Introduction

Research on the mental health of children exposed to armed conflict has undergone significant shifts in recent years. These include a greater focus on children in low- and middle-income countries [1]; an increase in intervention studies, relative to studies assessing the impact of organized violence [2, 3••]; and an emerging consensus that armed conflict threatens children’s wellbeing both directly—through exposure to war-related violence and loss—and indirectly—through its adverse impact on the social and material conditions of everyday life [4•, 5–7, 8••].

This last shift represents a paradigm shift from the “war exposure” model that guided much of the earlier research on children and armed conflict [9–11]. In that model, cumulative exposure to war-related violence was seen as the critical determinant of children’s mental health, and war-related trauma was the paramount outcome of interest. With some notable exceptions [12, 13], limited attention was paid to the psychological impact of “daily stressors,” the persistently stressful conditions of daily life that are caused or exacerbated by armed conflict. Examples include poverty, overcrowded and unsafe housing, family violence, impaired parenting due to parental distress, lack of access to education, the marginalization of children orphaned or disabled by violence or disease, and the stigma and social exclusion experienced by former child soldiers and young survivors of sexual violence.

Numerous studies have documented the negative influence of daily stressors such as these on children’s mental health in conflict and post-conflict settings. Their contribution to the levels of distress and clinical disorder has consistently equaled or exceeded that of direct war exposure [14–18]. Moreover, in several studies, daily stressors have been found to either partially or fully mediate the influence of armed conflict on mental health among children [19, 20•, 21] as well as adults [22, 23].

This more complex or mediated model including war exposure and daily stressors does not negate the toxic effects of exposure to the violence and destruction of armed conflict. Rather, it draws attention to the multiple pathways by which organized violence may endanger children’s mental health. It also underscores the salience of ongoing stressors in children’s environments that might be targeted for change (e.g., poverty, abusive or impaired parenting). Finally, the model serves as a useful reminder that even in settings of armed conflict, there are sources of childhood trauma other than direct exposure to the violence of war. For example, family violence, which tends to increase significantly in contexts of organized violence [24, 25], has been strongly linked to PTSD symptom levels among children in Afghanistan [15, 26], Uganda [17], and Sri Lanka [15, 19]. As we discuss below, the high visibility of war-related violence such as rocket attacks, suicide bombings, and assault by armed combatants can make it easy to overlook the violence to which children may be subjected in the privacy of their homes [26, 27].

The mediated model discussed above depicts children’s wellbeing as influenced by risk factors at all levels of their social ecology [28]. However, it is also helpful to consider protective factors that may buffer children from the various sources of stress to which they may be exposed [2, 28]. Although findings have been mixed and at times contradictory [29], there is compelling evidence that supportive relationships with parents, good parental mental health, and strong peer relationships may help protect children from the adverse effects of exposure to armed conflict [26, 29–32]. Unfortunately, organized violence often weakens or destroys these critical protective resources. Parents and other attachment figures may be killed, disabled, or traumatized; schools may be damaged or become targets for military attack; and opportunities for play and friendship are often diminished as families are displaced and safe communal spaces disappear. As a result, children must not only contend with war-related violence and loss amidst the stressful conditions of everyday life; they must also cope with these threats to their wellbeing under conditions of diminished interpersonal resources [5].

Our purpose in this paper is twofold. First, we briefly summarize recent findings regarding the influence of living in war-torn societies on children’s mental health and psychosocial wellbeing. We then consider the extent to which evidence in support of a comprehensive model of risk and protective factors has actually informed interventions with war-affected children. Systematic reviews have noted that while calls for ecological approaches are widespread, most evidenced-based interventions continue to prioritize direct work with children, with limited attention to addressing ongoing stressors and resource deficits in their families and broader social environments [3••, 33, 34•]. Possible reasons for the prioritization of intrapersonal child-focused approaches are considered, and examples of recent interventions are presented which suggest that a gradual broadening of focus may be underway.

War Exposure, Daily Stressors, and Children’s Mental Health

Recent studies have confirmed, with increasing methodological rigor, the findings of earlier research on the powerful contribution of daily stressors to the mental health of children in conflict and post-conflict settings. Fernando et al. [19], for example, studied the effects of trauma exposure and daily stressors on the mental health of children in eastern Sri Lanka, a region heavily impacted by civil war and natural disaster. They found that experiences of child abuse and material deprivation predicted levels of PTSD more than strongly than war and disaster exposure. Moreover, child abuse and material deprivation partially mediated the relationship between war/disaster exposure and all mental health outcomes, including PTSD, depression, anxiety, and a locally developed measure of psychosocial distress. Inter-parental violence was also significantly related to levels of PTSD, anxiety, externalizing behaviors, and general distress.

In a recent study of Afghan families, Panter-Brick et al. [26] found that family violence and maternal mental health were both strongly linked to multiple dimensions of children’s mental health status. These findings are consistent with those of Catani et al. [15], who found high levels of family violence, including violence against children and wife beating, in their study of Afghan and Sri Lankan children. In both samples, family violence occurred with greater frequency than exposure to armed conflict, despite recurrent war-related violent incidents in the areas where study participants lived.

Palosaari et al. [21] found that political violence in Gaza impacted children’s mental health not only via direct exposure but also through an increase in psychological maltreatment by their fathers, though not their mothers. Specifically, exposure to political violence increased men’s distress as well as their use of harsh parenting behaviors, which in turn adversely impacted children’s attachment security and level of post-traumatic stress symptoms. The influence of parental psychological health was also documented in a study by Khamis [20•], who found that parental distress fully mediated the relationship between violence exposure and Palestinian children’s mental health.

Taken together, these studies suggest the particular importance of the family environment as a powerful mediator of armed conflict on children’s mental health and psychosocial wellbeing. More specifically, three aspects of the family environment seem to be particularly impactful: harsh parenting including physical and psychological maltreatment; intimate partner violence, which although not caused by armed conflict, is clearly exacerbated by it; and impaired parenting due to parental psychological distress. The data also suggest that war-related conditions of material deprivation contribute to heightened levels of chronic stress among children and form the backdrop against which they must cope with exposure to violence, political as well as familial.

The adverse effects of family violence, impaired parenting, and poverty-related stressors on children’s mental health and psychosocial development have been well documented in countries not affected by armed conflict [5, 35]. Their salience as predictors of distress among children in war-affected settings is, therefore, not unexpected. What is perhaps surprising is the relative inattention to such variables, until fairly recently, in studies of conflict-affected children. We suspect this stems from several factors, including (1) the high visibility of war-related violence relative to family violence; (2) a perception among researchers that asking about family violence is culturally taboo and potentially dangerous for participants; and (3) a dominant focus on psychological trauma in the mental health field, which has tended to overshadow interest in non-traumatic but toxic stressors such as poverty and related forms of material deprivation (e.g., unsafe and overcrowded housing, malnutrition, lack of access to basic healthcare). Researchers such as Hobfoll [36] and Kubiak [37] have long argued that the chronic stress and vulnerability stemming from resource loss and material deprivation greatly increase the likelihood that potentially traumatic experiences will actually give rise to enduring symptoms of psychological trauma; however, this empirically supported idea has been slow to take hold in studies of war and mental health.

How Has Research on Risk and Protective Factors Informed Interventions for Conflict-Affected Children?

What impact has the evidence on risk and protective factors had on the design of interventions aimed at protecting or improving the mental health of children in conflict-affected settings? That is, to what extent has the paradigm shift from a war exposure model to a comprehensive risk and protection model informed a similar shift in the design of mental health and psychosocial interventions? Such a shift would be evident, for example, in preventive interventions aimed at reducing threats and/or increasing protective resources at levels that transcend (but may also include) the individual child. For severely distressed children, evidence of a paradigm shift would include the utilization of existing healing resources within a child’s ecosystem (e.g., parents and other caregivers, peers, teachers, and/or religious leaders), in addition to or as part of whatever therapeutic methods are used directly with children.

Systematic reviews of interventions aimed at improving the wellbeing of conflict-affected children have noted several trends pertinent to our discussion [2, 3••, 33, 34•]. First, the majority of mental health and psychosocial support (MHPSS) interventions, both preventive and treatment-focused, are provided in a group format, with schools being the most common setting for program delivery. School-based group interventions hold considerable potential to increase peer support and reduce social isolation among withdrawn or otherwise isolated children. By training teachers or other local community members as implementation agents, such methods also broaden children’s access to supportive adults. And by enhancing the capacity of schools to meet the psychosocial needs of their students, school-based interventions enhance the role of schools as supportive community resources—to the extent that interventions are sustainable and integrated into the school environment (versus one-off interventions that do not develop local capacity).

School-based group interventions clearly do have an emphasis on the development of protective and/or curative resources within a key community setting. However, a common feature of most MHPSS programs, whether based in schools or other community settings, is their primary emphasis on changing variables within the child, while failing to address critical sources of ongoing risk in the environment that threaten children’s wellbeing. Reviews have documented a consistent set of intrapersonally focused activities common to evidenced-based interventions with conflict-affected children. Whether they are aimed at fostering healthy psychosocial development, preventing the development of psychopathology, or reducing symptoms of clinical distress, programs typically emphasize several of the following: emotional regulation, behavioral activation, cognitive reframing, relaxation training, creative self-expression through play and expressive arts activities, processing of traumatic experiences, improving social skills, and activities aimed at fostering greater self-esteem. These are all theoretically and empirically sound foci, and the prevention and amelioration of distress, as well as the development of key life skills, are clearly important aims. However, the findings discussed earlier regarding the salience of intra-familial and community-level stressors call into question the wisdom of prioritizing interventions that rely primarily or exclusively on “change-the-child” approaches. For example, given the powerful contribution of family violence to children’s mental health, there is clearly a compelling need for interventions aimed at reducing child maltreatment. Similarly, robust findings regarding the impact of parental distress and impairment on children’s wellbeing suggest that psychosocial support for parents may hold considerable promise as a way of supporting children. Other sources of ongoing stress represent similarly promising intervention targets, including the use of physically abusive disciplinary practices by teachers, inter-parental violence at home, and dealing with stigma and social exclusion based on any of several factors (e.g., physical disability, status as a former child soldier).

Direct work with children may play a critical role by helping them learn to better regulate their emotional reactions to toxic stressors, identify potential sources of support, and learn strategies for deescalating or avoiding threatening situations. However, the failure to address the actual sources of children’s distress seems likely to undermine the impact of child-focused interventions. In the case of particularly acute stressors, such as child abuse, interventions may unwittingly be treating symptoms of trauma that are continuously generated by ongoing violent interactions in the home. A cardinal rule of trauma treatment, and of clinical interventions generally, is to first establish safety before engaging in the work of healing [38]. While we cannot alter the unfortunate reality of children’s exposure to armed conflict, the question is whether we are adequately addressing sources of risk in the environment that are amenable to change.

As Jordans et al. [3••] noted in their recent review, effect sizes of MHPSS interventions with children in conflict and post-conflict settings have generally been modest. We argue that larger effects could be achieved if interventions also target ongoing sources of the threat to children’s wellbeing. There are certainly other factors limiting the impact of existing interventions, including insufficient adaptation of intervention methods to local cultural contexts, the challenge of implementation in chaotic settings continuously or intermittently affected by armed conflict, and insufficient training and supervision of project staff. These implementation and contextual issues all represent critical foci for interventionists. However, the failure to address ongoing stressors in the social ecology may put a ceiling on the impact of even the most well-designed and implemented projects.

Reasons for the Predominance of Child-Focused Interventions

There are likely numerous reasons why MHPSS interventions in conflict-affected settings have tended to prioritize direct work with children. Here, we briefly consider three possible factors.

  1. (1)

    The continued influence of the war exposure model. To the extent that children’s distress is assumed to stem primarily from exposure to armed conflict, child-focused clinical interventions form a logical cornerstone of any theory of change. From this perspective, trauma treatment methods make sense even in the absence of intervention components targeting other sources of distress. In light of what we know about the salience of daily stressors, however, we suggest that direct work with children is best conceptualized as one element of a comprehensive approach that also targets ongoing stressors and increases the availability of supportive resources in the family and community. Such an approach is illustrated in a recent waitlist control study by O’Callaghan et al. [39••]. They achieved significant and lasting improvements in the mental health of sexually exploited Congolese girls by combining trauma-focused cognitive behavioral therapy with livelihood support and the active engagement of the girls’ caregivers in the intervention.

  2. (2)

    Working with children is relatively easier than working with parent and families because children are readily accessible in schools and other community settings. Parents and other caregivers often work during the day and are busy with childcare or other activities in the evening. Unquestionably, this makes working with parents more challenging. Adults may also be less inclined to participate in MHPSS programs than children, who may be readily engaged by the play and expressive arts elements common to many child-focused interventions [3••, 40, 41]. However, a small but growing body of research has shown that it is possible to actively engage parents and other caregivers in MHPSS interventions in conflict-affected communities [39••, 42, 43, 44••].

  3. (3)

    It is easier to target intrapersonal variables in children such as maladaptive cognitions and emotional regulation than it is alter the use of violence against children by parents and teachers and by men against women. It is certainly challenging to address sensitive issues such as intimate partner violence and the use of harsh physical and psychological parenting practices. They may be culturally ingrained, and asking about them may evoke strong reactions, as they involve power relations within the private sphere of the family. However, because familial and community violence against children increase significantly in contexts of armed conflict, and are powerfully linked to children’s mental health, they represent important targets for change regardless of the difficulties that such efforts may entail.

Three recent randomized control studies demonstrate the feasibility of targeting family violence [44••, 45] and the use of violence by teachers [46•] with conflict-affected populations. Sim et al. [44••] were able to significantly reduce the use of harsh parenting practices among displaced Burmese families in Thailand, using a methodology they successfully replicated in post-conflict Liberia [45]. Devries et al. [46•], working with the Ugandan organization Raising Voices, evaluated an intervention aimed at reducing the use of violent disciplinary practices by teachers. The intervention achieved a significant reduction, with 17 % fewer students in the intervention group than the control group reporting past-week incidents of violence by teachers.

Conclusions

There is by now robust evidence that armed conflict affects children’s mental health and psychosocial wellbeing via multiple pathways. These include direct exposure to the violence and destruction of war and a host of indirect pathways that reflect the adverse impact of organized violence on the social and material conditions of everyday life. Critical among these indirect pathways is a subset of daily stressors related to family functioning. Armed conflict increases the risk of inter-parental violence, child maltreatment, and impaired parenting related to parental psychological distress. This suggests the importance of interventions that reduce family violence and support parents’ own mental health and psychosocial wellbeing. More broadly, the salience of daily stressors at all levels of children’s social ecology underscores the need for multilevel interventions that target ongoing environmental stressors, while also helping children heal from, or become more resilient to, the effects of toxic stress, whether directly or indirectly related to armed conflict.

MHPSS interventions have lagged behind the evidence base concerning sources of distress among war-affected children. Despite growing calls for preventive interventions that target environmental stressors and strengthen protective resources, few evidence-based interventions transcend the narrow focus on direct work with children. Although child-focused interventions, particularly when school- and group-based, may strengthen social ties among children and between children and teachers, they do not address ongoing threats to children’s wellbeing at home or in the community. While the challenges to altering stressful social conditions are formidable, recent studies have documented that toxic stressors such as harsh parenting and the use of violence by teachers can be successfully reduced. We are not suggesting that such interventions should supplant direct work with children but that they should form essential components of multilevel approaches to supporting children in conflict and post-conflict settings.