Methods: Students at four universities (N = 416) completed written questionnaires approximately three months following Hurricanes KR. Measures of mental health included the 17-item Modified PTSD Symptom Scale (MPSS-SR), adapted to measure PTSD-like symptoms during the month following Hurricanes KR, and the 20-item Center for Epidemiological Studies–Depression Scale (CES-D) (Cronbach's = .92 and .90, respectively). Substance abuse and active coping were measured with subscales of the Brief COPE (Cronbach's = .84 and .64, respectively). First, bivariate analyses were conducted to examine zero-order correlations between major variables of interest. Next, multiple regression analyses were employed to examine the predictive ability of significant correlates on depression, PTSD-like symptoms, substance abuse, and active coping. Finally, significant predictors were entered into the regression equation to yield a parsimonious model.
Results: Mean scores on the CES-D and MPSS-SR were moderate, at 16.33 (SD = 10.7) and 7.75 (SD = 9.06), respectively. Fewer than one fifth (16.9%) reported substance abuse (M = .40, SD = 1.06) and over three fourths (80.7%) used active coping (M = 2.73, SD = 1.96). The final regression equation with significant correlates of depression included African American ethnicity, previous personal traumas, optimism (measured with Life Orientation Test), negative reactions, hurricane stressors, and negative peritraumatic emotions (adapted from Types of Peritraumatic Emotional Responses to Terrorist Attacks). The model was significant at F(5, 273) = 35.1, p < .001, R Square = .39, and explained approximately 40% of the variance in depression. The final regression equation with significant correlates of PTSD-like symptoms included African American ethnicity, negative peritraumatic emotions and negative reactions. The model was significant at F(3, 277) = 59.6, p < .001, R Square = .39, and explained approximately 40% of the variance in PTSD-like symptoms. The final regression equation with significant correlates of substance abuse included Caucasian ethnicity, negative reactions, optimism, volunteer activities, and previous personal traumas. The model was significant at F(3, 345) = 8.81, p < .001, R Square = .11, and explained 11% of the variance in substance abuse. The final regression equation with significant correlates of active coping included negative reactions, altruism (measured with Altruism Scale), and sociopolitical reactions (adapted from Sociopolitical Reactions to Attacks of International Terrorism). The model was significant at F(3, 382) = 30.05, p < .001, R Square = .19 and explained approximately 20% of the variance in active coping.
Results showed race-related differences regarding vulnerability to mental health problems and substance abuse. Disasters pose a major challenge to the social work profession and to students reporting previous traumatic experiences, in particular. Findings support research demonstrating links between altruism and well being.