Suicide and Accidental Death Among Women With Primary Ovarian Cancer: A Population-Based Study

Background Women with ovarian cancer had the highest suicidal rate among all patients with gynecological malignancies, but no large studies about suicide and accidental death for women with ovarian cancers in detail were conducted. We aimed to determine the relative risk of suicide and accidental death among patients with ovarian cancer to that of the general population, and to identify risk factors associated with suicide and accidental death. Methods Data are from the SEER (surveillance, epidemiology, and end results) cancer registry of women diagnosed with ovarian cancer data from 18 registries for the years 1973–2016. The study population comprised 149,204 patients after inclusion and exclusion criteria were applied. Standardized mortality ratios (SMRs) were calculated and Fine-Gray models were fitted to identify risk factors associated with suicidal and accidental death among cancer patients, with stratifications on demographic and tumor-related characteristics. Results Women with ovarian cancer had a higher risk of suicide and accidental death than the cancer-free group [SMR = 1.86; 95% CI (1.54–2.25) and SMR = 1.54; 95% CI (1.39–1.71)]. Subgroup analysis indicated that only patients with type II epithelial ovarian cancer [SMR = 2.31; 95% CI (1.83–2.91)] had an increased risk of suicide, and those with type I and type II epithelial ovarian cancer [SMR = 1.65; 95% CI (1.39–1.97) and SMR = 1.49; 95% CI (1.30–1.70)] were at a higher risk of accidental death. Patients with ovarian cancer who were younger, white, diagnosed with high-grade, non-metastatic cancer and pelvic exenteration were at a higher risk of suicide. The advanced age, earlier year of diagnosis, and non-metastatic cancer were associated with a higher risk of accidental death. Additionally, pelvic exenteration increased the risk of suicide but not the risk of accidental death among women with primary ovarian cancer. Conclusions Women with ovarian cancer had a higher risk of suicide and accidental death compared with the general population. The findings suggested that clinicians should identify high-risk subgroups of ovarian cancer patients for suicide and accidental death as early as possible, with appropriate prevention strategies.


INTRODUCTION
Suicide is the act of intentional death, which is the tenth leading cause of death in the US, claiming the lives of more than 48,000 individuals in 2018 (1). It not only represents an individual tragedy but also takes a staggering toll on global public health. Accidental death is an unnatural death that is caused by unintentional injury. The hitherto unknown suicidal deaths may be misclassified as death from accidental or unintentional injury (2,3). Both suicide and accidental death are regarded as external causes of death, and they share common risk factors such as physical, social, and emotional dysfunction (4). Though the 5year survival rates of cancer patients have increased over the past decades, the risk of dying from suicide and accidental death for those patients remains elevated over time (5). Epidemiological studies have shown that the rate of death from suicide and unintentional injury in cancer patients was 1.9 (5) and 1.6 (6) times those of the general population, respectively.
Gynecologic cancers represent a major problem affecting women's health. Kristy et al. reported that women with gynecological malignancies were at a higher risk of suicide compared with those diagnosed with non-gynecological malignancies (7). Among gynecologic cancers, the patients with ovarian cancer had the highest suicidal rate among patients with gynecological malignancies (8). It was estimated that approximately 75% of ovarian cancer occurs as an advanced disease (9) because of the lack of effective screening methods and non-specific symptoms (10). This resulted in a high recurrence and mortality rate in ovarian cancer patients. Severe pain, high recurrence rate, intense treatment, and inferior quality of life from the disease itself and adverse effects of treatment often caused serious depression and anxiety among patients with ovarian cancer, which could put them at high risk for suicide and accidental death (11).
Previous studies have shown higher suicide rates in ovarian cancer patients compared with other gynecological malignancies (8), however, a direct comparison of the risk of suicide and accident injury especially among ovarian cancer with the general population has not been undertaken, and there is currently no literature on characteristics, incidence, and risk factors for suicide and accidental death in women with ovarian cancers in details.
As such, to address the current lack of evidence, we conducted a population-based analysis of the suicide and accidental death among patients with ovarian cancer. The objective of this study was to determine the relative risk of suicide and accidental deaths among ovarian cancer patients compared with the general population and to identify demographic and tumor-related characteristics, such as age at diagnosis and histological subtypes, that are associated with a particularly high risk of suicide and accidental deaths among ovarian cancer patients.

Patient Selection
The Surveillance, Epidemiology, and End Results (SEER) database (12) was established by the National Cancer Institute, covering ∼28% of the US population. The Public Use version of data collected from the SEER18 registries (12) between January 1, 1973, and December 31, 2016, was used for this study. As a comparison, the mortality data of the general US population collected by the National Center for Health Statistics spanning from 1969 to 2016 was used.

Study Population and Study Variables
To avoid the influence of the second primary tumors on the outcome of cancer patients, we only included patients diagnosed with the first primary ovarian cancer at the beginning of this study. Patients were excluded if the diagnosis was made at autopsy or obtained solely from the death certificate, and those without definite data on age at diagnosis, income, educational level, and survival time were also excluded in this study (Figure 1).
Available data about demographic characteristics from the SEER database included age at diagnosis, race, and calendar year of diagnosis. The tumor-related variables encompassed pathological type, stage, and grade. Income (median family income) and educational level (percentage of residents >25 years of age with at least a bachelor's degree) were categorized into quartiles. Residence types (rural/urban) were obtained at the county level from the SEER program. Data on ovarian cancer treatment, follow-up time, and cause of death were also available. Ovarian cancer was categorized into five histological subgroups according to the International Classification of Disease for Oncology third revision (ICD-O-3) codes: type I epithelial, type II epithelial, germ cell, sex cord-stromal, and other (13,14). Surgeries for ovarian cancer were categorized into "Oophorectomy without hysterectomy, " "Oophorectomy with hysterectomy, " "Surgery, not otherwise specified (NOS), " "None, " "Debulking, " and "Pelvic Exenteration." Patients were considered to have committed suicide if the cause of death variable was coded as "Suicide and Selfinflicted Injury (50220)" including International Classification of Diseases, Eighth Revision codes (ICD-8): 950-959, ICD-9: 950-959, and ICD-10 codes U03, X60-X84, and Y87.0. Patients were considered to have committed accidental death if the cause of death variable was coded as "Accidents and Adverse Effects (50210)" including ICD-8: 800-949; ICD-9: 800-949; and ICD-10: V01-X59 and Y85-Y86.

Statistical Analysis
The number of suicides or accidental deaths divided by personyears of survival was calculated as the mortality of suicidal or accidental death. Among subgroups of cancer patients stratified by different characteristics, SMRs were calculated as the ratios of observed to the expected number of deaths (5,6,15), which provided the relative risk of death from suicide and accident injury for cancer patients compared with all general US population with the same distribution of age, sex, and race. The observed values represented the number of suicides or accidental deaths in patients with ovarian cancer, and the expected values represented the number of individuals who died by suicide or accident in the general population. A 5-year age range was used for standardization, and the 95% CI of SMR was determined by using the Poisson distribution approximation. Interaction tests were further carried out to investigate the potential differential effect of cancer subtypes on suicide risk during different follow-up periods. Given that other causes of death were considered as competing events, we also designed the Fine-Gray model to identify demographic and tumor-related characteristics associated with a higher risk of suicide and accidental death. Observations were censored if patients did not die from suicide or accidental death at the time of the last follow-up. The survival time recorded as 0 month in the SEER database was converted to one-half of a month based on accepted epidemiologic practices (15).
All statistical tests were two-sided, and values with P < 0.05 were considered statistically significant. The SEER database was accessed using SEER * Stat software 8.3.8 (12). The calculation of SMRs, cumulative incidence curves, Gray tests, and interaction tests were conducted in R version 3.51 statistical software. These histograms summarizing the results of subgroup analyses were drawn by Microsoft Excel and GraphPad Prism 8.0.

Incidence of Suicide and Accidental Death Among Patients With Ovarian Cancer
A total of 107 suicidal and 373 accidental deaths occurred among 149,204 patients with ovarian cancer, followed by 768,840.17 person-years ( Table 1). The suicide rate was 13.92/100,000 person-years and the accidental death rate was 48.51/100,000 person-years among ovarian cancer patients. Women with ovarian cancer had an 86% higher risk of suicidal death [SMR = 1.86; 95% CI (1.54-2.25)] and 54% higher risk of accidental death [SMR = 1.54; 95% CI (1.39-1.71)] than the general US population with the same distribution of age, sex, and race. The number of suicide and accidental deaths among patients with ovarian cancer ranks second in patients with gynecologic cancers, between uterine cancer and cervical cancer. From 1973 to 2017, the relative number of suicide and accidental deaths decreased gradually in patients with uterine cancer and unchanged among patients with cervical cancer, while the relative number of suicide and accidental deaths remained elevated among patients with ovarian cancer (Figure 2).

Demographic Characteristics Associated With Increased Incidence of Suicide and Accidental Death
There were 27 suicides and 77 accidental deaths among patients with ovarian cancer who were younger than 50 years old. These younger cancer patients had almost the same risk of suicide as the general population with the same distribution of age, sex, and race [SMR = 1.25; 95% CI (0.86-1.83)]. But a significantly higher incidence of suicide and accidental death in patients over 50 was observed. Among cancer patients diagnosed between 50 and 59 years old, there were 36 suicides of 32,298 patients, accounting for 33.6% of all suicides among patients with ovarian cancer ( Table 2). While for patients whose age at diagnosis was   more than 50 years old, the suicide risk gradually decreased with age but increased after the age of 80. Figure 3 showed

Suicidal and Accidental Death Risk in Different Histological Subtypes of Ovarian Cancer
Only patients with Type II epithelial ovarian cancer had an increased risk of suicide, and those with Type I and Type II epithelial ovarian cancer were at a higher risk of accidental death compared with the general population (Figure 4)

Internal Comparisons: Predictors of Suicide and Accidental Death Based on Fine-Gray Model
The Fine-Gray model revealed that age was found to be a protective factor for suicide [HR = 0.98; 95% CI (0.97-0.99); P < 0.001] but an independent risk factor for accidental death [HR = 1.02; 95% CI (1.01-1.03); P < 0.001] among patients with ovarian cancer (  Figure 5 showed the cumulative incidence of suicide and accidental death among patients who underwent different types of surgery throughout follow-up. Moreover, patients with localized or regional ovarian cancer had a significantly high risk of suicide and accidental death. The high-grade cancer population had a higher rate of suicide risk while the risk of accidental death was the same as that of the low-grade cancer population.

Suicidal and Accidental Death Risk Over Time After Diagnosis
The relative increase in suicide risk among ovarian cancer patients was the highest within the first year after initial diagnosis and subsided with longer follow-up time ( Table 5). Ten years after the cancer diagnosis, the suicide risk of cancer patients was equal to that of the general population. The SMR for accidental death declined within the first 5 years of cancer diagnosis and then elevated after 5 years ( Table 6). The highest relative risks of suicide and accidental death persisted over time in patients with type II ovarian cancer.

Main Findings
Using data from the SEER database, we first found that a relative risk of 1.9 for suicide and 1.5 for accidental death among women with ovarian cancers compared with the general US women population. Of note, we found that patients with Type II epithelial ovarian cancer faced an increased risk of both suicide and accidental deaths in comparison with the general population, and the patients with Type I epithelial ovarian cancer also had a greater risk of accidental deaths. Given that suicide is potentially preventable, this analysis is important as it identifies the high-risk groups of death from suicide and accidental injury among ovarian cancer patients, providing a preliminary heuristic framework for clinicians to develop suicide risk-identification and prevention strategies to reduce the risk of suicide among women with ovarian cancer.

Strengths and Limitations
Despite many important findings of an association between ovarian cancer and the risk of suicide and accidental death, there are some limitations in our study. First, the SEER database does not contain detailed information regarding chemoradiotherapy such as cycles and dosage, which renders it difficult to evaluate the impact of chemoradiotherapy on the risk of suicide and accidental death. Second, data on psychiatric status, quality of life, and social support are also unavailable in the SEER database. Third, data on a history of psychiatric disorders, complications of cancer and treatment, the quality of life, and social support are also unavailable in the SEER database, therefore, we cannot assess the direct association between these factors and the risk of suicide and accidental death among patients with ovarian cancer. Nevertheless, this study is the first large population-based study on the risk of suicide and accidental death in patients with  (16). In the context of the poor prognosis, patients with type II epithelial sarcoma would suffer more psychological concerns, including anxiety and depression. They could feel isolated and hopeless due to the lack of survivor groups. All these may lead to intense suicidal ideation among women with type II epithelial ovarian cancer. We also found that the risk of suicide was the highest within the first year after initial diagnosis among patients with ovarian cancer. This is a novel finding of the present study, and our results stand corroborated by other cancer studies which showed an elevated risk of suicide within the first year of diagnosis compared with cancer-free controls. Mahdi et al. reported the highest suicide rates within the first year following diagnosis among women with gynecologic cancer, with an SMR of 2.8 (8). Further, prior research has shown that significantly higher risks of suicide were observed among patients diagnosed with breast cancer (19), brain cancer (20), male genital-system (21), and skin malignant melanoma (22) within the first year of diagnosis. We thought it reasonable as the change from being a healthy individual to being a cancer patient who had a poor prognosis created a new perception of identity. During this process of shift, patients with ovarian cancer often suffer from severe psychological distress and problems in social life (23). This novel finding underlines the concept that healthcare providers should better understand the social psychology and identity changes of newly diagnosed patients with ovarian cancer to better guide ovarian cancer survivors during follow-up. Many methods could be utilized to evaluate the level of mental health and suicidal ideation among cancer patients, such as item 9 of the Patient Health Questionnaire depression module, Center for Epidemiologic Studies Depression Scale (CES-D), the Impact of Event Scale (IES), and the Profile of Mood States short-form (POMS-SF) (24).
Within the ovarian cancer cohort, we observed that those who underwent pelvic exenteration were at the highest risk of suicide. Pelvic exenteration is a radical surgical procedure that removes the visceral pelvic organs including the uterus, tubes, ovaries, parametrium, vagina, urinary bladder, urethra, and rectum, with or without the perineum in an en-bloc fashion (25)(26)(27). It is considered the last curative opportunity for malignant gynecological tumors (28). However, it is related to various complications, high costs of substantial treatment, and high mortality. A survey conducted in New Zealand reported 106 complications, such as intra-abdominal collection (43.7%) and wound infection (14.1%), out of a total of 646 consecutive patients who required extended surgery for local advanced pelvic malignancies (29). Moreover, severe gastrointestinal and urinary tract symptoms such as dyspareunia, as well as decreased sexual functions and sexual desire, have also been observed among ovarian cancer patients, especially for those who underwent pelvic exenteration (30,31). These adverse effects and accompanying symptoms had an obvious negative impact on global health status, body image, self-identity, social functioning, emotional response, and quality of life among these survivors (32). Therefore, these survivors who underwent pelvic exenteration have an increased risk of suicide and need close monitoring, and unnecessary operations should be avoided to reduce the risk of suicide in patients with ovarian cancer (33,34).
Ovarian cancer occurs more in patients older than 50 years, similarly, and we found that patients diagnosed over 50 were at higher risk of suicide, especially between the ages of 50-59. Our results stand corroborated by Stephanie's study which found that female patients with cancer whose age at diagnosis between 55 and 59 had the highest suicide risk among all female cancer survivors (5). Furthermore, it was well-recognized that older age was associated with a higher risk of suicide among patients with prostate cancer, lung cancer (35), colorectal cancer, and bladder cancer (36). The working hypotheses for our finding may be that ovarian cancer patients with older age may suffer more severe emotional and psychological distress. Payne reported that some symptoms, such as the anxiety and depression of recurrent disease and death, as well as sleep disorders, may persist longer in elder patients (37). An Australian survey using the Insomnia Severity Index (ISI) to assess the degree of insomnia also confirmed that ovarian cancer patients aged 50-59 present had clinically higher levels of insomnia (38), which has been proven to be correlated with suicidal thoughts and attempts (39). Moreover, older patients could get less information about cancer and psychological support than younger patients from the internet (40,41), therefore, they lack the confidence to defeat cancer (42), leading to relatively higher suicidal ideation (43). Given the great difference in suicide risk between young and old survivors with ovarian cancer, it is necessary to regard old patients as a distinctive group that warrants special attention when considering the suicide risk of cancer patients.
In conclusion, our study demonstrated that patients with ovarian cancer, especially type II ovarian cancer, had an increased relative risk of suicide and accidental death to that of the cancer-free population. Further, it is worth noting that pelvic exenteration for ovarian cancer would increase the risk of suicide, but not increase the risk of accidental death. These meaningful findings suggest a need in clinical work to detect suicidal ideation as early as possible especially for high-risk subgroups (44,45). We can develop an accurate predictive model to predict the risk of suicide and accidental death in patients and to pay more attention to the high-risk group, and appropriate psychosocial interventions and treatment are also essential, as clinicians could provide psychological counseling to cancer patients to reduce their anxiety, depression and loneliness, and strengthen their confidence to face cancer and help them return to community life, such as psychosocial and emotional support through interpersonal interactions during survivorship care, which potentially reduce the risk of suicide in cancer survivors. In this way, a better balance can be achieved between achieving a better overall prognosis and reducing the risk of suicide and accidental death (46,47). Future studies should explore the direct relationship between psychosocial factors and the risk of suicide and accidental death among ovarian cancer patients, which may discover new interventions that can improve the survival of patients diagnosed with ovarian cancer.