Parental cancer diagnosis and child mortality--a population-based cohort study in Sweden.

OBJECTIVE: Cancer diagnosis is known to induce severe psychological stress for the diagnosed patients; however, how it affects the next-of-kin is less well documented. This study aimed to assess the impact of parental cancer on the risk of childhood death. METHODS: A population-based cohort study was conducted using the Swedish national registries, including 2,871,242 children followed during the period of 1991-2009. Parental cancer diagnosis was defined as a time-varying exposure. We used Cox proportional hazards regression to calculate the hazard ratio (HR) and its corresponding 95% confidence interval (CI) as an estimate of the association between parental cancer and childhood mortality. We adjusted for attained age, sex, gestational age, mode of delivery and birth weight of the child, maternal age at child's birth, as well as educational level and socio-economic classification of the parents in the analyses. RESULTS: Among 113,555 children with parental cancer, 127 deaths occurred during 561,198 person-years of follow-up. A parental cancer diagnosis was associated with an increased rate of death among children at the age of 1-18 (HR for all-cause death: 1.39; 95% CI: 1.16-1.66). For young children (aged 1-12), an increased rate was only noted for death due to cancer (HR: 2.06; 95% CI: 1.13-3.75) after parental cancer diagnosis. Among adolescents (aged 13-18), an increased rate was noted for all-cause death (HR: 1.52; 95% CI: 1.25-1.86), and for both non-cancer-related (HR: 1.43; 95% CI: 1.14-1.79) and cancer-related (HR: 2.07; 95% CI: 1.33-3.24) death in the exposed children. CONCLUSION: Children have an increased rate of death if they have a parent diagnosed with cancer as compared to children without such experience; this association appears to be slightly stronger among adolescents.