Risk of suicide among individuals with a history of childhood cancer

Abstract

BACKGROUND: Previous studies have described suicidal ideation among survivors of childhood cancer, but small numbers of events limit the understanding of suicide risk. The objectives of this study were to assess whether childhood cancer survivors are at increased risk of suicide in comparison with the general population and to determine risk factors associated with risk in a population-based cohort. METHODS: First primary malignancies among individuals aged 0 to 19 years from 1975 to 2016 were identified from Surveillance, Epidemiology, and End Results (SEER) databases. Standardized mortality ratios (SMRs) of suicide were obtained via SEER*Stat software from SEER 9. Fine and Gray proportional hazards models were used to identify suicide-associated factors among childhood cancer patients included in SEER 18. RESULTS: In all, 96,948 childhood cancer cases and 89 suicides were identified. Across all attained ages, the suicide risk for individuals with a childhood cancer history (11.64 per 100,000 person-years) was similar to the risk for those without a cancer history (SMR, 1.14; 95% confidence interval [CI], 0.91-1.43). However, for survivors alive beyond the age of 28 years (the median age of death by suicide), the suicide risk was significantly elevated (suicides per 100,000 person-years, 22.43; SMR, 1.40; 95% CI, 1.02-1.87). Females (hazard ratio, 0.29; 95% CI, 0.18-0.59; P < .01) had lower risks than males. CONCLUSIONS: These results suggest that long-term childhood cancer survivors may be at increased suicide risk. Male sex is an independent risk factor for suicide. However, the absolute risk of suicide in older survivors is still low at ~1 per 5000 person-years. Future efforts should identify survivorship strategies to mitigate suicide risk.

Lifetime Burden of Psychological Symptoms, Disorders, and Suicide Due to Cancer in Childhood, Adolescent, and Young Adult Years

“…Results of risk of bias assessment using the Joanna Briggs Institute critical appraisal tools (eTable 17 in Supplement 1) showed that although studies overall were not at significant risk of bias, a proportion used normative population data rather than matched controls . Publication bias, outlier assessment, and leave-one-out analysis was also performed (eFigures 6-25 in Supplement 1).…”

“…From 7319 results (Figure 1), we included a total of 52 studies reporting psychiatric disorders and psychological symptoms of depression, anxiety, psychotic disorders (including schizophrenia), and suicide mortality in CYACs compared with controls without cancer (eTable 2 in Supplement 1). Most studies recruited healthy, age-matched comparators as a control group, whereas 13 studies recruited siblings and 3 studies recruited parents or caregivers of patients with cancer.…”

“…Three studies were included in a meta-analysis of suicide mortality per 100 000 person-years for CYACs. The overall risk was found to be increased at an RR of 1.63, not reaching statistical significance (95% CI, 0.78-3.40) (eFigure 5 in Supplement 1).…”

“…Eleven studies investigated the association between level of educational attainment and risk of mental health conditions, including depression, anxiety, schizophrenia, and/or suicide (eTable 14 in Supplement 1). Eight of these studies found a significant association between lower educational levels and increased risk of mental health conditions or suicide, whereas 3 studies found an insignificant association. Specifically, lower educational attainment was associated with an increased risk of depression, anxiety, and combined risk of either depression, somatization, or anxiety (ie, using the BSI-18) .…”

“…Fourteen studies investigated associations between income level and the risk of mental health conditions (eTable 15 in Supplement 1). Nine of these studies found a significant association between patients with lower income compared with patients with higher income and risk of depression.…”