A Pilot Assessment of Psychological Factors Associated with Hurricane Sandy Preparedness in Post-Earthquake Haiti

Publication Date: 2014

Abstract

The current study examined psychological factors associated with engagement in disaster preparedness behaviors among displaced Haitians affected by Hurricane Sandy. Seventy-four semi-structured interviews were conducted in a camp for internally displaced survivors of the 2010 earthquake in metropolitan Port-au-Prince. Multiple regression analyses revealed that prior trauma (including disaster) exposure was negatively associated with preparedness. Avoidance symptoms of posttraumatic stress disorder (PTSD) were also negatively associated with preparedness, while other symptoms of PTSD and depression symptoms were not significantly related to preparedness. Trust of sources of preparedness information, disaster preparedness efficacy (perceived possession of necessary finances, time, and knowledge) and perceived financial security overall were also significant predictors. Results suggest that avoidance of traumatic stimuli, including those associated with prior disasters, may impede engagement in preparedness for future disasters and thus imply the need for integration of mental health intervention into disaster preparedness training for chronically disaster-exposed populations.

Introduction and Research Objectives

In late October 2012, Hurricane Sandy swept through Haiti, resulting in floods and landslides that killed more than 50 people, destroyed 200,000 homes, and displaced 32,000 people (IDMC, 2012; USAID 2012)¹. This event was particularly devastating for the approximately 360,000 people already living in camps for internally displaced peoples (IDPs) as a result of the January 2010 earthquake and subsequent disasters. Nearly a fifth of Port-au-Prince’s 541 earthquake IDP camps experienced flooding and shelter damage, affecting 30,000 IDPs (IDMC, 2012; USAID, 2012)². Hurricane Sandy was preceded by succession of damaging natural disasters in Haiti, including Hurricane Isaac just three months prior and an earthquake in early 2012. Such disasters have occurred against a backdrop of chronic hardship, including a cholera epidemic that has killed 7500 people since 2010, insecure camp conditions and exposure to physical and sexual violence, inadequate shelter and toilets, food-insecurity, unemployment, and political and social instability (IOM, 2012³; 2013⁴).

In light of the high frequency and extreme impact of disasters in Haiti, the implementation of preparedness measures (e.g., basic first aid, community risk assessment, evacuation route mapping and planning for connecting with family members, health and hygiene practices in emergencies, and the use early warning systems) is critical, especially for displaced individuals residing in insecure shelters. The International Organization for Migration (IOM) identifies “lack of knowledge about natural disaster risk management among the population” as one of the primary reasons for Haiti’s vulnerability to severe disaster impact (IOM, 2012³, p. 42). The international community has made attempts to bolster both knowledge and resources in Haitian populations (e.g., Red Cross/IOM training of disaster management committees (IOM, 2012³) but the extent to which preparedness information is retained by camp residents or adopted in IDP camps is unclear. One of the main challenges facing preparedness activities is building an evidence base which demonstrates success in regards to beneficiary well-being and behavior change. Much of the disaster preparedness literature is based on opinions and anecdotal evidence, rather than empirical evaluation studies (Yeager et al., 2010⁵). This gap is particularly meaningful in light of indications that people don’t always engage in disaster preparedness or mitigation activities, even when they possess sufficient resources, receive preparedness training, and/or have a history of disaster exposure (McKay, 2012⁶). For example, even after the establishment of effective evacuation procedures for IDPs, many Haitian IDPs choose not to evacuate in advance of storms and flooding – despite having experienced devastating disaster impact in the recent past (personal correspondence, IOM-Haiti DRR).

The literature is ambiguous regarding the role of prior disaster exposure in predicting disaster preparedness engagement. In their review, Kellens and colleagues (2013⁴) report that most studies find a positive relationship between hazards exposure and disaster mitigation. However, conflicting findings suggest that a prior history of disaster exposure may actually make a person less likely to prepare (e.g., Lin, Shaw, & Ho, 2008[^ Lin, Shaw, & Ho, 2008]), and that reasons for this are sometimes psychological in nature and dependent on social, cultural, and religious context (Morrissey & Reser, 2003[^Morrissey & Reser, 2003]; Reale, 2010⁷). In light of the international community’s enormous financial commitment to disaster risk reduction in Haiti (total earmarked disaster risk reduction and recovery funding for 2013 exceeds $68 million, UN OCHA, 2013⁸), examination of factors with potential to impede training effectiveness is critical.

However, little is known about psychological factors contributing to the adoption or dismissal of preparedness measures in this population. In Western samples, factors such as locus of control and perceived risk of future disasters have been associated with preparedness efforts (Lindell & Perry, 2000[^Lindell & Perry, 2000]; McClure, Walkey, & Allen, 1999[^McClure, Walkey, & Allen, 1999]; Spittal, McClure, Siegert, & Walkey, 2008[^ Spittal, McClure, Siegert, & Walkey, 2008]), but such factors have not been adequately examined in other cultural contexts, developing countries, or in IDP settings. Even less is known about the role of mental health factors. Disasters are associated with severe mental health consequences (Norris, 2001⁹; Norris et al. 2002¹⁰). Posttraumatic stress and other forms of anxiety, depression, and somatic symptoms, including physical pain and sleep disturbances, are the most common forms of distress. In a meta-analysis of disaster survivors in 80 countries 25-49% of samples reported clinically significant psychological distress, with nearly one-quarter of these experiencing severe impairment (Norris, 2001⁹; Norris et al. 2002¹⁰).

For Haitian IDPs, who have almost universally endured disaster trauma, as well as other forms of trauma and the severe stress associated with survival in displaced contexts, the influence of mental health issues such as symptoms of PTSD and depression on preparedness efforts is particularly important. Although data documenting mental health of earthquake survivors in Haiti is scarce, existing research demonstrates significant emotional and somatic distress. A 2010 population-based study of 1324 adult earthquake survivors found that 29.7% of IDP camp residents met criteria for posttraumatic stress disorder, and another 28.8% met criteria for major depressive disorder, compared to 19.1% and 21.9% for PTSD and MDD respectively in community populations (Cerda et al., 2012¹¹). Risk factors for PTSD and MDD included pre-earthquake history of trauma, injury to self or family and damage or loss of home in the earthquake, and job loss and low social support after the earthquake. Prior studies suggest that many IDPs had significant trauma exposure and were at risk for depression and/or PTSD prior to the 2010 earthquake (Fawzi et al., 2009¹²; Pierre et al., 2010¹³). IDPs remaining in camps in late 2012 when Hurricane Sandy struck represent the most vulnerable of those displaced by the 2010 earthquake; they possess the fewest resources and have the most difficulty finding durable housing solutions (IASC, 2013¹⁴) and are thus likely to be disproportionately affected by stress and trauma-related distress. Even among non-IDPs, earthquake-related exposure is associated with psychiatric morbidity. A 2011 post-earthquake household survey conducted in Haiti’s Central Plateau found that 41.7% of respondents met criteria for major depression and 6.1% endorsed current suicidal ideation. Although this study area did not directly experience damage from the 2010 earthquake, depression was associated with earthquake-related variables, including having a relative die and having others move into one’s house after the earthquake (Hagaman et al., 2013¹⁵).

The relationship between distress and disaster preparedness engagement is thus far unclear. A number of studies find positive relationships among disaster exposure, distress, and preparedness, including among US earthquake and hurricane survivors and flood survivors in the Netherlands (e.g., Nguyen et al., 2006¹⁶; Sattler, Kaiser, & Hittner, 2000[^Sattler, Kaiser, & Hittner, 2000]; Siegel, Shoaf, Afifi, & Bourque, 2003[^Siegel, Shoaf, Afifi, & Bourque, 2003]; Zaalberg, Midden, Meijnders, & McCalley, 2009[^Zaalberg, Midden, Meijnders, & McCalley, 2009]). Explanatory frameworks in these studies commonly draw on decision-theory perspectives which link disaster-related exposure and distress to perceived risk and/or vulnerability, which in turn promote increased preparedness (e.g., Lindell & Perry, 2012[^e.g., Lindell & Perry, 2012]; 1992¹⁷; see also conservation of resources stress model, Hobfoll, 1989¹⁸).

However, distress symptoms also have potential to impede preparedness engagement through a variety of pathways, thus increasing vulnerability to disaster effects (Morrissey & Reser, 2003[^Morrissey & Reser, 2003]). For instance, feelings of hopelessness and powerlessness associated with depression may make preparedness feel overwhelming or pointless (e.g., Lin, Shaw, Ho, 2008¹⁹). Additionally, for those traumatized by prior disaster exposure (such as the 2010 Haiti earthquake), avoidance of traumatic stimuli associated with PTSD (DSM-V, APA, 2013²⁰) may include efforts to avoid thinking about or preparing for future disasters. Perhaps surprisingly, these possibilities have not been thoroughly researched. The aim of this pilot study is to clarify the use of disaster preparedness measures in a Port-au-Prince vicinity camp for internally displaced peoples (IDPs) affected by Storm Sandy and to conduct a preliminary investigation of psychological factors related to the adoption of such measures. Specifically, this study collects data from focus groups and interviews conducted with a sample of IDP camp residents affected by Hurricane Sandy, with the objective of determining the extent to which mental health factors contribute to disaster preparedness engagement. This study will assess hypothesized relationships between disaster preparedness and trauma history (including disaster exposure), PTSD (with special attention to avoidance symptoms), and depression. Other variables identified in the literature as important for preparedness engagement were also assessed, namely trust in sources of preparedness information (see Samaddar, Misra, & Tatano, 2012[^Samaddar, Misra, & Tatano, 2012]), efficacy (as defined by perceived knowledge, money and time needed to prepare), and demographic variables, including gender, age, and income (e.g., Lindell & Hwang, 2008[^e.g., Lindell & Hwang, 2008]; Sattler et al., 2000²¹; see Kellens, Terpstra, De Maeyer, 2013²² for a review). This preliminary research is designed to inform future study and intervention development with potential to enhance engagement in disaster preparedness strategies in Haiti and other vulnerable populations in diverse disaster prone contexts.

General Methods

Staff and training In March 2013, the PI trained a team of six Haitian research assistants to conduct focus groups and interviews and a project manager to provide oversight. Team members were veterans of a lay mental health worker project conducted by the PI from 2010-2012 (James, Noel, Favorite, & Jean, 2012[^James, Noel, Favorite, & Jean, 2012]; James & Noel, in press[^James & Noel, in press]; James, Noel, & Jean Pierre, in press[^James, Noel, & Jean Pierre, in press]), and so had intensive training and prior experience in mental health-focused clinical and research practices, including interview and focus group research as part of needs assessments and evaluation studies. Research assistants were also students at Ensite Travay Sosyal ak Syans Sosyal (Institute of Social Work and Social Science) in Port-au-Prince, where the PI has a visiting faculty appointment. In exchange for their participation, they received field placement course credit and a financial stipend.

Study site The Dye Done IDP camp in the Tabarre neighborhood of metropolitan Port-au-Prince was selected as the research site due to its relative lack of crime, its accessibility to research staff, and its location alongside a river, resulting in considerable flood damage from Hurricane Sandy. The PI and project manager secured permission from the Dye Done “camp committee” (the camp leadership contingency) to conduct research in the camp.

Focus Group Methods The PI (with interpretation by research team members) conducted two Creole-language focus groups, the first consisting of all available camp committee members (n = 8) and the second of IDP camp residents (n = 14). Focus groups with these populations were conducted separately due to the status differential between groups, so as to protect against the possibility that responses by camp residents may be influenced by the presence of committee members (and vice-versa). The aim of both groups was 1) to identify appropriate preparedness and mitigation measures for the camp setting and determine the extent to which these have been implemented and 2) to discuss factors contributing to likelihood of implementation, including both resource-oriented and psychological components.

Committee member focus group recruitment. All committee members were invited to participate in a focus group. Each member was consented separately. After determining whether each member met recruitment criteria (age 18 or over; a resident of the Dye Done camp; living in the camp at the time of Hurricane Sandy), a research assistant read the informed consent document and members provided oral consent. No names were collected. All available committee members agreed to participate and provided consent. Participants were asked to report directly to a large tent for their focus group. Following participation in the group discussion (approximately 60 minutes), committee members were compensated with phone credit valuing $10. They also received a bottled drink and a snack.

Camp resident (non-committee member) focus group recruitment. To identify camp resident participants as randomly as possible, research assistants (working in pairs) began on opposite sides of the camp and approached the first adult visible outside of a tent or other dwelling, and then outside of every 5-10 dwellings to follow. Research assistants used a recruitment script to assess interest in participation and determine whether participants met recruitment criteria (age 18 or over; a resident of the Dye Done camp; living in the camp at the time of Hurricane Sandy). Those who expressed interest and met criteria were escorted to a location outside of earshot of other residents where research assistants read an informed consent document and participants were asked to provide oral consent. No names were collected. All residents approached expressed interest in participation and provided oral consent. Sixteen residents were invited to participate in a focus group and were provided with a time later the same day to report to a specified location. Fourteen residents arrived as planned. Following participation in the discussion (approximately 90 minutes), participants were compensated with phone credit valuing $5. They also received a bottled drink and a snack.

Analysis Focus group content was audio-recorded, transcribed, and translated by the research team. Transcripts were reviewed by the PI and project manager to identify themes.

Focus Group Results Strategies identified during focus group sessions included: listening to the radio or arranging to listen to someone else’s radio; creating an emergency kit with food, water, and medicine in case of evacuation; keeping important documents in a safe, waterproof place; arranging for disaster notifications on one’s cell phone, developing an evacuation and family meet-up plan, training children in these procedures, taking steps to stabilize one’s dwelling, and keeping important documents in safe places. Several behaviors were identified to be enacted once a hurricane warning is issued: tying down tents and digging canals. Preparedness strategies were incorporated into the structured interview schedule subsequently conducted with camp residents.

Interview Methods

Recruitment Later the same week as the focus group, an additional sample of 74 camp residents were recruited for individual interviews using the methods described above for the camp resident focus groups. Again, all residents approached agreed to participate. Interviews were conducted in a quiet, private tent space provided by the camp committee. Participants were compensated with phone credit valuing $5, a bottled drink, and a snack.

Interview schedule The interview schedule utilized the following measures. All were translated to Haitian Creole by the project manager through translation, back-translation, and revision for clarity.

Disaster preparedness: Participants were asked to state whether or not they had engaged in 10 disaster preparedness items (developed by investigators based on focus groups; informed by the Earthquake Readiness Scale by Spittal and colleagues (2008²³), see also Sattler et al., 2000²¹). See Table 1 for items and descriptive statistics. A composite disaster preparedness variable was constructed using a sum of preparedness activities, M = 6.64, SD = 1.82. Two activities were excluded from the composite score; “Teach children what to do in the event of a disaster” was excluded because it was determined that participants who did not have children tended to state “no” rather than “not applicable” in response to this item, and “Sign up to receive disaster warnings or information by text message” was excluded because it was determined that some cell phone companies automatically send warning messages so responses were determined by cell phone possession rather than preparedness engagement.

Preparedness efficacy. Participants were asked three questions taken from the Vested Interest Scale (Miller, Adame, & Moore, 2013[^Miller, Adame, & Moore, 2013]) to measure perceived efficacy to prepare for disasters. Specifically, participants were asked to rate responses to the following questions on 7-point Likert scales: 1) “How able are you to take the time to prepare for a disaster?” (M = 3.23, SD = 2.11, range: 1-7); 2) “Can you afford to buy the items needed to prepare for disaster?” (M = 2.68, SD = 2.24, range: 1-7); and 3) “How much knowledge do you have about how to prepare for a disaster?” (M = 4.78, SD = 1.85, range: 1-7). Items formed a reliable scale (a = .789) so were averaged to create a single efficacy item, M = 10.69, SD = 5.22, range: 3-21.

Trauma exposure. Using an adapted Life Events Checklist (Blake, Weathers, Nagy, Kaloupek, Charney, & Keane, 1995[^Blake, Weathers, Nagy, Kaloupek, Charney, & Keane, 1995]), participants were asked to state whether or not 12 disaster exposure items and 8 non-disaster trauma exposure items “happened to me”. See Table 2 for items and descriptive statistics. A trauma exposure composite variable was created by summing items (two items, earthquake and hurricane, were excluded because nearly every respondent endorsed having these experiences; one item, tsunami, was excluded because almost no respondents endorsed having this experience) (M = 7.53, SD = 4.33, range: 2-16).

Posttraumatic stress disorder (PTSD) symptoms. The PTSD checklist portion of the Harvard Trauma Questionnaire (HTQ), created by the Harvard Program for Refugee Trauma, aims to assess the mental health of survey and interview respondents who have experienced displacement, disaster, and war (Mollica et al., 1992²⁴). It is designed specifically for cross-cultural use and has been implemented in a wide variety of settings (Mollica et al., 2004²⁵), including Haiti both before and after the earthquake (Kolbe et al., unpublished). The following assessments used only the 4th section of the HTQ, which consists of 16 self-report questions modeled after PTSD diagnostic criteria of the DSM-IV. Participants rate how much symptoms have bothered them in the last week on a four point scale ranging from “not at all” to “extremely”. Scores are averaged for the 16 items, with those with scores of 2.5 or above meeting cut-off criteria for PTSD 1 (Mollica et al., 1992). The mean HTQ score was 2.65 (SD = .77; range: 1.19-3.94), with 66.2% of respondents scoring 2.5 or above.

In line with evidence supporting a four-factor model of PTSD in which avoidance and numbing symptoms are distinct constructs (e.g., Asmundson, Stapleton, & Taylor, 2004[^e.g., Asmundson, Stapleton, & Taylor, 2004]), and to test hypotheses regarding the influence of avoidance symptoms of PTSD, items were also summed to create four symptom cluster scores: re-experiencing/intrusion (M = 2.77, SD = .92), avoidance (M = 2.5, SD =1.10), numbing (M = 2.47, SD = .73), and hyperarousal (M = 2.79, SD = .89). See Table 3 for items, clustering, and descriptive statistics by item.

Depression symptoms. Assessment of depression utilized the 16-item Hopkins Symptom Checklist (Derogatis, Lipman, Rickels, Uhlenhuth, & Covi, 2007[^Derogatis, Lipman, Rickels, Uhlenhuth, & Covi, 2007]). Participants rated how much symptoms have bothered them in the last week on a four point scale ranging from “not at all” to “extremely”. A composite depression score was constructed by summing item scores, M = 36.77, SD = 11.95, range: 15-57. See Table 4 for items and descriptive statistics.

Credibility of disaster information. Participants were asked to rate how likely they would be to act on disaster preparedness information from eight sources (e.g., the Haitian government; family and friends) using a five-point Likert Scale. See Table 5 for items and descriptive statistics. Items formed a reliable scale (a = .870) so were averaged to create a single source credibility item, M = 3.11, SD = 1.19.

Demographic information. Gender, age, number of children, marital status, relative financial security, employment and student status, and religion, were also assessed. Seventy-four (41 female, 34 male) participants were interviewed, with mean age of 32.4 years (SD = 9.93, range: 18-63). Participants reported a mean of 2.3 children (range: 0-8) and 20% reported that they were married. Sixteen percent of participants reported being currently employed and 20.3% reported current student status. To assess relative financial well-being, participants responded to “How well off is your family compared to others in the community?” on a five point scale (1 = “Much less than”; 5 = “Much more than”) (M = 3.20, SD = .97, range: 1-5).

The survey also included additional variables designed to assess cultural and religious factors, coping, and perceived risk, for use in future analyses to test additional hypotheses not associated with the current paper.

Analysis. Data were entered into an excel file by the interviewers and sent to the PI for analysis. Multiple regression analyses were used to examine the relationship between mental health predictor variables and disaster preparedness variables, controlling for demographic factors as appropriate.

Interview Results Multiple regression analyses were used to test hypothesized mental health predictors of disaster preparedness engagement: trauma exposure, PTSD symptoms (particularly avoidance symptoms) and depression symptoms. To test the hypothesis that avoidance symptoms of PTSD are particularly relevant in predicting preparedness engagement, rather than using a single HTQ sum score, four variables representing each of the PTSD symptom clusters (re-experiencing, avoidance, numbing and hyperarousal) were included in the model. Disaster preparedness efficacy, source credibility, and relative financial security 2 were also included in the model.

The results of the regression indicated that the predictors explained 67.3% of the variance (R2 = .673, F(9,61) = 13.96, p < .001). Trauma exposure significantly predicted disaster preparedness, such that those with more exposure were less likely to engage in preparedness behaviors. The avoidance symptom cluster of PTSD surfaced as a significant predictor of preparedness, such that avoidance symptoms were associated with less preparedness behavior. However, there were no significant effects of other PTSD symptom clusters or of the depression scale. Preparedness efficacy and relative financial well-being were associated with increased preparedness, while higher perceptions of source credibility were associated with less preparedness engagement. See Table 6 for betas and significance levels.

Discussion

Analyses revealed that engagement in disaster preparedness was predicted by trauma exposure and avoidance symptoms of PTSD. There was a negative relationship between trauma exposure and disaster preparedness, such that increased trauma (including disaster exposure) was associated with decreased preparedness. Results are congruent with findings by Lin, Shaw, and Ho (2008)²⁶ showing that Taiwanese flood and landslide victims were less likely than the general public to engage in activities such as relocating, purchasing insurance, or otherwise inconveniencing oneself or taking on expense to mitigate disaster. However, in some ways, this is a counter-intuitive finding in that exposure to prior trauma, including prior disasters, might be predicted to evoke increase preparedness due to greater and more salient knowledge regarding the risk and consequences of negative disaster impact. Indeed, in their review, Kellens and colleagues (2013)²⁷ report that most studies find a positive relationship between hazards exposure and disaster mitigation. They suggest that effects of disaster exposure may be mediated by other (often psychological) variables.

While other researchers have focused on psychological factors such as perceived risk and vulnerability (Kellens et al., 2013²⁸), the current study focused on the relevance of mental health symptoms, including those potentially stemming from prior disaster-related and non-disaster trauma and stress. In these data, avoidance symptoms of PTSD (“Avoiding activities that remind you of the traumatic or hurtful event predicted decreased preparedness” and “Avoiding thoughts or feelings associated with the experience”) was the only PTSD symptom cluster significantly predictive of preparedness. Presumably, avoidance symptoms related to prior disaster trauma (such as death of close others or being trapped in the 2010 earthquake) may discourage engagement with disaster-related content of all sorts – including preparedness efforts for future disasters. Interestingly, despite an intuitive connection between hopeless and helplessness associated with depression and decreased preparedness efforts, the depression scale was not a significant predictor of preparedness in this model. Further work is needed to assess the extent to which these early indications reflect consistent patterns; if so, results imply that attention may be better focused on decreasing avoidance symptoms than on resolving feelings of hopelessness in this population.

Trust of sources of disaster information was a significant predictor of preparedness engagement. Interestingly, while prior studies have found a positive relationship between trust and disaster mitigation (e.g., Samaddar, Misra, & Tatano, 2012[^e.g., Samaddar, Misra, & Tatano, 2012]), the opposite relationship surfaced in these data, such that increased confidence in sources of information was associated with decreased preparedness. While this initially appears contradictory, it may be explained by several factors, including the exceedingly high levels of corruption and low levels of trust in Haitian society (e.g., Pierre et al, 2010²⁹), and the fact that disaster preparedness activities assessed in this study were individual and family level, rather than organizational-level efforts (such as coordinated evacuation efforts). Respondents may have perceived this kind of preparedness ass more important when authorities cannot be trusted to help or provide accurate information in times of crisis.

In line with the literature (see Kellens et al., 2013²⁸), disaster preparedness efficacy, as defined by perceived possession of necessary finances, time, and knowledge, was also a significant predictor of preparedness, as was perceived financial security overall (relative to other community members). Results suggest that both mental health and other perceptions of preparedness ability are necessary contributors to self-reported engagement.

Limitations and implications The current analyses are preliminary. Future analyses will build on the current regression model by incorporating cultural and religious variables potentially contributing to psychological processes among Haitian participants (see Pierre et al., 2010¹³). Moreover, mediation and moderation analyses will be used to assess the extent to which the relationship between trauma and disaster exposure and disaster preparedness is explained by specific symptoms of PTSD and depression. A more rigorous analysis of the qualitative focus group data will allow for further assessment of pathways by which disaster preparedness is impeded as perceived by IDP camp residents. Generalizability of findings associated with this study is limited by the relatively small sample size and restriction to a single IDP camp. Additionally, the cross-sectional nature of this study impedes determination of causality (e.g., between avoidance and decreased preparedness). A longitudinal model would allow for assessment of the predictive power of avoidance over time. Finally, disaster preparedness is assessed by self report; a behavioral measure would allow for a more accurate assessment of preparedness engagement.

Despite limitations, the current study has significant implications for intervention. Haitian IDPs continue to be at significant risk of disaster exposure; as of April 2013, 86 camps, harboring nearly 40,000 families, are considered at high risk of further hurricane impact in metropolitan Port-au-Prince alone, according to the Inter-Agency Standing Committee in Haiti (Adams & Mellicker, 2013[^Adams & Mellicker, 2013]). In particular, the current results imply the importance of involving mental health-focused intervention in disaster risk reduction (DRR) efforts. By acknowledging psychological factors such as avoidance that may impede preparedness and teaching relevant coping skills, DRR trainers may increase the likelihood that trainees will be able to fully absorb training preparedness and mitigation content and translate knowledge to action.

For chronically disaster-exposed populations such as Haitian IDPs, a joint mental health and DRR initiative makes intuitive sense from both a prevention and treatment perspective. Just as mental health intervention may improve preparedness, preparedness training is in turn likely to benefit mental health. Disasters adversely impact mental health so to the extent that disaster impact can be mitigated, disaster-related distress can be prevented or decreased (Norris, 2001⁹; Norris et al. 2002¹⁰). In the shorter term, DRR training may improve participant well-being by increasing perceived safety and efficacy to keep one’s family and property safe in the aftermath of a feared future disaster, by facilitating building of social ties, and by providing an empowering opportunity to disseminate useful information to others. Indeed, there is evidence that disaster preparedness volunteerism is associated with posttraumatic growth among earthquake survivors (Karanci & Acarturk, 2005).

In sum, by beginning to clarify the pathways by which distress may affect disaster preparedness, the current study makes an empirical contribution to the development of a theoretical framework that can inform intervention models tailored to chronically disaster exposed populations in Haiti and elsewhere.

References