he Global Burden of Disease Collaboration reported 24.6 million incident cancer cases for 2017, causing 233.5 disability-adjusted life-years and an increasing cancer incidence rate during the last 1.5 decades in 123 of 195 countries 1 . For a patient, a diagnosis of cancer is one of the most dramatic and life-changing events, representing a severe psychosocial stressor 2 . Patients with cancer may die by suicide for a number of reasons, including anxiety, pain, loss of perspectives and previous coping strategies, treatment-related adverse events and fatigue 3 . Suicide is a global public health concern. More than 800,000 people die by suicide each year, with 20 suicidal attempts being carried out for each death by suicide 4 .Suicidal ideations among patients with cancer may arise through a patient's will for self-autonomy and self-control, but they can also occur in the context of severe depressive symptoms 5 . Whatever the cause, suicides are difficult to handle for next-of-kin and caregivers, who may be affected far beyond the loss of the patient 6 .The effectiveness of comprehensive psychological, psychiatric and psychotherapeutic management of patients with cancer is underscored by the estimation that every US$1.00 spent on psychotherapeutic interventions and interventions that strengthen linkages among different health-care providers saves US$2.50 in the cost of suicides 7 .Increased suicide rates in patients with cancer compared with the general population have been observed for decades in different populations around the globe 8-23 . However, only few meta-analyses examined suicide risk among patients with cancer and those studies lacked a comprehensive literature search, generated incomplete findings regarding certain risk groups and/or analyzed only specific cancer entities, and/or presented only incidence rates without comparisons with the general population [24][25][26][27][28][29] . The research objectives of the present study were, therefore, to quantify overall suicide mortality in patients with cancer compared with the general population, followed by comprehensive subgroup analyses to identify specific risk factors of particular interest, including cancer prognosis, cancer stage, time since diagnosis, gender, ethnicity, marital status and geographic region, which have not been performed in a meta-analysis to date. Our aim was to gain a deeper knowledge of risk factors for suicide in patients with cancer to enhance oncological and psychiatric practice, improve the quality of life of patients with a cancer diagnosis and help develop targeted interventions.
ResultsStudy population. Our database search led to 12,188 records; an additional 5 studies were manually added by the authors. After exclusion of duplicates, 7,565 records were screened by titles and abstracts. A total of 768 of 934 reports did not meet the inclusion criteria and were excluded during full text screening (Fig. 1). In 62 of the remaining 166 cohort studies, results were presented as standardized mortality ratios (SMRs) and were ther...