Abstract
The mental health burden of displaced communities is enormous and ever-increasing. Community-based Mental Health and Psychosocial Support (CB-MHPSS) interventions are seen as vital in addressing this challenge, and they have been lauded as being integral in the overall multi-level Mental Health and Psychosocial Support Intervention approach in humanitarian settings. This article assumes that CB-MHPSS innately require a participatory approach to truly benefit the targeted population. It explores important benefits and challenges of using a participatory approach with CB-MHPSS interventions, as well as identifying key considerations in their design and implementation. A literature review of the PubMed database, Web of Science, The Cochrane Library of Systematic Reviews, and defined ‘grey literature’ identified 42 relevant articles. Thematic analysis identified dilemmas raised by many of the authors, including: the disconnect between using a participatory community-based approach and evidence-based medicine; using locally derived versus standardised measures; incorporating local mental health expressions and idioms into the intervention versus using standardised diagnostic classifications; empowering communities versus falling into the hands of local power dynamics and agendas; and trying to allow for sufficient time to develop relationships and build trust with the targeted community. The findings can serve to encourage reflexivity and critical thinking in the design and implementation of future CB-MHPSS interventions, which will be required to develop robust evidence that supports CB-MHPSS interventions in displaced communities.
Introduction
The relationship between mental health, psychosocial wellbeing and displacement is complex. Accumulated evidence indicates that the mental health burden is significantly higher amongst displaced populations, reflecting the increased prevalence of mental health problems during humanitarian crises (UNHCR, 2012). Numerous models and conceptual frameworks attempt to explain why displaced persons suffer a higher mental health burden than the general population (IASC, 2007; JHSP & IFRC, 2009; Miller & Rasco, 2004; Miller & Rasmussen, 2010; Tay & Silove, 2017). A pertinent example is the Adaption and Development After Persecution and Trauma (ADAPT) model, which posits that mass conflict and displacement disrupt five core psychosocial pillars: safety and security, bonds and networks, justice, roles and identities, and existential meaning (Silove, 2013; Tay & Silove, 2017). Restoration of communal psychosocial well-being subsequently requires the repair of these disruptions (Silove, 2013). Many experts have also described how issues such as poverty, social inequity, social exclusion, social deterioration and lack of sense of coherence, all of which are potentially exacerbated by mass conflict and displacement, impact on communal mental health and wellbeing (Antonovsky, 1979; Eriksson & Lindström, 2006; Patel, Flisher, Hetrick, & McGorry, 2007; Patel & Kleinman, 2003; Somasundaram, 2007). For instance, Patel and Kleinman (2003) indirectly link displacement and poverty by discussing common factors such as insecurity, hopelessness, rapid social change, and limited opportunities as a result of less education. These factors have been shown to negatively affect mental health and psychosocial outcomes, including the wellbeing of children and young people (Patel et al., 2007; Patel & Kleinman, 2003).
A community-based mental health and psychosocial approach
The term community-based (CB) is used in a wide variety of contexts and with a diverse range of meanings. McLeroy and colleagues (2003) describe four broad categories of community-based interventions, based on how interventionists employ their construction of the community: community as setting; community as target; community as agent; and community as resource. Interventions may have characteristics of one of more of these categories, illustrating the difficulty in summarising findings from across the scope of community-based interventions (McLeroy, Norton, Kegler, Burdine, & Sumaya, 2003).
In the mental health and psychosocial support (MHPSS) field, Miller and Rasco (2004) echo many by highlighting that, in conflict settings, MHPSS interventions must go beyond the provision of mainly specialised clinical treatments (Allden et al., 2009; Miller & Rasmussen, 2010; Silove, Ventevogel, & Rees, 2017; Tol et al., 2011), and be coupled with an ecological approach that acknowledges the potent impact of the social context on mental health and well-being outcomes (Miller & Rasmussen, 2010). Advocates of the ecological approach emphasize that MHPSS interventions need to focus on the provision of supportive environments that allow displaced persons the ability to restore their resource base on the personal, familial, social and material levels (Silove et al, 2017). This notion is in accord with Hobfoll’s ‘Conservation of Resources’ theory, which focuses on the effect of resource deprivation, and the shared meanings of these losses within a community as determinants of mental health outcomes (Hobfoll, 1989).
The Inter-Agency Standing Committee (IASC) guidelines for MHPSS in emergencies conceptualises the recommended multilevel MHPSS approach as a four-tiered pyramid (IASC, 2007). The first tier, or the base, represents social considerations in basic services and security; the second tier represents the strengthening of community and family supports. The third and fourth tiers represent focused non-specialised support and specialised services that target the individual respectively (IASC, 2007). This conceptually implies that MHPSS interventions are not ‘structurally sound’ unless the overall approach considers the ecological context as foundational.
Using the participatory approach with community-based MHPSS interventions
The disappointing results of many public health interventions have been attributed in part to the lack of meaningful community engagement in the planning, implementation, and evaluation of these initiatives. (Guta, Flicker, & Roche, 2013, p.432)
The underlying assumptions and values of the participatory approach and the community-based approach are compatible, thus making participatory research a ‘good fit’ for community-based MHPSS interventions (Nelson, Ochocka, Griffin, & Lord, 1998). One of the principles of applying the ecological framework is to utilise a participatory approach that subsequently allows for interventions to be integrated into existing community settings and activities (Miller & Rasco, 2004). Genuine collaboration among community members and researchers in each of the research or interventional stages – identifying the goals, research questions, methods, interventions, data analyses, interpretation and dissemination of results – is vital in this integration process by, amongst other things, recognising a shared vision of the target community, and by facilitating a mutual agreement to move towards an improved future for all (Farwell & Cole, 2001; Goodkind et al., 2017; Wallerstein & Duran, 2008). This is paramount in ensuring that the intervention intends to improve the lives of the targeted persons and their communities without doing any harm (Goodkind et al., 2017, Wallerstein & Duran, 2008; Wessells, 2008).
This article aims to guide researchers in understanding the benefits and challenges of using a participatory approach with community-based MHPSS interventions in displaced populations, and by outlining key considerations when contemplating their design and implementation.
Methods
This literature review explores the benefits and challenges of using participatory approaches in community-based MHPSS interventions for displaced populations. The literature search was conducted in September 2017 using PubMed Central, Web of Science and The Cochrane Library of Systematic Reviews. In addition, a ‘grey literature’ search was conducted using Google Scholar, Intervention, the UNHCR website, the IOM website, Humanitarian Response and MHPSS.net. Relevant primary references were also added to the search. The following search terms were used: (Mental Health OR Psychosocial) AND (Community Based OR Participat*) AND (Refugee* OR Displaced Persons) AND (Intervention OR Program* OR Evaluation OR Implementation). No restriction was set on publication year.
The inclusion criteria for the review were: (i) the article was written in English; (ii) the abstract stated that both a participatory and community approach was used in the intervention; (iii) the abstract explicitly mentioned that a mental health and/or psychosocial support intervention was undertaken, with two exceptions: the article was a relevant review; or the abstract mentioned key relevant recommendations, even if not being an intervention itself; (iv) the article was relevant to a refugee or displaced persons setting, such as a camp. Articles were excluded if (i) they only focused on the post-migratory phase (i.e. resettlement in host countries); or (ii) the benefits and/or challenges of using a participatory approach were not explicitly mentioned.
Passages that explicitly discussed the benefits and challenges of participatory approaches were highlighted. Thematic analysis was used to identify recurrent themes, which were then either framed as a dilemma (when the benefits and challenges of multiple options were considered by the authors to be related to the same decision or choices), or added to a list of further benefits and challenges. The recurrent themes were also framed in terms of key considerations. All of the dilemmas and considerations discussed in the results are derived from the reviewed literature.
Results
The original search produced 371 articles, 28 of which were grey literature. After excluding 26 repeats, 345 relevant articles remained. After screening the title and abstract (inclusion criteria two and three), 261 articles were excluded, with a further 48 excluded after full text review (inclusion criteria four and five). Six articles were included after exploring relevant primary references. Finally, a total of 42 articles were selected for thematic analysis (refer to online supplement for the related PRISMA flow diagram). Of these 42 articles, seven were reviews, 21 were intervention articles, and the remaining 14 contained relevant intervention recommendations without being a review or a primary intervention. Of the 21 intervention articles, 14 described the process of the intervention without describing the outcomes, while seven described the intervention outcomes. Four of the articles that focused on outcome results were randomised controlled trials; two were cohort studies; and one was quasi-experimental. Six of the articles described interventions based in Sub-Saharan Africa, six in the Eastern Mediterranean region, three in East and South-East Asia, two in Central America, and one in Eastern Europe. Three of the interventions were multi-sited studies across these regions. The publication dates of the interventions ranged from 1994 to 2017. The intervention strategies were markedly diverse, and ranged from multilevel approaches in numerous countries, to small scale community-based activities in a single displaced community. Please refer to Table 1 (Appendix) for a summary of further benefits and challenges of using the participatory approach.
Common dilemmas faced when using a participatory approach
A community-based versus an evidence-based approach
Kirmayer and Pedersen (2014) summarise this dilemma succinctly when they write about the tension between a public health approach grounded in biomedicine and current evidence-based practices, and a socially and culturally informed community-based approach that emphasises the importance of the social determinants of mental health, listening to local priorities, and creating endogenous solutions. Bolton et al. (2014) argue that focusing on community-based concerns using a participatory approach could create a divergence from evidence-based treatments; however, this does not consider the fact that the majority of evidence stems from high-income countries, and therefore may not be relevant in the context of the intervention (Kirmayer & Pedersen, 2014). Bolton et al. (2014) argue that a community-based approach may be justified when there is no evidence-based treatment to follow, but also caution that culturally adapted interventions could divert attention from targeting the basic social determinants of health (Metzl & Hansen, 2014). In his article titled ‘Do No Harm’, Wessells (2008) states that it is ethically important to avoid imposing outside approaches. Siriwardhana, Adikari, Jayaweera, & Sumathipala (2013) supports this ethical claim that studies should integrate the ideas and views of the participants.
Local versus standardised indicators and outcome measures
Among the articles, the studies indicated a dilemma between using ‘standardised’ or local approaches to indicators and outcome measures. While some researchers advocated for the use of local outcome measures and indicators as a means of providing more meaningful results (Ager et al., 2011; Ager & Metzler, 2012; Bolton et al., 2014; Kirmayer et al., 2014), others qualified this with the argument that local indicators limit possibilities for comparison or aggregation in meta-analyses (Tol et al., 2011). The drawback of using locally-derived outcomes could perhaps be overcome by the application of the same instrument development process in different socio-cultural contexts, such as measures developed for function impairment (Tol et al., 2011). Other researchers recommended standardised measures because they are more practical – it is easier to use a measure that has already been developed and used in other contexts (Hubbard & Miller, 2004).
Local MHPSS idioms versus standardised diagnostic classifications
Literature indicated that some local explanatory models of psychological illness are not readily translated into ‘Western diagnostic syndromes’ or integrated into Western diagnostic frameworks. Interventions that entailed participatory or community-based approaches were able to integrate local idioms into the diagnostic component of interventions (Eisenbruch et al., 2004; Miller & Billings, 1994; Vukovich & Mitchell, 2015). For example, interventionists in Guatemala came to understand that the word ‘susto’ referred to a “complex psychological response to a sudden fright” (Miller & Billings, 1994), and that, in parts of Myanmar, being ‘buried in the heart’ referred to feeling sad (Vukovich & Mitchell, 2015). Failing to capture local expressions and idioms of distress could result in the ‘under-enumeration’ of mental health problems, or failure to grasp the ‘full set’ of issues at hand (Silove et al., 2017). Understanding the ‘full set’ of mental health issues as perceived by the community is important for numerous reasons, particularly when trying to develop locally informed training materials (Bass et al., 2016) and ensuring that interventions are consistent with the cultural and educational knowledge and values of participants (Lykes & Crosby, 2014). While the literature examined these benefits, pragmatic challenges to using local indicators – such as ease, convenience and time – were also mentioned (Hubbard & Miller, 2004; Miller & Billings, 1994).
Community empowerment versus local power dynamics
A common argument for using a participatory approach in community-based MHPSS was that it was important for fostering participant empowerment (Afifi, Makhoul, El Hajj, & Nakkash, 2010; Guta et al., 2013; Pedersen, Kienzler, & Guzder, 2015; Quosh, 2013; Ventevogel, Ndayisaba, & van de Put, 2011; Wessells, 2008; Wessells & Kostelny, 2013; Williams & Thomspon, 2011). Often this argument was implicit, or just mentioned in a fleeting statement. Pedersen et al. (2015) reasoned in their review that MHPSS interventions were more likely to be beneficial if they used a participatory approach allowing for community involvement from the beginning, with such an approach ultimately promoting empowerment. One author suggested that community-based participatory research has resulted in unintended consequences, such as reinforcing local power structures through constructed hierarchies and norms (Guta et al., 2013). Authors suggest that when using a participatory approach, researchers should be cautious of local power dynamics and political agendas (Allden et al., 2009; Guta et al., 2013, Quosh et al., 2013). One study found that using a participatory approach made the researchers potentially susceptible to the agendas of the youth groups they engaged, which was not necessarily described in a negative light but rather as an important aspect of the study that warranted reflection (Chatty, Crivello, & Hundt, 2005). Afifi et al. (2010) note that ‘true’ participation might have been impeded in their intervention by cultural norms because women may have refrained from speaking out or getting involved due to their local patriarchal context.
Development of trust versus time-constraints
Researchers touted the participatory approach as essential to developing relationships and building trust among the targeted community (Afifi et al., 2010; Miller & Billings, 1994; Wessells & Kostelny, 2013; Bangpan, Lambert, Chiumento, & Dickson, 2017). Bangpan et al. (2017) in their systematic review identified that building trust and supportive relationships with the target community was important in implementing an intervention with positive outcomes. Miller and Billings (1994) stated that taking time to build local relationships was essential to advance their primary mental health intervention project with Guatemalan refugee children towards their goal.
While none of the reviewed articles discounted the importance of trust, several authors identified the need for compromise sometimes when considering the length of the proposed intervention (Afifi et al., 2010; Miller & Billings, 1994). Afifi et al. (2010) mention that using the participatory process resulted in stages of their project lasting longer than anticipated, and that this was difficult for some NGOs involved. Richters Dekker, & Scholte (2008) recognised that it would have been beneficial to ‘avoid the pitfall’ of preparing and programming too much before actually being able to develop relationships with the targeted community, but that such an approach would not have been acceptable for the donor. However, most studies urged that developing relationships was important because relationships with participants improved the intervention and data. For example, when implementing an intervention for participants living in refugee settings, a substantial amount of would likely be required during the intervention to develop relationships with participants and to better understand their ‘day-to-day’ experience (El-Khani, Ulph, Peters, & Calam, 2017).
Discussion
The dilemmas identified in designing and implementing a CB-MHPSS intervention using a participatory approach should not be seen as insurmountable barriers. However, they must be addressed in order to design and implement CB-MHPSS interventions that are community-based, locally accepted, contextually relevant, beneficent, non-maleficent, resource-efficient and evidence-based.
Theoretically, there is no reason as to why a community-based approach cannot also be an evidence-based approach. However, many of the authors reviewed saw this as a practical challenge that requires the further research. This appears to be particularly true for CB-MHPSS interventions that focus on displaced communities during their migrations as refugees, and relatively less the case than those who are dealing with the post-migratory factors of displacement. This is unsurprising considering that research on during refugee migrations on these communities tends to be done in unstable contexts and often in emergency settings (Allden et al., 2009). A Mental Health Working Group Report recommended that community-based approaches and evidence-based medicine be better connected and noted that the absence of relevant research on MHPSS interventions in emergency settings was unethical (Allden et al., 2009). The report urged the need for research to incorporate existing evidence-based practices for interventions with culturally relevant assessment methods (Allden et al., 2009). The report highlighted the impetus for more robust research of community-based MHPSS interventions in displaced populations. Silove’s overview (2017) of mental health challenges outlined a gap in the literature when it came to evidence for social interventions that improved MHPSS outcomes among refugee populations. This research will need to consider dilemmas when it comes to participatory approaches, including how to determine the indicators and outcome measures to be used and how to incorporate local expression and idioms pertaining to MHPSS.
Much of the literature found that the participatory approach promoted empowerment. But what does empowerment actually entail and what are the potential negative consequences of being in a position to ‘empower’? Gaventa (1993) suggests that participatory research “attempts to break down the distinction between the researchers and the researched, the subjects and objects of knowledge production, by the participation of the people-for-themselves in the process of gaining and creating knowledge” (cited page numbers in original article). However, Cruikshank (1999) points out that participatory research as community empowerment has been operationalised in the past, as researchers fail to consider that “the will to empower contains the twin possibilities of domination and freedom.”
Local power dynamics also exist within a targeted community, which could influence not only participation, but also negatively affect the intervention’s outcomes (Quosh, 2013). What is the advantage of using a participatory approach if we cannot guarantee equal and just participation of all community members – an important pre-requisite of empowerment (Nelson et al., 1998)? Community members might not participate in a community-based intervention because they are concerned about potential stigma, because they do not have time to participate due to work or family obligations, or because they are not aware of what the intervention involves (Hubbard & Miller, 2004). Ben-Zeev at al. (2017) discuss that creative and contemporary opportunities, such as promoting bidirectional communication via social media platforms or messaging, could be a useful strategy to overcome participatory challenges. These arguments point to the need for ongoing reflexivity, creativity and critical thinking when designing and implementing a CB-MHPSS intervention in order to have a true and just participatory approach.
Another key discussion point is the highlighted constraints of participatory research, which may require more time. Due to the interventionist agency–donor institutional relationship, time may not be an option for some researchers. Many studies found that more time with the targeted participants fostered relationships and built trust, which in turn increased participant motivation and a real sense of ownership of the research (Afifi et al., 2010; Eisenbruch et al., 2004; Tyrer & Fazel, 2014; Ventevogel et al., 2011; Wessells & Kostelny, 2013). Of course, donors may be driven by the need for short-term results (Kopinak, 2013), which can impose limits on a participatory process. However, such compromises of the participatory process must be questioned in light of the repeated warnings from many authors.
The alarming reality is that most displaced persons who need mental health and psychosocial support will not receive the appropriate care they need (Silove, 2017). Participatory CB-MHPSS interventions offer a means of maximising available resources, and increasing access and availability of MHPSS services to a larger target population (Somasundaram, 2007). Participatory CB-MHPSS interventions are well placed to be integrated into existing community-level programmes, such as existing preventive and promotional public health activities (Somasundaram, 1998), as well as inter-sectoral programmes, such as income generation or educational programmes (Blas & Kurup, 2010; Patel et al., 2007). Please refer to Table 2 (Appendix) for a number of key considerations when using the participatory approach with CB-MHPSS compiled by the authors.
Limitations
This review did not examine the outcomes of the interventions that used a participatory approach and therefore cannot address their efficacy. The thematic analysis focused on benefits and challenges and was not exhaustive so that other important considerations may have been missed. This review also did not consider how the participatory approach may differentially affect participants according to their gender, age, or relationship to marginalised or vulnerable groups.
Conclusion
This review highlights important considerations in the design and implementation of community-based MHPSS interventions in displaced populations that use a participatory approach. These considerations, which have been framed as practical dilemmas, are based on the recommendations of previous researchers and interventionists. Anticipating and overcoming the outlined challenges will serve to strengthen socially and culturally informed community-based MHPSS interventions among displaced populations.
A failure to incorporate participatory community-based interventions into the MHPSS approach is essentially a failure to recognise that displacement is experienced at a level beyond the ‘individual unit.’ Participatory and community-based interventions recognise an individual as embedded within a community (Somasundaram, 2014). A careful review of the literature leads to one conclusion: participatory CB-MHPSS interventions are no longer just optional, they are imperative.
Footnotes
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) received no financial support for the research, authorship, and/or publication of this article.
Note
The asterisks indicate the articles that were included in the review.
Appendix
lists identified recurrent themes taken from the selected literature that have not been mentioned in the aforementioned dilemmas. Key considerations when using the participatory approach with CB-MHPSS interventions.
Further benefits
Further challenges
Ownership and sustainability
Participation does not ensure fulfilment of needs
Increases community ownership of interventions (Boniface et al., 2009; Chatty et al., 2005)Using the participatory approach with the evaluation process builds local capacity for ongoing, long-term evaluation (Eisenbruch et al., 2004) Important for long-term programming (Madfis, Martyris, & Triplehorn, 2010) Encourages ownserhip and taking responsibility for protecting and supporting vulnerable groups, such as children (Wessells & Kostelny, 2013)
Although being engaged in the intervention can create a sense of meaning and enhance wellbeing, the activities may not be what many of the participants truly need (Quosh, 2013)
Utilisation and prioritisation of local resources
Participatory research potentially more susceptible in highly dynamic populations
Important for the mobilisation and utilisation of local resources, and decreasing reliance on external resources (Eisenbruch et al., 2004; Ertl et al., 2011; Lykes, 1994, as cited by Farwell & Cole, 2001; Jordans et al., 2010; Kirmayer & Pederson, 2014; Miller & Billings, 1994; Pedersen et al., 2015; Quosh, 2013; Stepakoff et al., 2006; Ventevogel et al., 2011; Weinstein, Khabbaz, & Legate, 2016, Wessells & Kostelny, 2013) Provides a platform for individuals and their collectives to mobilise their own natural capacities for recovery (Silove, 2004) Makes it possible to prioritise community needs, and to generate more accurate information on specific needs and assets (Afifi et al, 2010; Kozariæ-Kovaèiæ et al., 2002) Important for task-shifting, which is vital in low resource settings (Bolton et al., 2014) To avoid the error of over-rating the ability of outside helpers to understand and shape the recovery process, and under-rating the capacity of affected communities to draw their own resources to guide and lead the activities (Silove, Steel, & Psychol, 2006)
There may be a high number of participants not available for follow up given the dynamic nature of the population (Ager et al., 2011) Researchers should consider the reality that dynamic populations may not benefit directly from the eventual results of the intervention (Allden et al., 2009)
Avoidance of stigma
To avoid stigmatisation by using locally appropriate terminology (Quosh, 2013)
Key considerations
How will the planned intervention bridge the perceived disconnect between a community-based approach with an evidence-based approach?
Which outcomes will be measured, and how will these fulfil the criteria of being locally derived, as well as standardised?
How have local expressions and idioms been integrated into diagnostic classifications?
How does the intervention intend to empower its participants, and how will local power dynamics and agendas be taken into account?
How will the intervention ensure ‘true participation', including the equal participation of marginalised groups?
Has an adequate amount of time been allocated for relationship development and trust building within the targeted community?
Have participants been made aware that participation will not necessarily ensure a fulfilment of perceived needs?
Has consideration been given to the potential effect of a dynamic population on the intervention?
