Suicidal Ideation in Long-Term Adult Cancer Survivors: Findings of a Nationwide Cross-Sectional Study

Cancer survivors experience emotional distress on a spectrum that extends from financial worries to facing death (1); suicide is a critical psychologic outcome. A growing body of evidence indicates adult cancer survivors have a higher risk of suicide than the general population (2). However, vulnerability to suicide tends to normalize over time (2). The early phase after a cancer diagnosis is critical, but little is known about such vulnerabilities during later phases (3). Furthermore, factors related to suicide have seldom been considered in this patient population, although it has been previously established that depression is more common in patients with cancer than in the general population (4). Our aim was to confirm the association between long-term cancer survivorship and suicidal ideation after adjusting for potential confounders.

The Institutional Review Board of Gachon University Gil Medical Center approved the study protocol (GFIRB2021-481). Detailed methodology information was reported elsewhere.

We used Korean National Health and Nutritional Examination Survey (KNHANES) data collected in 2007 to 2013, 2015, 2017, and 2018 that included information on suicidal ideation and age at cancer diagnosis. Initially, we identified 56,247 subjects ages 19 to 79 years who had provided data on suicidal ideation (in people ≥ 80 years of age, age at cancer diagnosis was not specified). Cancer survivors were defined as individuals who had received a cancer diagnosis more than 5 years in the past. The time from cancer diagnosis was calculated by subtracting the age at diagnosis from the age at survey completion. After excluding patients recruited within 5 years of their cancer diagnosis, 55,128 subjects were included in the final analysis.

The primary outcome of our study was suicidal ideation, which was addressed using the self-reported yes/no question: “Have you thought about committing suicide within the last 12 months?” The following were included as covariates: demographics (age, sex, economic status, highest education level, marital status, and current job status), health-related habits (smoking status and drinking frequency), comorbidities (obesity, hypertension, type 2 diabetes, and major cardiovascular diseases such as myocardial infarction and stroke), and depressive mood. To assess depressive mood, we used responses to the yes/no questionnaire item: “Have you experienced sadness or despair continuously for more than 2 weeks during the past year?” Patients who were currently being treated for depression were considered as having depressive mood regardless of their response. Suicidal ideation is associated with depressive mood, though sometimes it can occur without depressive mood. Given the significant collinearity between cancer history and depressive mood, logistic regression models for evaluating suicidal ideation in cancer survivors were applied among depressed or nondepressed individuals. All analyses were performed using STATA MP software, version 17.0 (STATA Corp.), and P values <0.05 were considered statistically significant.

The characteristics of participants according to their cancer history are summarized in Table 1. Approximately 1.9% of the 55,128 study subjects were long-term cancer survivors, who were more likely to have comorbidities and to be older, female, poorer, less educated, and unemployed, but had healthier habits than subjects without a cancer history. Higher proportions of survivors reported experiences of depression but not of suicidal ideation. In unadjusted models, cancer survivors did not experience suicidal ideation significantly more often than subjects without a cancer history, both in depressed and nondepressed individuals (Table 2).

Many factors, such as catastrophic reactions to a diagnosis or treatment responses, aggravate the risk of suicide among adult cancer survivors during the early phase after a cancer diagnosis (2, 3). Nonetheless, it remains unclear whether this increased risk persists beyond 5 years. Demonstrating that this is the case has been difficult because few data are available on suicide rates of cancer survivors 5 years or more after a cancer diagnosis. Some researchers (5) have reported suicide rates among long-term survivors and the general population are not significantly different, while others (6) have reported significantly higher suicide rates for cancer survivors. However, these studies did not consider important confounders, especially depression. In addition, most studies focused on actual suicide rates, rather than on suicidal ideation, which is one of the most accurate determinants of the intention to commit suicide (7).

This study has several limitations. First, given the cross-sectional nature of the KNHANES, causal and longitudinal relationships between independent variables and mental health cannot be determined. Future longer-term prospective studies are required to evaluate the risks of psychologic distress caused by a diagnosis of cancer. Next, our results focus on the general Korean population and may not be applicable to Western cultures, in which suicide rates are lower, strategies for dealing with distress are better, and less stigma is attached to cancer (2, 8). Despite these limitations, our findings suggest that long-term adult cancer survivors do not have higher rates of suicidal ideation than members of the general population without a cancer history.

No disclosures were reported.

I.C. Hwang: Conceptualization, data curation, funding acquisition, methodology, writing–original draft, writing–review and editing. H.Y. Ahn: Data curation, formal analysis, supervision, writing–review and editing. H.S. Suh: Supervision, investigation, writing–review and editing.

This work was supported by the Gachon University Gil Medical Center (grant no. FRD2021-14).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.