This report looks at research that assesses the effectiveness of measures taken to mitigate the impact of direct experience of armed conflict on the psychosocial and cognitive development of children aged 0-8.
War is characterised at the beginning of the twenty-first century by its impact on civilian populations. Whereas the casualties of war were once predominantly the contending armies, now eight out of ten casualties are likely to be civilians, of which children are estimated to form 90%, according to a 2002 UNICEF (United Nations Children's Fund) report. The same report highlighted that over a million children were orphaned or separated from their families, 12 million left homeless, 2 million slaughtered and 6 million seriously injured or permanently disabled, as a result of armed conflicts in the last decade of the twentieth century. Globally, some 20 million children are currently affected directly by armed conflict. While previously the impact of war on children has been equally catastrophic, the sheer scope and breadth of its recent and current impact on children is unprecedented. As well as the threat to their physical health, very young children are likely to be vulnerable from an educational, psychosocial and welfare point of view, and there are many reports of severely traumatised children who have escaped from war zones. These include children actively engaged in military activities, as a recurrent feature of current armed conflict is the use of child soldiers, some as young as seven. Evidence is strong that the psychosocial and cognitive implications for the youngest children affected by armed conflict may be particularly serious and long lasting.
Research into the development of young children in situations of armed conflict has identified a lack of evidence for the effectiveness of interventions and called for careful evaluation and action research. One in-depth review of studies of psychosocial group intervention programmes for children fleeing from armed conflict in majority world countries found that most of the authors only gave anecdotal evidence for the effects and therefore the effectiveness of the programme in terms of child outcomes.
As well as informing practice in relation to children affected by armed conflict, it is expected that the conclusions of this review will be of relevance in relation to children experiencing other stressors, such as AIDS, environmental disasters or sexual exploitation. However, the search for effective interventions along the lines described in this review, can only ever be justified alongside concentrated efforts to address and eliminate the horrifying effects of armed conflict, of which children are, totally unjustifiably, the primary victims.
The current review had two stages. The aim of the first stage of the review was to produce a systematic map of research by identifying and describing studies that examined the outcomes for children of interventions to mitigate the effects of direct experience of armed conflict on children aged 0-8. As a second stage of the review, we reviewed a smaller set of studies in-depth. To do this, we applied further criteria aimed to ensure that the research under consideration provided reliable and sufficient information for effective utilisation.
The broad review question is as follows:
How effective are measures taken to mitigate the impact of direct experience of armed conflict on the psychosocial and cognitive development of children aged 0-8?
The Review Group included academics and practitioners in care and education. This was supplemented by the Advisory Group, which included a wider range of academics and practitioners who were consulted at various stages in the procedure: the formulation of the research question, and the writing of the protocol and the draft report.
Initial work concentrated on the development of definitional statements, inclusion criteria and codes to describe studies. Armed conflict was defined as any experience of armed conflict: that is, conflicts on a continuum ranging from war between states to organised crime and large-scale violations of human rights. Inclusion criteria for the map were that the provision under study should be an intervention aiming to mitigate the effects of armed conflict on children; that the children under study should be aged eight or under and have experienced armed conflict; that the provision should be aimed at psychosocial or cognitive development or wellbeing; and that the study should be evaluative, and published in English in or after 1939.
A highly sensitive search strategy was developed based around terms describing children under eight and terms relating to armed conflict; major databases, websites and library catalogues were searched. The abstracts were scanned to make an initial decision about whether they met the inclusion criteria. Those where determination was positive or unclear were obtained, and where they still met the criteria on examination of the documents, they were keyworded using generic and review-specific keywords.
Following this exercise, a map of relevant literature was produced and further criteria were developed for the in-depth review. These were that the study should be a primary study and not a review, and should include a comparison group. Most importantly, they referred to quality of reporting. Studies were required to state the aims of the research unambiguously, specify the study design, specify the tools used to collect the data, and give details about sampling and recruitment methods. Data extraction was undertaken in pairs, using guidelines to describe and apply weight of evidence to studies. A synthesis was conducted that pulled together the characteristics, findings and weight of evidence of the studies as a whole.
The map includes 16 papers, describing 13 studies. All the studies report interventions relating to conflicts in Africa or the former Yugoslavia; most took place in the country of conflict or a neighbouring country. Most provide a range of complementary interventions, including direct interventions with children, interventions with parents and/or foster carers, and interventions with service providers.
Three of these 13 studies, contained within six reports, met the further criteria for the in-depth review. These are an evaluation of a psychosocial intervention for Bosnian refugees (Dybdahl, 2001); an evaluation of two different psychosocial interventions for Sudanese refugees in Ethiopia (Paardekooper, 2002); and the evaluation of a reorganised orphanage for Eritrean refugees (Wolff et al., 1995a. While children in our target age group were among the study populations in all three studies, only the study in Bosnia focused expressly on the target age group of our review. The context of the studies selected for the map did not vary significantly from that of the studies selected for the in-depth review.
Research methods varied. All three studies selected for the in-depth review used comparison groups; two used random allocation to these groups. The studies in Bosnia and Ethiopia were prospective; the Eritrean one was retrospective. In terms of weight of evidence, the Dybdahl study was rated high, the Paardekooper study was rated medium-high and the Wolff et al. study was rated medium.
All three studies targeted refugee children from mixed social groups in the same country who had extensive direct experience of armed conflict and flight, and who were living under exceedingly difficult material circumstances. The overall context of the three studies varied considerably, ranging from an intervention study that used mothers as the mediators of the intervention for their young children (Dybdahl, 2001b) to a study (Wolff et al., 1995a) that evaluated an intervention for children whose parents had been killed, were missing or presumed dead. In between the two was Paardekooper's (2002) study of an intervention aimed at refugee children living with one or two parents.
Paardekooper (2002) and Dybdahl (2001b) provided statistically significant evidence of a beneficial impact of interventions on children, including children in the age group 0-8, as compared with a comparison group. In both cases, the interventions found to be beneficial focused on 'normalisation' of the children's daily living situation and on strengthening their coping mechanisms. The reorganisation of the Eritrean orphanage into family groups (Wolff et al., 1995a) could also be considered a normalising strategy. In the case of both Paardekooper (2002) and Wolff et al. (1995a), the interventions involved the active participation of children in identifying coping strategies. Paardekooper found that the development of problem-focused coping strategies was more effective than emotion-focused ones. Although Paardekooper argued for the effectiveness of group interventions, the review found no studies of one-to-one interventions with which to make a comparison. However, one-to-one studies tend to be reported as case studies, which were excluded from the review, and this limited the comparisons which were possible.
The range of studies found provide little evidence of effects on children aged under four, and we cannot be sure that the conclusions described in this report are generalisable to this age group.
We consider that the three studies included in this systematic review constitute evidence that interventions can help improve aspects of psychosocial functioning in children and that the evidence is strongest for group interventions focusing on normalisation. This review also shows that valid evaluations can be attempted even in situations of continuing armed conflict, and that these therefore serve as a significant example to all those working with and/or researching interventions with young children affected by armed conflict.
If we accept that armed conflict affecting young children is not likely to be totally eliminated in the near future, then we implicitly accept that local and international state agencies, non-governmental organisations (NGOs) and individuals will continue to explore optimal ways of supporting children's development and improving their living conditions under such circumstances. From this follows the need to continue the search for effectiveness.
- Support for group interventions with young children affected by armed conflict should definitely be a policy priority, both with displaced populations and refugee communities, and in-country or out-country.
- The extent to which a proposed intervention focuses on 'normalisation' should inform policy decisions about practical support and funding. Evidence from all three studies suggests that interventions focused on 'normalisation' of the daily living conditions of children affected by armed conflict and strengthening their coping mechanisms were more successful than other types of interventions or more useful than 'usual services'. Evidence of the effectiveness of residential care for this group is less convincing.
- The involvement of non-policy-maker service users in these processes is now commonly held to be critical in developing sound interventions. This approach should encompass the perspectives of both adults and children as active or prospective recipients of interventions.
- The role of children's 'agency' in achieving significant positive results; although it is not supported unequivocally by the review's findings, this may eventually have to be taken into account in policy decisions about interventions with this target group.
- The impact of context-specific factors and the role of indigenous practitioners in designing interventions goes way beyond acting as interpreters/translators of Western/Eurocentric approaches. Further research may confirm that these factors are likely to be a decisive influence on the effectiveness of any such interventions. However, such a conclusion cannot yet be drawn.
- Up-to-date and robust evidence needs to be available to practitioners in accessible formats. Government agencies should produce and disseminate authoritative research summaries and digests via the internet, programmes of practitioner seminars, and so on. NGOs should expand their research and development capacity and invest in strengthening the capacity of their employees and the agencies they fund, to be reflective and critical users of research findings.
- In order for practitioners to be well placed to use research evidence in shaping interventions, an adequate project development phase should be built into project management plans. Such a phase would allow for surveying the available information on effectiveness.
- The importance of well-designed evaluations cannot be underestimated and funding bids should explicitly request support for formative and summative external evaluation as a matter of course. NGOs should moreover enable and support their staff to engage in ongoing monitoring and self-evaluation.
- A great deal of work remains to be done on developing pathways for optimal co-operation and improving mutual understanding of culture-specific contextual features between practitioners from the countries affected by armed conflict, particularly those in majority world countries, and those from minority world countries.
- The importance of cross-cultural understanding needs to be recognised in designing evaluative research; in the relationships between minority and majority world practitioners and researchers; and in the relationships between users and practitioners and/or researchers.
- Active encouragement and support is needed for the rigorous evaluation of interventions, employing robust methodologies. Sufficient resources and evaluation expertise should also be allocated strategically and on an international scale to develop what is currently an extremely patchy evidence base for these interventions.
- A systematic review of qualitative research and the case study literature on one-to-one interventions would bring under consideration a wider range of interventions undertaken in the minority world, where the latter strategy is more widespread.
- Further research is urgently needed on the interaction between genetic and environmental factors in responding to adversity.
- The issue of children's agency deserves further research attention, both in this and other contexts, as does the concept of resilience.
Dybdahl R (2001) Children and mothers in war: an outcome study of a psychosocial intervention program. Child Development 72: 1214–1230.
Paardekooper B (2002) Children of the forgotten war: A comparison of two intervention programmes for the promotion of well-being of Sudanese refugee children. Amsterdam: Vrije Universiteit, Academic Proefscrift.
Wolff PH, Tesfai B, Egasso H and Aradom T (1995) The orphans of Eritrea: a comparison study. Journal of Child Psychology and Psychiatry and Allied Disciplines 36: 633–644.
This report should be cited as: Lloyd E, Penn H, Barreau S, Burton V, Davis R, Potter S, Sayeed R (2005) How effective are measures taken to mitigate the impact of direct experience of armed conflict on the psychosocial and cognitive development of children aged 0–8? In: Research Evidence in Education Library. London: EPPI Centre, Social Science Research Unit, Institute of Education, University of London.