2010a. Mental health and armed conflict: the importance of distinguishing between war exposure and other sources of adversity: a response to Neuner. Soc. Sci. & Med. 71 (8), 1385e1389. Miller, K.E., Rasmussen, A., 2010b. War exposure, daily stressors, and mental health in conflict and post-conflict settings: bridging the divide between traumafocused and psychosocial frameworks. Soc. Sci. & Med. 70 (1), 7e16. Mollica, R.F., Sarajli c, N., Chernoff, M., Lavelle, J., Vukovi c, I.S., Massagli, M.P., 2001. Longitudinal study of psychiatric symptoms, disability, mortality, and emigration among Bosnian refugees. JAMA J. Am. Med. Assoc. 286 (5), 546e554. Momartin, S., Silove, D., Manicavasagar, V., Steel, Z., 2003. Dimensions of trauma associated with posttraumatic stress disorder (PTSD) caseness, severity and functional impairment: a study of Bosnian refugees resettled in Australia. Soc. Sci. Med. 57 (5), 775e781.

291

Moser, C., 2002. Gender Planning and Development: Theory, Practice and Training. Routledge. National Statistics Directorate [Timor-Leste], 2010. Timor-leste Demographic and Health Survey 2009. National Statistics Directorate (NSD) [Timor-Leste], ICF Macro International Inc., Dili, Timor Leste. Niner, S., 2012. 3 Women in the Post-conflict Moment in Timor-leste. In: Security, Development and Nation-building in Timor-leste: A Cross-sectoral Assessment, vol. 32, p. 41. Orcutt, H.K., King, L.A., King, D.W., 2003. Male-perpetrated violence among Vietnam veteran couples: relationships with veteran's early life characteristics, trauma history, and PTSD symptomatology. J. Trauma. Stress 16 (4), 381e390. Panter-Brick, C., 2010. Conflict, violence, and health: setting a new interdisciplinary agenda. Soc. Sci. Med. 70 (1), 1e6. Panter-Brick, C., Burgess, A., Eggerman, M., McAllister, F., Pruett, K., Leckman, J.F., 2014. Practitioner Review: engaging fatherserecommendations for a game change in parenting interventions based on a systematic review of the global evidence. J. Child Psychol. Psychiatry 55 (11), 1187e1212. Panter-Brick, C., Goodman, A., Tol, W., Eggerman, M., 2011. Mental health and childhood adversities: a longitudinal study in Kabul, Afghanistan. J. Am. Acad. Child Adolesc. Psychiatry 50 (4), 349e363. Pease, B., 2012. The politics of gendered emotions: disrupting men's emotional investment in privilege. Aust. J. Soc. Issues 47 (1), 125. Pope, C., Ziebland, S., Mays, N., 2000. Qualitative research in health care: analysing qualitative data. Br. Med. J. 320 (7227), 114. Rees, S., Silove, D., Chey, T., Ivancic, L., Steel, Z., Creamer, M., Teesson, M., Bryant, R., McFarlane, A.C., Mills, K.L., 2011. Lifetime prevalence of gender-based violence in women and the relationship with mental disorders and psychosocial function. JAMA J. Am. Med. Assoc. 306 (5), 513. Rees, S., Silove, D., Verdial, T., Tam, N., Savio, E., Fonseca, Z., Thorpe, R., Liddell, B., Zwi, A., Tay, K., 2013. Intermittent explosive disorder amongst women in conflict affected Timor-Leste: associations with human rights trauma, ongoing violence, poverty, and injustice. PLoS One 8 (8), e69207. Saile, R., Ertl, V., Neuner, F., Catani, C., 2013. Does war contribute to family violence against children? Findings from a two-generational multi-informant study in Northern Uganda. Child Abuse Negl. 38 (1), 135e146. Sandelowski, M., 1995. Sample size in qualitative research. Res. Nurs. Health 18 (2), 179e183. Sandelowski, M., 2000. Focus on research methods-whatever happened to qualitative description? Res. Nurs. Health 23 (4), 334e340. Scambary, J., 2009. Anatomy of a conflict: the 2006e2007 communal violence in East Timor. Confl. Secur. Dev. 9 (2), 265e288. Schrager, S., 2011. The Cycle of Violence. In: Family Violence: What Health Care Providers Need to Know, p. 23. Shonkoff, J.P., Garner, A.S., Siegel, B.S., Dobbins, M.I., Earls, M.F., McGuinn, L., Pascoe, J., Wood, D.L., 2012. The lifelong effects of early childhood adversity and toxic stress. Pediatrics 129 (1), e232ee246. Shuval, K., Harker, K., Roudsari, B., Groce, N.E., Mills, B., Siddiqi, Z., Shachak, A., 2011. Is qualitative research second class science? a quantitative longitudinal examination of qualitative research in medical journals. PLoS One 6 (2), e16937. Silove, D., 1999. Health and human rights of the East Timorese. Lancet 353 (9169), 2067. Silove, D., Brooks, R., Bateman Steel, C.R., Steel, Z., Hewage, K., Rodger, J., Soosay, I., 2009. Explosive anger as a response to human rights violations in post-conflict Timor-Leste. Soc. Sci. & Med. 69 (5), 670e677. Silove, D., Ivancic, L., Rees, S., Bateman-Steel, C., Steel, Z., 2014a. Clustering of symptoms of mental disorder in the medium-term following conflict: an epidemiological study in Timor-Leste. Psychiatry Res. 219 (2), 341e346. Silove, D., Liddell, B., Rees, S., Chey, T., Nickerson, A., Tam, N., Steel, Z., 2014b. Effects of recurrent violence on post-traumatic stress disorder and severe distress in conflict-affected Timor-Leste: a 6-year longitudinal study. Lancet Glob. Health 2 (5), e293ee300. Sunar, D., Kagitcibasi, C., Leckman, J., Britto, P., Panter-Brick, C., Pruett, K., Reyes, M., Hodges, W., Zonderman, A., Yazgan, Y., 2013. Is early childhood relevant to peacebuilding? J. Peacebuilding Dev. 8 (3), 81e85. Taft, C.T., Creech, S.K., Kachadourian, L., 2012. Assessment and treatment of posttraumatic anger and aggression: a review. J. Rehabil. Res. Dev. 49 (6). Thornberry, T.P., Knight, K.E., Lovegrove, P.J., 2012. Does maltreatment beget maltreatment? A systematic review of the intergenerational literature. Trauma Violence Abuse 13 (3), 135e152. Tong, A., Sainsbury, P., Craig, J., 2007. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. Int. J. Qual. Health Care 19 (6), 349e357. Trevillion, K., Oram, S., Feder, G., Howard, L.M., 2012. Experiences of domestic violence and mental disorders: a systematic review and meta-analysis. PLoS One 7 (12), e51740. UNFPA, 2012. Gender Based Violence in Timor Leste. A Case Study. Retrieved 13.03.2012, from. http://www.unfpa.org/women/docs/gbv_timorleste.pdf. Widom, C.S., Maxfield, M.G., 2001. An Update on the “Cycle of Violence”. US Department of Justice, Office of Justice Programs, National Institute of Justice. Winters, R., 1999. Buibere: Voice of East Timorese Women. Australia, East Timor International Support Centre, Darwin: Northern Territory.