Evidence-based mental health integrated disaster preparedness in Nepal and Haiti
Mental health integrated disaster preparedness (MHIDP) was inspired by an observation, made by ourselves and our colleagues in Nepal and Haiti, that some community members seemed resistant to engaging in disaster preparedness. Some Nepalis living in areas at high risk from natural hazards such as flooding and landslides seemed uninterested in putting in place even relatively easy disaster risk reduction strategies. Similarly, in makeshift settlements on hillsides in Port-au-Prince, regularly flooded by seasonal hurricanes, some community members did little to try to mitigate impact. This behaviour could not be explained purely by lack of funds or insufficient knowledge; some community members failed to adopt even low-cost or free measures such as developing household emergency action plans or the safe storage of documents. This behaviour is not specific to communities at risk in Nepal or Haiti; rather, studies have shown that households across the world are generally underprepared for disaster. S. Kohn et al., ‘Personal Disaster Preparedness: An Integrative Review of the Literature’, Disaster Medicine and Public Health Preparedness 6(3), 2012.
Those involved in emergency management have spent years asking themselves why some people seem to ignore risk. Research points to a range of reasons, including psychological biases, such as a tendency to underestimate risk, R. L. Dillon, ‘Book Review: The Ostrich Paradox: Why We Underprepare for Disasters. Robert Meyer and Howard Kunruether. Philadelphia: Wharton Digital Press, 2017’, Risk Analysis 37(12), December 2017. demographic factors such as gender, education and age, levels of trust in the preparedness efforts of others and social values and norms. C. Solberg, T. Rossetto and H. Joffe, ‘The Social Psychology of Seismic Hazard Adjustment: Re-evaluating the International Literature’, Natural Hazards and Earth System Sciences 10(8), 2010. Some individuals, such as those with physical and mental challenges, the elderly and the homeless, may face particular barriers to engaging in disaster mitigation. V. Walters and J. C. Gaillard, ‘Disaster Risk at the Margins: Homelessness, Vulnerability and Hazards’, Habitat International 44, October 2014.
In our early work with earthquake- and flood-affected communities in Haiti and Nepal, we began to wonder whether mental health difficulties might exacerbate vulnerability and interfere with preparedness. The communities we worked with were struggling, not only from exposure to recurring natural hazards, but also due to stress associated with poverty, displacement and lack of healthcare, education and employment. Pilot data indicated that symptoms of depression and PTSD were common, and those reporting higher levels of distress were less likely to prepare for disasters than other community members. L. E. James, ‘A Pilot Assessment of Psychological Factors Associated with Hurricane Sandy Preparedness in Post-earthquake Haiti’, Quick Response Grant Report series, Natural Hazards Center, Institute of Behavioral Sciences, University of Colorado, Boulder, 246, 2013. These findings are consistent with other research indicating that people with mental health concerns may be less likely to engage in preparedness. D. P. Eisenman et al., ‘Variations in Disaster Preparedness by Mental Health, Perceived General Health, and Disability Status’, Disaster Med Public Health Prep. 3(1), 2009. Mental health symptoms might interfere with preparedness in various ways, ranging from depression-related lack of motivation and low efficacy to avoidance of disaster- and preparedness-related thoughts stemming from symptoms of PTSD. Some anxiety-prone individuals may be reluctant even to discuss preparedness.
Tackling mental health and preparedness simultaneously, within a single intervention, may benefit both wellbeing and preparedness-related outcomes. Despite this potential, disaster preparedness training curricula typically do not address mental health concerns. With this in mind, we developed a three-day MHIDP intervention, working with Soulaje Lespri Moun (SLM) in Haiti and Transcultural Psychosocial Organization in Nepal. The intervention has been tested across three studies involving flood- and earthquake-affected communities in Haiti and Nepal, using randomised controlled trials and a matched cluster comparison. Funding for the study was provided by Elrha’s Research for Health in Humanitarian Crises (R2HC) programme, supported by the UK Department for International Development and the Wellcome Trust. MHIDP has been shown to be effective in increasing disaster preparedness, decreasing mental health symptoms and increasing social cohesion. It has also been associated with increased use of specific coping skills, and both disaster- and mental health-related help-seeking and help-giving.
Key elements of the approach
Engaging in-country colleagues in the development of the MHIDP model is crucial. L. Ferrer-Wreder, K. Sundell and S. Mansoory, ‘Tinkering with Perfection: Theory Development in the Intervention Cultural Adaptation Field’, Child & Youth Care Forum 41(2), 2012. In Haiti, the earliest version of the intervention was created by Leah James and founding members of SLM in the aftermath of the 2010 earthquake. L. James, J. Noel and Y. Roche Jean Pierre, ‘A Mixed Methods Assessment of the Experiences of Lay Mental Health Workers in Postearthquake Haiti’, American Journal of Orthopsychiatry, 84(2), 2014. Elements of this initial framework, including coping mechanisms drawing on Haitian stories, songs, dance and humour, and discussion around culturally specific belief systems (such as religious attributions for mental health problems and disasters) were integrated into the MHIDP model, along with specific preparedness training. Subsequently in Nepal, the Haiti-specific intervention was revised and adapted through formal and informal meetings with Nepalese team members from Transcultural Psychosocial Organization, including members of the local communities targeted for programming. Team members commented on drafts of the manual, shared ideas about experiential exercises and provided examples of culturally specific stories, including one highlighting the relationship between karmic beliefs and preparedness. Although some components used in Haiti were changed for Nepal, there were some surprising examples of cultural compatibility between the two contexts. For example, some coping exercises developed by Haitian team members were well received in Nepal.
In addition to tailoring the curriculum to the cultural context, it was important to ensure that the MHIDP intervention was implemented by local teams. This both maximises the possibility that the content will fit the community and increases the likelihood that associated messages will be well received. In both Haiti and Nepal, we provided training and supervision for lay mental health workers and early career clinicians, all of whom directly facilitated the three-day intervention.
This process of co-creation, adaptation and facilitation by local staff helps to ensure that the MHIDP intervention does not challenge or undermine existing belief systems or practices – a key consideration, not only in terms of ethical practice, but also to increase the likelihood of community acceptance and the effectiveness of the intervention. Prior research suggests that spiritual/religious narratives for explaining natural hazards such as earthquakes can exist in parallel with scientific explanations. L. E. James et al. ‘Challenges of Post-disaster Intervention in Cultural Context: The Implementation of a Lay Mental Health Worker Project in Post-earthquake Haiti’, International Perspectives in Psychology: Research, Practice, Consultation 1, no. 2, 2012. Religious attributions for disasters and spiritual and religious practices to mitigate risk are common in both Haiti and Nepal. For example, some community members in Nepal derive comfort from engaging in offerings designed to appease the gods after an earthquake or flood. Families may also use traditional, culturally specific approaches to preparedness, such as storing food on the ceiling using woven mats and a rope and pulley system.
Such pre-existing beliefs and behaviours do not preclude the adoption of new preparedness techniques, including household emergency action plans and the collaborative development of community risk and resource maps. Indeed, results from our work in both settings indicate that such beliefs and practices continued alongside the new knowledge and skills acquired during the intervention. As explained by a Haitian lay mental health worker involved in the early phases of the project: ‘If we take the time to teach them a technique to use to feel better, they will listen. They may see a Houngan (voodou practitioner) also, but in the meantime, they have a technique they can use’.
Other elements of MHIDP include small group discussion, experiential components and a peer-based framework. Small group discussions involve community members sharing opinions (and sometimes engaging in spirited debate) about topics such as local idioms of distress, beliefs about the causes of mental health symptoms and related stigma and local preparedness techniques, rather than simply receiving content from facilitators. Experiential components include opportunities to practice new coping skills, such as breathing and relaxation techniques, and interactive and engaging means of sharing content, including games. The peer-support approach entails creating opportunities for community members to provide and receive social support regarding mental health needs and to encourage collective approaches to disaster preparedness.
The intervention curriculum
Day 1 begins with establishing ground rules and explaining the importance of the peer support framework, including guided opportunities throughout the workshop to provide support to other group members. Day 1 also entails discussion about mental health and psychosocial reactions to disaster-related stress and associated coping strategies, with the purpose of teaching immediately applicable coping skills and self-soothing techniques. Utilising coping strategies during workshop discussions about stressful topics helps to encourage engagement and reduces the risk that participants may shut down or disengage in order to avoid anxiety-provoking content. Specific activities during Day 1 include self-reflection through a ‘drawing feelings’ exercise; observing bodily reactions through a mildly stress inducing game; self-calming through breathing; culturally adapted grounding, mindfulness and muscle relaxation exercises; and small group discussion about mental health symptoms. Throughout Day 2 and Day 3, participants practice, with increasing autonomy, the coping skills learned on Day 1.
On Day 2, participants continue the mental health and peer support focus through an art-based activity entailing creation of culturally specific symbols of safety. The workshop then transitions to focus on disaster preparedness, including facilitated discussions regarding the links between common attributions for disasters (natural causes, will of the gods) and preparedness motivation. Facilitators introduce common scientific explanations for disasters such as earthquakes and floods and share recommended preparedness strategies. This is done without discouraging pre-existing cultural and religious beliefs or indigenous approaches to preparedness, which participants are encouraged to maintain alongside new information.
Ending Day 2 and moving into Day 3, participants work on providing disaster- and mental health related peer support to each other, drawing on techniques associated with Psychological First Aid (PFA), and role-playing a variety of brief disaster scenarios. Day 3 also entails discussion about the impact of disasters on particularly vulnerable groups – children and people with more serious mental health challenges – and practical skills for supporting them. Finally, a ‘Tree of Hope’ exercise is used to encourage individual goal-setting and future planning, and a ‘Forest of Hope’ is constructed to demonstrate community-level resilience through collaboration. The workshop concludes with a ‘mini disaster simulation’, in which participants demonstrate skills learned over the three days. At the closing ceremony participants receive certificates and disaster preparedness materials and discuss achievements and next steps.
Mechanisms of change
The research methodology and associated mediation models have allowed us to begin to understand the potential mechanisms of change underlying intervention results. Specifically, the impact of the intervention on social cohesion and mental health symptoms appears to partially explain the increase in preparedness. Similarly, the influence of the intervention on preparedness and social cohesion partially explains the positive mental health outcomes. Feedback collected from participants after the intervention provides additional insights into the importance of the peer support framework, disaster preparedness knowledge and mental health components. When asked about the training, we received the following responses: ‘After taking this training we learned the importance of helping each other more. Without depending on any other organisation or government, we learned that we can do much better if we work together in the community’; ‘We have learned to share the information with family and friends about disaster preparedness’; ‘We were very much scared and disturbed, but because of this training, it has helped us to come out from this fear. It has brought peace in our lives.’
Given the effectiveness of this model in increasing preparedness and social cohesion and decreasing distress, we suggest MHIDP should be scaled up for use elsewhere in Nepal and Haiti, and in other countries prone to natural hazards. The intervention can be culturally adapted relatively easily. It is cost effective and can be implemented by lay mental health workers or trained clinicians, using standardised curriculum manuals. MHIDP manuals are available for Nepal and Haiti (in English, Nepali and Haitian Creole) at: http://www.elrha.org/researchdatabase/?searchTerm=community-based+disaster+mental+health+intervention&searchSubmit=searchSubmit
Those interested in similar models should also be aware of the challenges we encountered during implementation. During the initial planning stages, we met community leaders and local government officials to address scepticism about the value of an intervention focused on mental health and lowcost preparedness. Various stakeholders, including some participants, initially suggested that we should provide tractors, building materials or other material goods instead. Some project staff were directly impacted by the earthquake and flooding and were experiencing associated distress. In the aftermath of the Nepal earthquake, we incorporated staff care training modules into the intervention and emphasised regular debriefing sessions. Occasionally we noted power differentials between group members, with men and those from ethnic groups with greater influence dominating discussions. To address this, facilitators organised small group discussions with individuals of similar social status. We also struggled with how best to address culturally specific beliefs that were potentially harmful to some group members. For example, during one discussion in Nepal some members evoked karma to suggest that those who had lost family members in the earthquake were to blame for their deaths. Facilitators created space for alternative views within the group, and during breaks we checked in with those who had lost family and offered referrals for follow-up support.
On a final note, it is important to highlight the need for additional research examining the efficacy and specific components of the intervention. We need to determine whether three days is the optimal timeframe for the intervention, especially when considering the duration of effects. While many of the results were sustained many months after the intervention, the impact did diminish over time. Might an intervention of longer initial duration, or a one-day ‘booster’ mid-year, be useful? Future research should also compare a standard disaster preparedness curriculum to the mental health integrated version (in our research the intervention was compared to a waitlist control condition). It is also essential that we examine the relationship between the intervention and mental health outcomes more closely. Unlike the robust results for disaster preparedness and social cohesion across two countries and three studies, the impact of the intervention on mental health outcomes varied by context, with stronger results in communities facing more acute disasters. Regardless of the challenges and the work that remains, MHIDP has brought us one step closer to building an evidence base to determine what type of mental health and psychosocial interventions work in humanitarian crises.
Courtney Welton-Mitchell is a Research Associate with the Institute of Behavioral Science’s Natural Hazards Center, University of Colorado, Boulder. She is the co-founder and director of the Humanitarian Assistance Applied Research Group at the Josef Korbel School of International Studies, University of Denver. Leah James is Research Associate with the Natural Hazards Center. In recent years, Leah has also served in technical mental health and research roles with organisations such as the Center for Victims of Torture, International Medical Corps and the International Rescue Committee.
Comments are available for logged in members only.