Predisaster predictors of posttraumatic stress symptom trajectories: An analysis of low-income women in the aftermath of Hurricane Katrina.

Sarah R Lowe,Ethan J Raker,Mary C Waters,Jean E Rhodes,Peter G Van Der Velden

doi:10.1371/journal.pone.0240038

Abstract

Exposure to disasters is associated with a range of posttraumatic stress symptom (PTSS) trajectories. However, few studies have tracked PTSS for more than a decade postdisaster, and none to our knowledge has explored the role of predisaster resources and vulnerabilities in shaping the likelihood of trajectory membership. The current study included participants from the Resilience in Survivors of Katrina Study (N = 885). Participants were originally part of a community college intervention study and were assessed prior to the hurricane (6-21 months predisaster), and approximately 1 year, 4 years, and 12 years postdisaster. Latent class growth analysis identified PTSS trajectories. Perceived social support, probable mental illness, and physical health conditions or problems-all assessed predisaster-were examined as predictors of trajectory membership at the univariate level and in multivariable models without and with adjustment for disaster exposure. Three PTSS trajectories were detected: Moderate-Decreasing (69.3%), High-Decreasing (23.1%), and High-Stable (7.6%). In the multivariable predictive model without adjustment for disaster exposure, probable predisaster mental illness was significantly associated with greater odds of the High-Decreasing and High-Stable trajectories, and physical health conditions or problems with greater odds of the High-Decreasing trajectory, relative to the Moderate-Decreasing trajectory. However, when disaster exposure was adjusted, only the association between predisaster mental illness and the odds of the High-Stable trajectory remained statistically significant. Lower predisaster perceived social support was significantly associated with membership in the High-Decreasing trajectory, relative to the Moderate-Decreasing, at the univariate level, but not in either multivariable model. Whereas predisaster mental illness confers risk for chronic postdisaster PTSS, other impacts of predisaster resources and vulnerabilities on elevated PTSS trajectories do not go beyond those of disaster exposure. The results support disaster preparedness efforts targeting those with mental and physical health conditions, and postdisaster mental health services addressing preexisting vulnerabilities in addition to disaster-related PTSS.

Highlights

A large body of research has shown that natural disasters, including hurricanes, floods, and earthquakes, are associated with a range of adverse mental health outcomes [1,2,3]
A focus on posttraumatic stress disorder (PTSD) as a dichotomous outcome neglects the heterogeneity in posttraumatic stress symptoms (PTSS) among those who do and do not meet criteria for the disorder, such that those with subclinical symptoms might benefit from mental health services, and those with extreme symptoms might have different needs from those who just surpass the diagnostic threshold
We note here that no differences in postdisaster mental health outcomes were observed between participants who were in the intervention and control conditions; as such, intervention status has been excluded from primarily analyses of the Resilience in Survivors of Katrina (RISK) data [18, 21, 28], and was excluded from the current study as well

Summary

Introduction

A large body of research has shown that natural disasters, including hurricanes, floods, and earthquakes, are associated with a range of adverse mental health outcomes [1,2,3]. At least two prospective studies to date have found predisaster support to be associated with mental health symptom trajectory membership [13, 14] Both studies included both pre and postdisaster symptoms in estimating trajectories and, as such, neither study assessed disaster-related PTSS. Research on predisaster predictors of disaster exposure is lacking, prior analyses of the RISK study have shown that participants with lower predisaster support and higher predisaster psychological distress tended to face exposure to a greater number of disaster-related trauma [24, 25] These associations, in turn, could account for their enhanced risk for PTSS. RISK analyses have shown lower perceived social support to be indirectly associated with higher postdisaster PTSS via exposure to a greater number of disaster-related trauma (12) It is unclear how such relationships could influence the likelihood of various postdisaster symptom trajectories. Such findings would suggest that the influence of predisaster factors on PTSS trajectory membership might be indirect via disaster exposure

Participants and procedures

Results

Discussion

26. EM-DAT