Factors influencing the course of posttraumatic stress following a natural disaster: Children's reactions to Hurricane Katrina

Abstract

This investigation examined psychosocial and behavioral factors involved in the course of post-traumatic stress disorder (PTSD) symptoms in youth affected by Hurricane Katrina. Participants (N = 152; 54% female; 61% Caucasian; mean age = 11.5 years) self-reported on hurricane exposure, PTSD symptoms, fear reactivity, regulatory abilities, social experiences, and coping 1.5 months following the storm, and on their PTSD symptoms again 8 months following the storm. Fear reactivity predicted more severe initial PTSD symptoms. Peer victimization added to the prediction of subsequent PTSD symptoms. Whereas regulatory abilities protected against PTSD symptoms both initially and across time, a negative coping style enhanced vulnerability to PTSD symptoms. Thus, mental health service providers should work to minimize peer victimization experiences, improve regulatory abilities, and promote alternatives to negative coping in youth following disasters.

Introduction

Hurricane Katrina resulted in one of the worst natural disasters in U.S. history, and exposure to such disasters is associated with adjustment difficulties in youth (see Silverman & La Greca, 2002 for a review). However, a variety of personal and contextual factors also contribute to negative outcomes in youth following disasters (e.g., Earls et al., 1988, Vernberg et al., 1996). Yet little research has examined if these personal characteristics and contextual factors interact with disaster exposure to impact adjustment following a disaster. Given the negative effects of disaster exposure among youth, a better understanding of the processes whereby factors impact risk and resilience in youth following disasters can aid in intervention efforts. In this context, the present investigation examined the additive and interactive influences of theoretically relevant emotional, social, and behavioral factors on the course of Posttraumatic Stress Disorder (PTSD) symptoms in youth following Hurricane Katrina.

Children's exposure to disasters has been linked to negative outcomes, which include anxiety and depression (Groome and Soureti, 2004, McDermott and Palmer, 2002). Most often studied, however, is the link between disaster exposure and PTSD symptoms (e.g., Garrison et al., 1995, Shaw et al., 1995), which include negative re-experiencing, avoidance, emotional numbing, and hyperarousal (American Psychiatric Association, 2000). The course of PTSD symptoms can vary widely and a better understanding of the factors that influence changes in the severity of PTSD symptoms over time is needed (Schnurr, Lunney, & Sengupta, 2004). This is particularly true for youth exposed to disasters, as youth are less likely to have the maturity necessary to effectively respond to disasters, and they must often rely on adults, who themselves are affected by the disaster, for support (Silverman & La Greca, 2002).

When examining the impact of hurricane exposure on youth, children's characteristics (e.g., sex, age, and ethnicity), intellectual and emotional functioning, factors in the post-disaster environment, and coping efforts are all theorized to influence the process whereby disasters lead to the emergence and course of PTSD symptoms (La Greca et al., 1996, La Greca et al., 1998, Vernberg et al., 1996). Research with youth generally supports this assertion. Although sex differences tend to be small in size, females report more disaster-related PTSD symptoms than males (Groome and Soureti, 2004, Yule, 1992). Preexisting emotional and behavioral difficulties increase the risk for PTSD symptoms as early as 10 days post-disaster and the persistence of these symptoms for as long as a year (e.g., Lengua et al., 2005, Nolen-Hoeksema and Morrow, 1991). Longer community recovery (Galante & Foa, 1986) and parental responses to the disaster (Earls et al., 1988) also are associated with elevated PTSD symptoms in youth up to 18 months following disasters.

When examining children's reactions to hurricanes more specifically, being female increases the risk for short-term and long-term PTSD symptoms (e.g., Garrison et al., 1995, Russoniello et al., 2002). The effects, however, are small and have not been found in all studies. Furthermore, the impact of ethnicity on PTSD symptoms following hurricanes is unclear, with some studies showing minorities are at increased risk (La Greca et al., 1996, La Greca et al., 1998), others finding Caucasians at greater risk (Garrison, Weinrich, Hardin, Weinrich, & Wang, 1993); and still others reporting no link to PTSD symptoms (e.g., Russoniello et al., 2002, Vernberg et al., 1996).

Predisaster negative emotionality (La Greca et al., 1998, Weems et al., 2007), postdisaster anxiety (Lonigan, Shannon, Taylor, Finch, & Sallee, 1994), predisaster attention and academic difficulties (La Greca et al., 1998), a lack of social support and maladaptive coping efforts (Vernberg et al., 1996) all increase the risk of more severe PTSD symptoms in youth within 3 months of hurricane exposure. Regarding the course of PTSD symptoms, predisaster anxiety predicts more severe PTSD symptoms for up to 10 months following exposure (La Greca et al., 1998). Additionally, post-hurricane factors such as stressful events, low social support, and coping efforts increase the risk for a prolonged course of PTSD symptoms in youth (Garrison et al., 1995, La Greca et al., 1996, La Greca et al., 1998).

Among youth, developmental differences in reactions to disasters are unclear (La Greca & Prinstein, 2002), and examining the impact of disasters on youth transitioning into adolescence is needed. Although older youth might be better able to understand the disaster and its effects than their younger peers, during late childhood and early adolescence youth are still unlikely to be mature enough to deal with their disaster related experiences without support, and as youth approach adolescence, peers become increasingly important (Buhrmester, 1996, Rubin et al., 2006), presumably playing more significant roles in adolescents reactions to disasters. Additionally, adolescence is a particularly important period of brain development involving regulatory systems (Dahl, 2004), which potentially increases their risk as regulatory difficulties are central characteristics of post-traumatic stress reactions (APA, 2000).

Whereas positive treatment from peers (e.g., social support) decreases the risk of PTSD symptoms in youth (La Greca et al., 1996, Vernberg et al., 1996), negative peer treatment (i.e., peer victimization) likely increases the risk for PTSD symptoms. As previously discussed, post-disaster stressful life events are associated with the emergence and course of PTSD symptoms in youth (Garrison et al., 1995, La Greca et al., 1996), and peer victimization is one of the more common and distressing social stressors that youth face in late childhood and early adolescence (Crick and Bigbee, 1998, Nansel et al., 2001, Sharp, 1995). Thus, frequent peer victimization likely increases the risk of hurricane-related PTSD symptoms and reduces the likelihood that these symptoms will abate over time, particularly as youth approach adolescence, but previous research has not examined the role of negative peer treatment on adjustment following disasters such as Hurricane Katrina.

Models applied to studying adjustment following disaster exposure and stressful environments also suggest that some factors play more complex roles in the link between hurricane exposure and PTSD symptoms (La Greca et al., 1996, La Greca et al., 1998, Luthar et al., 2000, Vernberg et al., 1996). The social support buffering hypothesis, for example, argues that social support interacts with stressors to weaken the links between stressors and negative outcomes (Cohen & Willis, 1985). Peer support might be particularly important following hurricanes, as peers can decrease isolation and assist in coping efforts (Prinstein et al., 1996, Vernberg et al., 1996), and peer support was found to predict both the initial severity and course of PTSD symptoms following hurricanes (La Greca et al., 1996, Vernberg et al., 1996).

Coping efforts might also moderate the link between hurricane exposure and PTSD symptoms, as youth who cope less effectively with stress will be less prepared to cope effectively with their hurricane experiences, strengthening the link between hurricane exposure and PTSD symptoms. Typically, negative forms of coping (e.g., ruminating, venting frustration, and avoiding the stressor) increase maladjustment in the face of stress (Dempsey, 2002, Rosario et al., 2003). Thus, negative coping likely also strengthens the link between hurricane exposure and PTSD symptoms in youth.

Regarding children's personal characteristics, a tendency to experience fear (i.e., fear reactivity) in potentially dangerous or emotionally arousing situations likely increases the risk for PTSD symptoms following disasters. Fear during disasters is a strong predictor of PTSD symptoms in children (Evans & Oheler-Stinnett, 2006) and adults (Başoğlu, Kiliç, Şalcioğlu, & Livanou, 2004), and youth with high fear reactivity likely experience higher levels of fear than their peers during hurricane experiences, increasing the association between hurricane exposure and PTSD symptoms in these youth. Alternatively, children's ability to regulate their emotions and behaviors likely attenuates the link between hurricane exposure and PTSD symptoms, as youth with better regulatory abilities are presumably better equipped to deal with disaster related stress. Supporting this point, better regulatory abilities weaken the link between adversity and maladjustment (e.g., Loukas and Roalson, 2006, Silk et al., 2006), and improvements in regulatory abilities relate to successful treatment of PTSD in adults (Cloitre et al., 2002, Cloitre et al., 2004).

The current study was conducted as part of a project originally planned to examine factors associated with peer victimization in youth. Following Hurricane Katrina, the project expanded to also examine the impact of the hurricane on students' adjustment. Participants were students attending a school located in an area of southeastern Louisiana under an evacuation order during Hurricane Katrina. The area serviced by the school was located approximately 35 miles from Hurricane Katrina's path, and research indicates that youth do not have to be at the center of a disaster to be affected negatively (e.g., Bulut et al., 2005, Shaw et al., 1995). The area experienced flooding, severe wind damage (e.g., destruction from large trees blown onto homes and cars), and a disruption of most services (e.g., electricity, phone, gas, groceries, etc.) for weeks (Crain, 2005, Louisiana Speaks: Long-Term Recovery Planning, n.d.). After the storm, unemployment claims in the area increased 193%; 25% of homes were seriously damaged; and 30% of residents became eligible for disaster aid (LSLTRP, 2007). Data were collected 1.5 months and 8 months after the storm.

Within this context, we sought to extend extant research by examining the direct and moderating associations of personal characteristics and treatment by peers on the course of PTSD symptoms after Hurricane Katrina in a sample transitioning from late childhood to early adolescence. Although developmental differences in reactions to disasters are currently unclear, early adolescents are a particularly interesting sample to study as they are old enough to understand the disaster but still not necessarily mature enough to deal with disaster experiences on their own. Additionally, peers begin to play an increasing role in children's lives as they approach adolescence (Buhrmester, 1996, Rubin et al., 2006), and presumably play a larger role in adolescents' adjustment following disasters. Guided by models of risk and resilience in the face of adversity and disasters, we examined the associations of previously unexamined emotional and social factors (i.e., fear reactivity, emotional regulatory abilities, and peer victimization) with PTSD symptoms (La Greca et al., 1996, La Greca et al., 1998, Luthar et al., 2000, Vernberg et al., 1996). We hypothesized that children's fear reactivity, regulatory abilities, peer victimization, peer support and negative coping style (characterized by externalizing, internalizing, and avoidant coping efforts) would independently add to the prediction of both initial and subsequent PTSD symptoms. We also hypothesized that greater fear reactivity and a negative coping style would interact with hurricane exposure, strengthening the link between exposure and subsequent PTSD symptoms. Alternatively, better regulatory abilities and peer support were expected to disrupt the link between hurricane exposure and PTSD symptoms.