Prevalence and predictors of posttraumatic stress disorder in adult survivors of childhood cancer: A report from the Childhood Cancer Survivor Study

CRA10002 Background: This study examined prevalence and demographic- and disease-related correlates of posttraumatic stress disorder (PTSD) among 6,542 adult childhood cancer survivors and 368 siblings from the Childhood Cancer Survivor Study. Methods: Subjects were dichotomized based on full PTSD criteria, using the Foa PTSD self-report measure to assess posttraumatic stress symptoms of intrusion of unwanted memories, avoidance of event reminders, and increased startle response, and the Brief Symptom Inventory-18 and the SF-36 subscale, role limitation due to emotional health, to evaluate clinical distress or impaired function. A self-report questionnaire provided demographic information and medical abstraction provided cancer reatment data. Multivariable generalized linear models were used to compare prevalence of PTSD among cancer survivors to siblings and to examine relationships between PTSD and demographic and disease-related factors. Relative risks (RR) were calculated based on a Poisson distribution with robust error variances. Results: Five hundred eighty-nine (9%) childhood cancer survivors and 8 (2%) siblings met criteria for a diagnosis of PTSD (RR = 3.83, 95% CI 1.96–7.48, p < 0.0001). Among cancer survivors, there was significantly more PTSD reported by women from minority backgrounds (p < 0.05). Other demographic factors associated with PTSD were having less than a college education (p < 0.05), being unmarried (p < 0.001), having an annual income less than $20,000 (p < 0.05) and being unemployed (p = 0.001). Risk of PTSD was significantly higher for survivors diagnosed at ages 15 to 20 years (p < 0.05). PTSD was more common among survivors treated with intensive chemotherapy (p < 0.05) or radiation therapy (p < 0.001) and who relapsed or developed a second malignant neoplasm (p < 0.001). Neuroblastoma and Wilms tumor survivors had a significantly lower risk of PTSD than did leukemia survivors (p < 0.05). Conclusions: While the majority of childhood cancer survivors did not demonstrate PTSD, a clinically significant number did meet diagnostic criteria. Prospective assessment of survivors with high risk demographic-, diagnosis-, and treatment-related characteristics should be considered as part of long-term health screening. No significant financial relationships to disclose.