Suicide and violent deaths in survivors of cancer in childhood, adolescence and young adulthood-A national cohort study

Abstract

Suicide risk in adult cancer patients is found to be elevated, but limited information exists regarding risks of suicide and non-suicidal violent deaths when diagnosed with cancer in young age. We investigate suicide and violent deaths in a national cohort including individuals diagnosed with cancer before age 25. Through the linkage of different national registries (Cancer Registry of Norway, Norwegian Causes of Death Registry and the National Registry) a cohort of all live births in Norway during 1965-1985 was defined and followed up through 2008. Individuals diagnosed with cancer before age 25 and the cancer-free references were compared using an extended Cox proportional hazard regression model. The cohort comprised 1,218,013 individuals, including 5,440 diagnosed with cancer before age 25. We identified 24 suicides and 14 non-suicidal violent deaths in the cancer group. The hazard ratio (HR) of suicide in the cancer group was 2.5 (95% confidence interval (CI) 1.7-3.8), and was increased both when diagnosed with cancer in childhood (0-14 years of age); HR = 2.3 (95% CI: 1.2-4.6), and during adolescence/young adulthood (15-24 years); HR = 2.6 (95% CI: 1.5-4.2). Survivors of bone/soft tissue sarcomas, CNS tumors and testicular cancer were at particular risk. The risk of non-suicidal violent death was not increased in the cancer survivors (HR = 1.0; 95% CI: 0.6-1.7). Although based on small numbers and the absolute risk of suicide being low, these are novel findings with important implications for establishing adequate follow-up including suicide prevention strategies for young cancer survivors.

Suicides and deaths linked to risky health behavior in childhood cancer patients: A Nordic population‐based register study

“…27 However, the Norwegian study did not Cancer October 15, 2019 find any risk elevation for external nonsuicidal causes of death, whereas we found a significantly elevated risk for these causes in our data (RR, 1.25; 95% CI, 1.01-1.54; data not shown). These results are in concordance with those previously reported from Norway, in which a 2.5-fold risk of suicide was found in patients with cancer diagnosed before age 25 years when compared with cancer-free comparison subjects.…”

“…These results are in concordance with those previously reported from Norway, in which a 2.5-fold risk of suicide was found in patients with cancer diagnosed before age 25 years when compared with cancer-free comparison subjects. 27 However, the Norwegian study did not Cancer October 15, 2019 find any risk elevation for external nonsuicidal causes of death, whereas we found a significantly elevated risk for these causes in our data (RR, 1.25; 95% CI, 1.01-1.54; data not shown). Compared with the Norwegian study, the cohort in the current study consisted of younger patients and population comparisons and the follow-up time was longer.…”

“…Studies with information regarding hospitalizations for such conditions are needed to explore the subgroups of patients who would benefit the most from psychosocial support. 27,33,34 The results of the current study indicate there is an elevated risk of death resulting from risky health behavior-related causes among patients diagnosed with CNS tumors. That is one reason why we wanted to examine suicides and risky health behaviorrelated deaths both together and separately.…”

“…6 In studies examining suicidal ideation and suicides among childhood cancer survivors, patients with brain tumors have been particularly vulnerable to developing these outcomes. 27,33,34 The results of the current study indicate there is an elevated risk of death resulting from risky health behavior-related causes among patients diagnosed with CNS tumors. The psychological burden caused by excess psychiatric morbidity could expose patients with CNS tumors to this risk.…”

“…Compared with the Norwegian study, the cohort in the current study consisted of younger patients and population comparisons and the follow-up time was longer. Moreover, the Norwegian study initiated follow-up for nonsuicidal violent causes of death at the age of 15 years, 27 whereas we did not have such restrictions herein.…”