Résumé
OBJECTIVE: A history of trauma is associated with poor mental and physical health, but the specific impact of posttraumatic stress disorder (PTSD) symptoms on physical health using objective indicators of health status has rarely been evaluated in elderly civilians. This study investigates the long-term consequences of a lifetime exposure to trauma on health in a French elderly general population. METHOD: Data from this retrospective study were derived from a longitudinal study (the Enquête de Santé Psychologique-Risques, Incidence et Traitement [ESPRIT]) of community-dwelling participants. Psychiatric health, medical history, and clinical examination (ICD-10 criteria) were assessed in 1,662 subjects (mean [SD] age = 72.5 [5.2] years). Lifetime traumatic exposure, PTSD, and psychiatric diagnoses were obtained using the Watson PTSD Inventory and the Mini-International Neuropsychiatric Interview. The outcome measures used were the Mini-International Neuropsychiatric Interview, Center for Epidemiologic Studies Depression Scale, Mini-Mental State Examination, and measures of physical health. RESULTS: We observed an increase in the number and severity of health-related outcomes between groups, with nontraumatized subjects having the lowest risk and those with trauma leading to recurrent reexperiencing of events (nonresilient subjects) having the highest risk. Traumatized persons who did not report reexperiencing symptoms (resilient subjects) showed better current mental health than traumatized subjects who did and nontraumatized subjects. Nonresilient subjects were more likely to have current depressive symptoms (P = .003), current major depression (P < .0001), current anxiety disorder (P = .032), and psychiatric comorbidity (P = .002) than nontraumatized subjects. Resilient subjects differed from nontraumatized subjects in having significantly less current suicidal ideation (P = .054) and psychiatric comorbidity (P = .035). Both groups of traumatized subjects showed a higher rate of cardio-ischemic diseases, notably current angina pectoris (multivariate, adjusted OR = 2.27; 95% CI, 1.31-3.91; and OR = 2.34; 95% CI, 1.22-4.49; for resilient and nonresilient groups, respectively). Traumatized persons, specifically those nonresilient, showed a higher waist-hip ratio, higher triglyceride levels, and a greater frequency of hypertension. CONCLUSIONS: Our findings suggest that trauma could be associated with cardio-ischemic diseases independently of PTSD symptoms expression. However, the presence of these symptoms appears associated with additional metabolic risk factors.