Challenges Associated with Childhood Exposure to Severe Natural Disasters: Research Review and Clinical Implications

Abstract

Natural disasters cause widespread destruction, economic loss, and death, leaving children to cope with the devastating aftermath. The research literature has demonstrated that children are at risk postdisaster for negative mental health outcomes, such as posttraumatic stress disorder and depression. The purpose of this review is to highlight the challenges associated with childhood exposure to severe natural disasters and to summarize the current research on clinical interventions for children postdisaster. Specific challenges reviewed include disaster-related deaths, secondary stressors typical in postdisaster environments, disruption in social networks, and the threat of recurrent disasters. A public health framework for school-based mental health services is discussed and the need for research on effective intervention models for youth in postdisaster environments is highlighted.

Keywords:

• disasters
• youth
• mental health
• interventions

Disasters are situations or events that create a serious disruption in the functioning of a community and overwhelm its capacity to deal with the widespread loss and destruction created by the event. Researchers often distinguish between different types of disasters, including natural, technological, and mass violence (Norris et al., 2002; Silverman & La Greca, 2002). Natural disasters are further distinguished based on the hazard associated with the disaster, including geophysical (e.g., earthquakes, tsunamis), hydrological (e.g., floods), meteorological (e.g., hurricanes, tornados), and climatological (e.g., droughts, wildfires; International Strategy for Disaster Reduction [ISDR], 2006; Löw, 2009).

Between 2000 and 2005, there were an average of 395 natural disasters globally each year, which means that a disaster occurred somewhere in the world every day (ISDR, 2006). In general, the incidence of natural disasters has demonstrated an increasing trend; since the 1980s, hydrological events have tripled in their average annual occurrence and meteorological and climatological events have almost doubled (Löw, 2009). Natural disasters exact heavy human and economic tolls. Between 1991 and 2005, more than 230 million people worldwide were affected by disasters annually with an average of 56,078 people killed each year (ISDR, 2006). During the same time period, economic losses averaged $80 billion annually with notable spikes in 1995, when losses exceeded $200 billion as a result of the earthquake in Kobe, Japan, and in 2005, when losses exceeded $150 billion as a result of Hurricane Katrina, which occurred in the Gulf Coast area in the United States (ISDR, 2006).

The literature on children affected by severe natural disasters highlights a number of negative mental health consequences linked with disaster exposure. Posttraumatic stress disorder (PTSD) and depression are the most commonly observed psychological reactions in children following severe natural disasters (La Greca & Prinstein, 2002; Norris et al., 2002). In fact, anywhere between 30% to 90% of youth display severe posttraumatic stress reactions following exposure to severe natural disasters (La Greca, Silverman, Vernberg, & Prinstein, 1996; Wickrama & Kaspar, 2007). Osofsky, Osofsky, Kronenberg, Brennan, and Hansel (2009) conducted a study of over 7,000 children from four parishes in Louisiana heavily affected by Hurricane Katrina. Based on a screening tool measuring symptoms of PTSD and depression (National Child Traumatic Stress Network [NCTSN], 2005), they found that 49% of fourth through twelfth graders exceeded the cutoff for a mental health referral in the year following the disaster and 41.6% of youth exceeded the cutoff the following year.

Although PTSD and depression seem to be the most common psychological reactions to disasters, the current diagnostic formulations of those reactions are based largely on symptoms most often observed in adults (American Psychiatric Association [APA], 2000; World Health Organization [WHO], 1992). Neither of those formulations comprehensively captures the range of symptoms experienced by children following exposure to a severe natural disaster (Cohen et al., 1998). An emerging body of literature on childhood reactions to traumatic events expands on these definitions by highlighting how children of varying ages and developmental levels express psychological distress (Cohen et al., 1998; Scheeringa, Zeanah, Myers, & Putnam, 2003) following disaster exposure (NCTSN, n.d.). In young children, research has found that psychological reactions are closely tied to the reactions of their caregivers (Scheeringa & Zeanah, 2008; Swenson et al., 1996). For example, in a sample of 70 preschool-aged children and their mothers assessed six months after Hurricane Katrina, one study found that new mental health problems among children were significantly correlated with new mental health problems in their mothers (Scheeringa & Zeanah, 2008). Because preschool children may lack the language necessary to express feelings of helplessness and fear, their distress is often manifested in behavioral indicators such as loss of previously acquired developmental skills (e.g., toileting and speech), disruptions in sleep, and excessive clinginess with adults (NCTSN, n.d.). Similar reactions may occur in younger school-aged children. Additional research on school-aged youths has documented decreases in academic performance and increased inattention (La Greca & Prinstein, 2002) as well as somatic complaints such as headaches and stomachaches (Hensley & Varela, 2008; NCTSN, n.d.) and aggressive behavior (Terranova, Boxer, & Morris, 2009). Adolescents' reactions are more closely aligned with adult conceptualizations of PTSD, although academic declines (Dyregrov, 2004) and aggressive behavior have also been observed (Marsee, 2008).

Essential Traumatic Elements of Exposure to Severe Natural Disasters

It is not surprising that exposure to severe natural disasters is related to a variety of negative mental health outcomes because natural disasters often represent traumatic exposure. In fact, exposure to severe natural disasters commonly involves what is often considered the hallmark of traumatic exposure: perceived life threat (APA, 2000). Youth exposed to natural disasters often perceive their lives or the lives of loved ones to be threatened, and both perceived and actual life threat during the disaster has been shown to be a powerful risk factor for poor mental health outcomes in children (La Greca et al., 1996).

Many disaster researchers also consider exposure to loss and destruction in the immediate aftermath of the disaster to be an essential part of the traumatic experience (La Greca et al., 1996; Scheeringa & Zeanah, 2008; Weems & Overstreet, 2008). Loss of one's possessions or the destruction of one's community has the potential to seriously challenge a child's basic sense of safety and elicit the state of panic typically associated with traumatic exposure (Scheeringa & Zeanah, 2008). In fact, existing research conceptualizes exposure to disasters as a multidimensional experience that involves both life threat and experiences of loss and destruction (for reviews, see Norris et al., 2002; Silverman & La Greca, 2002). Regardless of whether studies examine exposure as a composite measure that combines life threat, loss, and exposure to destruction (e.g., Bokszczanin, 2003; Hensley & Varela, 2008) or whether studies examine these dimensions of exposure separately (e.g., La Greca et al., 1996; Thienkrua et al., 2006), findings across studies indicate a clear dose–response relationship: As exposure across these dimensions increases, emotional distress increases.

Unique Challenges Associated with Exposure to Severe Natural Disasters

Severe natural disasters present a unique set of challenges that further complicates initial traumatic exposure, including the death of loved ones, ongoing loss and disruptions associated with the community-level impact of the disaster, and the recurring threat related to geographical location and recurrent seasons of increased risk of specific natural hazards. Each of these unique challenges and their relation to child mental health are discussed in the remainder of this section.

Disaster-Related Deaths

Deaths are one of the defining features of severe natural disasters. On average, natural disasters cause 184 deaths per day; however, not all countries or all people share an equal mortality risk (The United Nations Development Programme [UNDP], 2004). More than 53% of disaster-related deaths worldwide are exacted from just 11% of individuals in developing countries (WHO, 2009). Poverty plays a major role in the impact of disasters (UNDP, 2004), even in industrialized countries like the United States (McMahon, 2007). For example, African Americans in Orleans Parish, who had a higher poverty rate than Whites (U.S. Census Bureau, 2005), had an average Hurricane Katrina mortality rate 2.5 times higher than Whites (Brunkard, Namulanda, & Ratard, 2008).

In addition to injury-related deaths during and immediately following severe natural disasters, research has also documented higher mortality rates over a longer period of time, suggesting more enduring health consequences of disasters (Armenian, Melkonian, & Hovanesian, 1998; Stephens et al., 2007; Trichopoulos, Katsouyanni, Zavitsanos, Tzonou, & Dalla-Vorgia, 1983). For example, the mortality rate in the greater New Orleans area between five and 10 months after Hurricane Katrina was 47% higher than the pre-Katrina mortality rate (Stephens et al., 2007). Aremenian et al. (1998) also documented an increase in mortality in the year following a major earthquake in Armenia, with most deaths resulting from heart disease. As noted by Stephens et al. (2007), the underlying causes of increased mortality are “complex, multifactorial, and persistent” (p. 19). In addition to increased levels of stress, damage to the medical infrastructure and professional community in postdisaster environments undoubtedly plays a role in this phenomenon (Needle, 2008).

Children who encounter the death of a loved one following a disaster are faced with the challenges of compounded disaster exposure. These children must contend with the stressors associated with the disaster and the profound grief of losing a loved one. Disaster-related death can lead to the development of traumatic grief, which occurs when the child perceives the death of someone close to him or her as traumatic (see Mannarino & Cohen, this issue, for a review of traumatic loss in children and adolescents; Stoplebein & Greening, 2000; Taylor, Weems, Coasta, & Carrión, 2009). Posttraumatic symptoms can interfere with the bereavement process (Cohen, Mannarino, Greenberg, Padlo, & Shipley, 2002), which in turn complicates adaptive coping with disaster exposure, increasing the likelihood of PTSD and depression.

Studies have demonstrated that childhood loss of a family member following a natural disaster is associated with symptoms of PTSD (Bhushan & Kumar, 2007; Hsu, Chong, Yang, & Yen, 2002; Pfefferbaum et al., 1999; Zhang, Wang, & Sun, 2000). For example, Hsu et al. conducted a study of 323 Taiwanese adolescents six weeks after a devastating earthquake that resulted in over 2,000 deaths and severe infrastructure damage. They found that adolescents who had experienced the death of a close family member were 5.6 times more likely to meet criteria for PTSD than adolescents who had not experienced a family death. Results from another study suggest that the effects of this risk factor appear to be long lasting. Zhang and colleagues found that adolescents who experienced the death of a family member at the time of the 1976 Tangshan earthquake in China were more likely to be diagnosed with PTSD as an adult than similarly exposed adolescents who had not experienced a death.

Death of a family member due to disaster has also been associated with depression among youth (Eksi et al., 2007; Goenjian et al., 2009; Wickrama & Kaspar, 2007). Eksi and colleagues conducted a study of 160 children (aged 9 to 16 years) between one and five months after a deadly 7.4 magnitude earthquake in Turkey. All children were recruited from schools located in the epicenter of the earthquake. Their results indicated that children who had lost a family member in the quake were almost 11 times more likely to develop depression than children who did not lose a family member. Loss of a family member also conferred increased risk for PTSD with comorbid depression: Children who had lost a family member were almost 7 times more likely to meet that diagnosis than those who had not.

As noted previously, loss of a family member appears to confer long-term risk for emotional distress. Six and a half years after the 1988 Spitak earthquake, Goenjian et al. (2009) compared depression and posttraumatic stress symptoms among children who had both parents die, one parent die (mother or father), or no parent die after the earthquake. Children who lost both parents had the highest depression scores, which were in the clinical range, followed by those who lost a father. The researchers suggest that, although the orphaned children were the most vulnerable to depression, those who lost a father may also have been at risk due to loss of financial support. They also hypothesized that cultural norms may have led to less social and familial support extended to the family of paternally bereaved children as opposed to maternally bereaved. In general, more research is needed with children exposed to natural disasters to identify common and unique pathways to traumatic grief, PTSD, depression, and physiological changes in a variety of postdisaster environments.

Postdisaster Environment

As noted previously, the immediate loss and disruption wreaked by natural disasters is often considered part of the initial traumatic exposure and is strongly associated with mental health outcomes. However, severe natural disasters engender prolonged disruption over a period of months and years due to the significant damage to personal property and community infrastructure as well as population displacement. It is an accepted fact that recovery after severe natural disasters is a long-term process (U.S. Government Accountability Office [GAO], 2008). In developed countries, restoration of electricity, water, and other basic services tends to occur within the first days and weeks; however, restoration of other services, infrastructure repair, and the return of residents and businesses are more likely to take months to years (Hayashi, 2007; U.S. GAO, 2008).

Hurricane Katrina serves as an example of the challenges of living in a postdisaster environment. On August 29, 2005, the landfall of Hurricane Katrina, combined with the catastrophic failure of the federal levee system, inundated 80% of the city of New Orleans with floodwater, damaged 71.5% of the housing stock (U.S. Department of Housing and Urban Development, 2006), and resulted in almost 400 deaths in Orleans Parish alone (Bourque, Siegel, Kano, & Wood, 2006). Because of the degree of devastation, residents were under a mandatory evacuation for five weeks (Greater New Orleans Community Data Center [GNOCDC], 2007). Approximately six months after the disaster, only one third of the hospitals were functioning at 20% of their former capacity and just 14% of public schools had reopened. The number of mental health care professionals and agencies in the New Orleans area was significantly diminished to almost one sixth of the pre-Katrina resources (Abramson & Garfield, 2006). More than one year after the storm, one third of the medical facilities, two thirds of the city's child care centers, and more than half of the public schools remained closed (GNOCDC, 2007). Approximately half of all bus and streetcar routes were running, but only 17% of buses were in use (Streips & Simpson, 2007). Finally, less than 10% of homeowners had received federal rebuilding funds and up to one third of former residents were unable or unwilling to return home (GNOCDC, 2007). Using other postdisaster recovery efforts as a guide, postdisaster disruptions are likely to continue for New Orleans residents in the foreseeable future (U.S. GAO, 2008). In fact, in a population-based survey of over 1,000 adults living in New Orleans almost three years post-Katrina, 40% reported that their personal lives continued to be at least somewhat disrupted by Hurricane Katrina and the recovery process and 58% reported ongoing problems with access to health care (The Henry J. Kaiser Family Foundation, 2008).

Research has found that postdisaster factors in the months and years after a disaster become more important than initial exposure factors (e.g., life threat, acute loss, and disruption) in predicting long-term psychological outcomes (McFarlane, 1987; Norris, Perilla, Riad, Kaniasty, & Lavizzo, 1999). These stressors, which stem directly from the disaster, are often referred to as secondary stressors and are potent risk factors for mental health because they can serve as reminders of the trauma, further tax the coping capacity of individuals, and create distress in their own right (La Greca et al., 1996). In addition, secondary stressors impede a return to normal conditions and routines in affected communities, which also poses a challenge to mental health (Sattler, Preston, Kaiser, Olivera, & Schlueter, 2002), particularly for youth. In summary, secondary stressors can serve to both complicate children's ability to cope with initial trauma exposure as well as present new challenges to mental health.

There is a growing body of research highlighting the importance of secondary stressors. For example, a study conducted following a severe flood in Thailand found that adults who reported negative changes in their lifestyles and relationships as a result of the disaster were 1.5 times more likely to have a mental health problem than those who did not report such changes (Assanangkornchai, Tangboonngam, & Edwards, 2004). A similar study conducted 20 months after a major flood in Poland found that the immediate trauma experience predicted the initial development of symptoms of PTSD and depression in youth. However, secondary stressors such as parental unemployment and continued damage to the home influenced the severity and maintenance of these problems nearly a year after the disaster (Bokszczanin, 2003). Finally, a study of 271 adolescents conducted 18 months after Hurricane Katrina found that the vast majority of adolescents (91%) experienced secondary stressors (e.g., home still damaged, difficulty seeing friends), with more than 40% experiencing more than three (Overstreet, Salloum, & Badour, 2010). Exposure to secondary stressors was a significant predictor of PTSD symptoms; 17% of the least exposed youth reported clinically significant PTSD symptoms compared to 41% of the most exposed.

The majority of the research conducted on the occurrence and consequences of secondary stressors is cross-sectional. More longitudinal research is needed to gain a clearer sense of how the relationships among initial exposure, secondary stressors, and mental health functioning change over time. Equally important to this line of research is a focus on the coping strategies individuals employ to deal with secondary stressors and resulting distress, as these strategies have important implications for mental health symptoms (Assanangkornchai et al., 2004). For example, we have found that the relationship between secondary stressors and re-experiencing symptoms of PTSD was stronger for adolescents who reported problem substance use (Overstreet et al., 2010).

Disruptions in Social Networks

Severe natural disasters result in the disruption of social networks. In any given year, as many as 50 million people are estimated to be displaced due to natural disasters, resulting in the often prolonged separation of children and their families from peer groups, extended family networks, neighborhoods, schools, and day care centers (Weems & Overstreet, 2008). For example, Hurricane Katrina resulted in the displacement of more than 2.5 million people, which was more than the 1.5 million people displaced in all of the countries affected by the 2004 Indian Ocean tsunami (Larrance, Anastario, & Lawry, 2007). Nine months after Hurricane Katrina, 85,000 people remained housed in temporary Federal Emergency Management Agency (FEMA) trailer communities (Larrance et al., 2007) and thousands of others chose (or were forced by various factors) to relocate to other parts of the country. Residents who were able to return to New Orleans often faced sparsely populated and still heavily damaged neighborhoods. In a city where an unusually high number of multigenerational family members resided prior to the Katrina disaster, family members faced prolonged separation, children were enrolled in schools outside of their neighborhood, and many high school students returned to the city without their parents (Weems & Overstreet, 2008). For example, in a study of secondary stressors 18 months post-Hurricane Katrina, 36% of adolescents reported that family members had not returned to the city and 82% reported separation from friends (Overstreet et al., 2010).

Displacement, relocation, and resettlement present physical challenges to social support, but even when usual sources of social support are accessible, they may not always be available to provide support in postdisaster environments. Adults in postdisaster environments are dealing with their own exposure, disruptions, and emotional distress (Kessler et al., 2008) and may not always be able to provide the social support needed by children. The loss of social support may contribute to “resource spirals,” where the continued and additive loss of resources results in significant distress (Hobfoll, 1989; Sattler et al., 2002).

The ability to overcome these challenges to social support in a postdisaster environment appears to be critically important in reducing the effect of the disaster on a child's mental health for a number of reasons. First, the support a child receives from significant others is important for emotional validation (Bokszczanin, 2003); this is imperative for youth because it demonstrates that they are loved, valued, and cared for and that they are not the only person experiencing negative feelings (Dubow & Ullman, 1989). Second, when children engage in conversations about the trauma in the context of supportive relationships, they may become less reactive to their extreme negative emotions. This may allow youth to become less physiologically and psychologically aroused to thoughts and memories about the trauma (Prinstein, La Greca, Vernberg, & Silverman, 1996). Consequently, individuals may be less likely to experience symptoms of hyperarousal. Third, support from significant others can involve suggestions for adaptive coping (Dubow & Ullman, 1989; Prinstein et al., 1996). Extreme stress often depletes an individual's independent coping abilities and resources; therefore, having others model healthy coping strategies is vital because it provides support during the struggle with ongoing loss and disruption (Green, 1991; Prinstein et al., 1996).

A handful of studies provide empirical support for the importance of social support for positive mental health in children living in postdisaster environments (Bokszczanin, 2003; La Greca et al., 1996; Pina et al., 2008; Vernberg, La Greca, Silverman, & Prinstein, 1996). For example, in a study of 46 school-aged youth six months post-Hurricane Katrina, Pina et al., (2008) found that perceived helpfulness from extrafamilial sources of support (e.g., teachers, friends, church members) was related to decreased posttraumatic stress reactions in youth (i.e., PTSD, anxiety, and depression). Similarly, in a study 20 months after the 1997 flood in Poland, Bokszczanin (2003) found that perceived social support from parents, teachers, and friends was associated with lower levels of PTSD symptoms in youth. Finally, in the only longitudinal study examining the influence of social support on children's postdisaster mental health, La Greca and colleagues (1996) found that, following Hurricane Andrew, the availability of social support predicted decreased PTSD symptoms over time, although the relative importance of different sources of support (e.g., parents, teachers, friends) varied over the course of the study.

Although social support has been a long-recognized protective factor for children dealing with stress and trauma (Compas & Epping, 1993; Ozer, Best, Lipsey, & Weiss, 2003), more research is needed. First, although most studies examine the direct effects of social support on mental health in postdisaster environments, no published studies have examined whether social support actually moderates the relationship between exposure to disasters and mental health. In other words, researchers have yet to provide empirical support for the idea that social support actually influences the relationship between disaster exposure and mental health, rather than (or in addition to) having a direct influence on mental health. Second, the lack of longitudinal studies limits our understanding of how the importance of social support might change over time. A meta-analysis of predictors of PTSD conducted by Ozer and colleagues (2003) indicated that across different types of trauma exposure, the lack of social support became a more important predictor of PTSD symptoms in adults as the length of time since the trauma increased. Finally, we need to expand the focus of research on social support to include an assessment of the nature of the social support being received (e.g., instrumental support, emotional support) so that we can begin to identify the pathways through which social support is having its positive effects on mental health.

Context of Recurrent Threat

Natural disasters such as earthquakes, floods, hurricanes, volcanoes, and fires occur in geographically vulnerable areas where the threat of recurrence is high. In fact, the most common natural disasters are recurrent rather than single events (Dilley et al., 2005). Dilley and colleagues (2005) note that 3.4 billion people are at risk for experiencing one natural hazard and 22.6% of those individuals are threatened by more than one. Although it is certainly possible that the integration of persistent and familiar threat with cultural norms and everyday life fosters disaster preparation and has a buffering effect on mental health (Alexander, 1993; Crittendon & Rodolfo, 2002), it is also possible that recurrent threat poses specific risks to mental health.

Recurrent seasons that bring the increased likelihood of specific natural hazards make it more likely that the increased threat of disasters occurs around the same time as the initial threat. This could complicate anniversary reactions, which tend to occur around the same date as the initial disaster and trigger upsetting memories of the disaster experience and fears that the disaster will recur (APA, 2009; Hamblen, Friedman, & Schnurr, n.d.). When those fears are accompanied by the actual possibility of recurrence, distress can increase even further, as was so painfully illustrated by Hurricane Gustav during the 2008 hurricane season.

Hurricane Gustav threatened New Orleans on the 3-year anniversary of Hurricane Katrina. Forecasters included New Orleans in the predicted path of the storm, which was expected to be very severe. In light of the initial forecasts, the mayor of New Orleans issued a mandatory evacuation for all residents, warning that Hurricane Gustav could be the “mother of all storms” (Williams, 2008, para. 1). Given the weeks of media coverage commemorating the anniversary of Katrina combined with the timing and the initial projections about Hurricane Gustav, the days preceding the storm were anxiety provoking for an already traumatized New Orleans population (Witness Justice, 2008). In fact, the results of a study of 122 school-aged children living in New Orleans indicated that the relationship between exposure to Hurricane Gustav and PTSD symptoms was amplified for children who also had high prior exposure to Hurricane Katrina (Salloum, Carter, Burch, Garfinkel, & Overstreet, 2011). The authors argue that it is unlikely that a “near miss” disaster such as Hurricane Gustav would have been viewed as traumatic without prior exposure to Hurricane Katrina. Other research with adults indicates that adults who were most affected by a prior disaster or who have PTSD or other mental health problems tend to experience the strongest anniversary reactions (Assanangkornchai, Tangboonngam, Sam-angsri, & Edwards, 2007; Hamblen et al., n.d.).

Studies examining the impact of recurrent threat on children's anniversary reactions to disaster exposure are limited. Future research should examine whether recurrent threat of a natural disaster serves to maintain mental health problems that arise following initial exposure or increase the intensity of anniversary reactions. In addition, research should determine whether disaster preparation programs and other disaster-related education efforts might ameliorate any of the increased risk associated with recurrent threat. Given the trend toward more frequent natural disasters, and with certain countries and regions more prone to recurrent disasters than others (Scheuren, le Polain, Below, Guha-Sapir, & Ponserre, 2008), it is critical that we increase our understanding of the impact of recurring disaster threat and exposure.

Implications for Treatment

Despite compelling findings that a considerable proportion of children develop and maintain symptoms of PTSD and other forms of emotional distress in the wake of natural disasters, major obstacles often exist in providing mental health services to those children. In postdisaster environments, it is often the case that, as the need for mental health services grows, access to services is diminished. A rapid needs assessment following Hurricane Katrina documented major gaps in mental health services for children and recommended school-based health services as one mechanism to address those gaps (Abramson & Garfield, 2006). As noted by Dean et al. (2008), schools are often the only institutions up and running in the immediate aftermath of severe natural disasters. Following Hurricanes Katrina and Rita in 2005, Dean et al. noted that “when many people in the communities lost everything they had, they went to the school as a community center and a source of help and support” (p. 56). As Weems and Overstreet (2008) point out, school-based mental health services may be an important way to provide access to intervention to ethnic minority youth, who tend to perceive less potential benefit from clinic-based treatment for anxiety related problems (see Ellis, Miller, Baldwin, & Abdi, this issue, for comments on the importance of schools for mental health services in refugee populations; Chavira, Stein, Bailey, & Stein, 2003).

A Public Health Framework for School-Based Mental Health Services

Many researchers argue that a public health approach is the best service delivery system in postdisaster environments (e.g., Stephan, Weist, Kataoka, Adelsheim, & Mills, 2007). The key components of a public health approach to school-based mental health involve offering a continuum of services using an ecologically based, culturally sensitive perspective; using empirically supported or evidence-based practices, coupled with systematic program evaluation; and employing a participatory approach to develop services and build collaborations. Each of these components is discussed in the context of creating, offering, and evaluating interventions in postdisaster environments (see Ellis, Miller, Baldwin, & Abdi, this issue, for a discussion of some of these factors with regard to refugee children and adolescents).

Continuum of Ecologically Based and Culturally Sensitive Services

School-based tiered approaches based on a public health model provide universal programs for all children, intervention for those at risk, and selected treatment for those with more severe problems (Brown, McQuaid, Farina, Ali & Winnick-Gelles, 2006; Layne et al., 2008). This tiered approach maximizes resources and matches children to the level of care needed. In addition, it also tends to involve teachers in the provision of classroom-based services and programs. As Layne et al. (2008) point out, teachers are part of youths' natural ecologies and their involvement in school-based mental health services can help to repair disrupted social networks.

One strategy for maximizing the percentage of youth who reliably respond to treatment may be to assign them to a specific level of treatment within a tiered system based on their needs. Information from screening tools could be used as an efficient way to identify the level of care and services needed (Chemtob, Nakashima, & Hamada, 2002; Layne et al., 2008; Roussos et al., 2005). In postdisaster environments, screening measures must adequately assess indicators of posttraumatic stress and depression, as well as indicators of traumatic grief and adaptive functioning (Layne et al., 2008). In addition to screenings, Chemtob et al. (2007) proposes the development of registries of bereaved children when catastrophic events occur so that services can be adequately provided. As part of disaster preparedness, a systematic plan to identify, track, assess, and provide intervention to children bereaved by the disaster will help ensure that these children receive the services that they need not only in the immediate aftermath but in the long term.

Another type of assessment that is critical prior to the implementation of treatment is a needs assessment. Trauma exposure and children's reactions to the exposure must be viewed through a contextually and culturally sensitive lens. Postdisaster environments create an urgency for the need to establish interventions, and although prior research provides valuable information to that end, attention needs to be given to the unique context where mental health services will be provided. Developing culturally appropriate services is dependent on knowledge of the respective cultures and on being culturally sensitive without engaging in stereotypic thinking. In the context of mental health service delivery, this means awareness of how psychological distress is conceived and manifested (e.g., behavioral vs. somatic symptoms), available individual and social resources for coping (coping skills, social support network, communal forms of coping), and how to best intervene to build on individual and cultural strengths (Nastasi & Varjas, 2008).

A resource instrumental to the integration of cultural sensitivity into evidence-based practice is the NCTSN's Trauma Informed Interventions project (de Arellano, Ko, Danielson, & Sprague, 2008). de Arellano and colleagues surveyed the developers of empirically supported treatments and promising practices for trauma affected youth in an effort to compile “Culture Specific Fact Sheets” for each program. The purpose of the project was to gather information regarding the cultural sensitivity of trauma-focused interventions and to summarize the level of research supporting each treatment with culturally diverse groups. Information was elicited regarding culture-specific engagement strategies, attention to culture-specific symptom expression, and availability of culture-specific adaptations to the treatment. The resulting guide provides information on 22 trauma-focused interventions and can be used by clinicians to assist in the selection of culturally sensitive treatments for youth exposed to disaster-related trauma.

Evidence-Based Practices and Program Evaluation

It is well recognized that evidence-based treatments should serve as the basis for postdisaster school-based mental health services; however, there are only a few published studies on the effectiveness of postdisaster psychological interventions for children and adolescents (Taylor & Chemtob, 2004). Common, core components of many child trauma-focused treatments include psycho-education, the creation of a coherent trauma narrative, anxiety management and emotional regulation skills, adaptive coping skills, parenting skills and behavior management, the promotion of safety and positive relationships, and relapse prevention (Amaya-Jackson & DeRosa, 2007). Most of the treatments that have been studied are grounded in cognitive behavioral therapy (CBT), one of the few treatment approaches for PTSD in children with solid empirical support (Feeny, Foa, Treadwill, & March, 2004). In addition to trauma-focused cognitive behavioral strategies, postdisaster treatments tend to include a focus on loss and traumatic grief (Goenjian et al., 1997; Layne et al., 2008; Salloum & Ovestreet, 2008) to explicitly help children cope with the pervasive loss issues that co-occur with large-scale natural disasters (see Mannarino & Cohen, this issue, for further discussion of traumatic grief in children and adolescents).

School-based interventions that have incorporated these various treatment components have the potential to have a positive impact on both PTSD symptoms and depression. For example, Goenjian et al. (1997) found that youth (sixth and seventh graders) who participated in a six-week, structured, school-based grief and trauma intervention 1.5 years post earthquake reported significantly lower levels of posttraumatic stress and depression compared to youth who did not participate. Three and a half years postintervention, youth participants were three times more likely to report a decrease in posttraumatic stress than those youth not treated. Further, those youths who were not treated reported worsening of depressive symptoms over time (Goenjian et al., 2009). The authors suggest that by focusing treatment on concerns that directly relate to depression, such as lost opportunities and disruptions in social relationships, the escalation of depressive symptoms in adolescents who received treatment may have been curbed. Their findings suggest that the dual treatment focus on trauma symptoms and grief reactions may have a positive impact on both PTSD symptoms and depression.

In a randomized study with children postdisaster, Chemtob and colleagues (2002) found that children (N = 248, ages 6 to 12) who participated in a four-session school-based group or individual cognitive-behavioral psychosocial intervention two years after a major hurricane reported significantly more decreases in posttraumatic symptoms than a comparison group at posttest and one year later. Building on the work on Chemtob and associates, Salloum and Overstreet (2008) used a randomized design to compare the effectiveness of individual and group forms of a grief and trauma intervention in a sample of 56 children (ages 7 to 12 years) returning to school four months after Hurricane Katrina. The intervention, which was based on cognitive-behavioral and narrative techniques, included a grief component in addition to the trauma treatment and consisted of a total of 11 school-based sessions and one home or school-based parent meeting. Across both treatment modalities, children reported significant decreases in posttraumatic stress, depression, traumatic grief, and global distress (Salloum & Overstreet, 2008). Specifically, of the 24 participants who fell above the clinical cutoff for PTSD symptoms prior to treatment, 75% no longer exceeded the clinical cutoff at post treatment. In addition, moderate to large treatment effects were observed for the other dependent variables, including depression and traumatic grief.

Participatory and Collaborative Approaches to Service Delivery

Developing comprehensive, culturally sensitive, ecologically valid services in a postdisaster environment is likely to challenge the competencies and capacities of any individual professional or agency, thus necessitating partnerships among professionals (e.g., psychologists, educators, medical personnel, public health specialists), institutions (e.g., schools, community mental health agencies), and the individuals they serve. Participatory approaches to service delivery are characterized by building equal partnerships, fostering ownership, and building capacity within the existing systems (e.g., providing professional development for local school and community personnel). Participatory models can help build cohesive social networks, which promote resilience and healing in postdisaster environments (Norris et al., 2002).

A critical partner in school-based mental health services is parents (Brown et al., 2008). Involving parents in the treatment process and teaching them about what their child is learning in therapy (such as relaxation techniques) helps parents to reinforce those skills at home and provides education and normalization of postdisaster reactions. It also provides the opportunity for supportive counseling and additional referrals, if indicated (Salloum, Garside, Irwin, Anderson, & Francois, 2009). In addition, because exposure to distressing disaster-related images via media has been associated with increased distress among youth (Pfefferbaum et al., 1999), meeting with parents allows for a discussion about limiting media exposure and about having developmentally appropriate conversations with children about disaster-related images they have witnessed. Despite its importance, gaining parental involvement may be challenging due to the varied practical demands parents are faced with in postdisaster environments. Therefore, to achieve parental involvement, treatment approaches should be flexible and include a range of options for children and their available support networks.

Summary

Exposure to severe natural disasters changes the behavior of youth and creates “developmental hazards” (Dean et al., 2008, p. 56) that, in turn, can lead to the emergence of psychopathology and declines in adaptive functioning. In addition to the embodiment of traumatic exposure, severe natural disasters present several unique challenges to youth. School-based mental health services created within a public health framework hold promise for the treatment of problems that emerge in youth following exposure to severe natural disasters. However, more research is needed to strengthen the evidence base for specific treatments to be offered within that framework.

Evaluating interventions in a postdisaster context can be challenging due to timely institutional review board approvals, design challenges related to control groups or adequate comparison groups, difficulty carrying out follow-up assessments with an often transient population, overcoming beliefs about research adding an additional burden to an already stressed population (Steinberg, Brymer, Steinberg, & Pfefferbaum, 2006), having readily available trained clinical staff, balancing between postdisaster service needs and research, and allowing for flexibility of the treatment yet adhering to fidelity (CATS Consortium, 2007). We must work on plans to address these issues prior to a disaster, especially in communities that are vulnerable to disasters, so that research can be an integral part of the service delivery response.

As the efficacy and effectiveness of postdisaster treatments becomes more established, it will be critical for public policy to align with those findings so that resources are appropriately allocated to provide the best mental health services for children and adolescents. In the aftermath of a disaster, the federal dollars and volunteer assistance applied to mental health services are often designed to meet the immediate needs of a population in crisis (Weems & Overstreet, 2009), which is not necessarily conducive to a public health approach to mental health service delivery (Drury, Scheeringa, & Zeanah, 2008; Vernberg, 2002). There is little evidence that a short-term approach to mental health treatment works and it is clear there are long-term mental health needs in postdisaster environments (Kessler et al., 2008). More attention needs to be focused on creating funding policies that will address the long-term needs of children and families affected by severe natural disasters as well as enhance the capacity of schools and community agencies to provide needed services. Long-term disaster planning for mental health treatment should occur on many different levels including the individual provider level, the municipal and local level, and the state and federal policy level to ensure that mental health services are available long after disasters.