Suicide was ranked as the 11th leading cause of death in the United States, responsible for 33,289 deaths¹ in 2006. Risk factors associated with suicide in the general population include chronic diseases, pain, depression, both older age and youth, living alone, and unemployment. Gender also influences suicide rates with males at 4 times higher risk of suicide than females. Suicide was the seventh leading cause of death for males and the 16th for women² in 2006. In addition, Americans 65 years or older have a higher incidence of suicide. According to the National Institute of Mental Health, 14.2 of 100,000 older Americans died as a result of suicide in 2006 as compared to 10.6 of 100,000 people in the general population. More than 90% of people who commit suicide in the general population have depression, mental illness, and/or substance abuse.²

The suicide rate for persons with cancer is estimated to be twice the rate in the general US population.³ Yet oncology nurses are often unable to identify factors that may put the cancer patient at higher risk.⁴ Failure to recognize suicide risk in patients with cancer may lead to further morbidity and mortality. Therefore, the purpose of this review of the literature was to identify (1) risk factors associated with suicide in cancer patients and (2) evidence-based assessment methods.

Methods

Search Strategy

PubMed, CINAHL, and PsycINFO databases were searched and literature identified within an 11-year period (1999–2010); these articles reported data collected over several decades. The literature search was followed by a hand reference list checking for relevant articles from these 3 databases. Medical Subject Headings terms used in PubMed were neoplasm, psychology, and suicide in adults. MH exact Subject Headings used for CINAHL were (MH “Neoplasm’s/PF”) and (MH “Suicide”). PsycINFO search used DE “Neoplasms” and DE “Suicide”. Searches were confined to human, English, and adults.

Inclusion Criteria

The sources reviewed were research articles published in peer-reviewed journals within the specified 11-year period. The studies were peer reviewed and were relevant to cancer and suicide in the adult population. They included 1 or more of the variables associated with cancer and suicide such as population demographics (gender, age, type, and site of cancer), suicide rates, risk factors including psychological well-being, suicide and depression, assessment tools, and prevention measures. Articles using meta-analysis and case-controlled studies were included.

Data Extraction

A summary table to catalog articles alphabetically by author including publication date, citation, sample, methods, findings, and study limitations was created (see Table). Key findings were organized in this table for further analysis.

Results

After the database search, 193 articles were identified that mentioned suicide and cancer. However, when the inclusion criteria were applied, 22 articles were selected; 2 additional articles were retrieved from reference lists from these 22 articles for a total of 24 articles for this review. Among the 24 articles reviewed, 13 reported suicide risk factors, 9 evaluated tools used to detect increased suicide risk, and 2 discussed suicide prevention strategies for the terminal cancer patient. Major findings associated with suicide rates included type of cancer, gender, age, depression, and time from diagnosis.

Cancer Type

Four studies identified specific cancer sites associated with increased suicide risk, whereas 21 articles reported results under the broad category of cancer. The 4 site-specific malignancies with higher suicide rates included prostate, pancreatic, lung, and head and neck. In a population-based retrospective cohort study of older men living in South Florida from 1983 to 1993 who committed suicide, a total of 667 suicides were completed of which 20 were in men with prostate cancer.¹² The incidence of suicide in this study population was 55.32 per 100,000 older men, compared with 274.7 per 100,000 for older men with prostate cancer. Overall, the risk of suicide in men with prostate cancer was 4.24 times the age- and gender-specific cohort.¹² Miller et al¹⁴ examined 1 of the first population-based studies to establish the relative risk of suicide in Americans 65 years or older while controlling for medical and psychiatric comorbidity. In a multivariate analysis, the only medical illness associated with suicide was cancer (odds ration [OR], 2.3; 95% confidence interval [CI], 1.1–4.8). There were a total of 19 suicides among patients with a cancer diagnosis of which 8 were in men with prostrate cancer. This study demonstrated a suicide rate of 23% (8 of 35) among those patients with prostate cancer.¹⁴

In a review of patients with cancer, Passik and Breitbart¹⁵ focused on the connection between pancreatic cancer, depression, and suicidal ideation. Examination of psychiatric consultation data at Memorial Sloan Kettering Cancer Center revealed that one-third of cancer patients found to have suicidal ideation were patients with major depression.¹⁵ The study reported that for cancer patients with localized disease, only 25% experienced depressive symptoms compared with 77% of patients with advanced cancer disease. Similar findings have been reported in patients with pancreatic cancer who are usually diagnosed at an advanced stage. In a summary of 52 case reports of patients with pancreatic cancer, 71% had symptoms of depression, 48% had anxiety-related disorders, and 29% had both.¹⁵ The authors reported that suicidal ideation is likely to be associated with depression in patients with advanced cancer.

Akechi et al⁷ evaluated predictive factors for suicidal ideation in 89 newly diagnosed patients with unresectable lung cancer. He found that 15% of patients had some degree of suicidal ideation at 6 months after initial diagnosis. Risk factors, identified using multivariate analysis, indicated that pain at baseline (OR = 3.72) and the development of a depressive disorder (OR = 27.97) were significantly associated with development of suicidal ideation. The report concluded that comprehensive cancer care should include pain management and appropriate psychiatric intervention in an effort to minimize suicidal ideation. A retrospective review by Kendal¹¹ of 1.3 million cancer cases from the SEER database found a significant association between head and neck cancers in men and suicide (0.32%; 95% CI, 0.26–0.39).

In summary, having prostate, lung, head and neck, or pancreatic cancer was associated with higher rates of suicide or suicidal ideation. These studies were limited as they were primarily retrospective analyses of case reports. The Akechi trial was small, and it is difficult to apply these findings to other cancer populations. It is noteworthy that there have not been prospective studies evaluating depression and suicide in common malignancies such as breast and colon cancer.

Gender Differences

In the general American population, males are 4 times more likely to commit suicide than females.¹¹ Five articles reported male gender as a risk factor for completed suicides. One retrospective study from the Cancer Registry in Norway reported that of the 490,245 patients who died from cancer, 589 committed suicide. The relative suicide risk was increased for both men and women; however, men were at higher risk with a standardized mortality ratio (SMR) of 1.55 (95% CI, 1.41–1.71) as compared with women, who had an SMR of 1.35 (95% CI, 1.17–1.56).¹⁰

A large Danish study included 564,517 cancer patients diagnosed between 1971 and 1999 and found that 1241 had died as a result of suicide (0.22%).¹⁹ The suicide rate in this study was higher for men than women. The age standardized (world) suicide rates (WSTP) per 100,000 ranged from 16.8 to 30.2 for men and 6.1 to 16.3 for women. A similar large retrospective study from the Thames Cancer Registry in England from 1996 and 2005 reported a significantly increased risk of suicide in men. The study reported an SMR of 1.45 (95% CI, 1.20–1.73) for men compared with 1.19 (95% CI, 0.88–1.57) for women.¹⁷ A gender comparison study of 1.3 million cancer patients from the US Surveillance, Epidemiology, and End Results (SEER) registry, a population-based database regarding the pathology, disease extent, social factors, treatment, and causes of death of people with cancer in the United States, also noted that male sex is a risk factor.¹¹ The results of this study concluded that women completed suicide rate (0.02%) was about one-fifth that of men (0.1%). The suicide hazards ratio for women was 1 and men, 6.2. This study showed a 4.8 times excess in overall number of male suicides over female suicides, which is consistent with the general American population ratio¹¹ of 4.5.

In summary, as in the general US population, suicide risk is higher in male cancer patients than in female cancer patients. Extensive population-based studies have confirmed the increased rate in the male population. There are no studies that make this comparison by gender or in a specific cancer diagnosis such as lung cancer.

Older Oncology Patients

In the US population, suicide in the context of medical illness is higher in older adults.¹⁴ Miller et al¹⁴ focused on the risk of suicide in older Americans with cancer. In a case-controlled study of 1408 patients aged 65 years and older, he noted that 128 had died as a result of suicide during a study period from 1994 to 2002. In an adjusted analysis, the only medical condition associated with suicide was cancer (OR = 2.3). The authors concluded that the risk of suicide in older adults was higher among patients with cancer than among patients with other medical illnesses.

Misono et al³ reviewed the SEER database in an effort to identify cancer patient and disease characteristics associated with higher suicide rates. This study found higher suicide rates in men (SMR = 2.09), white race (SMR = 1.88), and older age at diagnosis (SMR = 2.42 in patients aged 65–69 years). The age-adjusted suicide rate in the general population was 22.0 per 100,000 person years for ages 80 to 84 years, yet this study revealed a suicide rate of 52.4 among the same age group with cancer. The higher rates were notable among men who demonstrated a suicide rate of 100.3 per 100,000 in those patients aged 80 to 84 years. This study also found an increased suicide rate in older (>65-year) patients with cancer. This risk increased as age increased and older men were at the highest risk. An analysis of cancer type would help further define special populations at higher risk of suicide.

Depression/Hopelessness

In a recent study of patients with cancer, depression was determined to be the major risk factor for suicidality (r = 0.36–0.39, P < .01). Hopelessness was also reported as increasing the risk for suicide (r = 0.45–0.49. P < .01). In this study of advanced cancer patients by Wilson et al,²⁰ “feeling oneself a burden to others” was found to be a moderate to extreme concern to patients (39.1%). A study by Breitbart et al⁸ found that depressed cancer patients were 4 times more likely to have a desire for hastened death (DHD) compared with those patients without depression (47% vs 12%). A retrospective study of 1721 cancer patients referred for a psychiatric consultation found that 220 had major depression (12.8%) and more than half of these (113) demonstrated suicidal ideation. Major depression was a significant risk factor for suicidal ideation in this study (OR = 1.80; 95% CI, 1.89–2.37; P = .0001).⁷

Recent Cancer Diagnosis

Hem et al¹⁰ studied a cohort of patients from the Cancer Registry of Norway who were linked to a suicide diagnosis in the Registry of Deaths from 1960 to 1997. During this period, 589 cancer patients committed suicide; 407 were males. The risk of death from suicide was highest in the first months after diagnosis. The SMR was 3.09 for men and 2.18 for women within the first 5 months of diagnosis. After 12 months from diagnosis, the SMR decreased to 1.57 for men and 1.72 for women. Hem et al¹⁰ concluded that the relative risk was elevated for both sexes in the first months after diagnosis (P < .001) and significantly decreased with time (P = .005).

In a recent study of suicide in cancer patients from England, Robinson et al¹⁷ found that the relative risk of suicide was greatest in the first year after cancer diagnosis. The SMR was 2.42 for men and 1.44 for women in the first year. The authors concluded that there was a critical period just after diagnosis in which suicide risk was high. Yousaf et al¹⁹ explored suicides among Danish cancer patients from 1971 to 1999. They reported that after a cancer diagnosis, suicide risk was highest in the first 1 to 3 months for men and between 3 and 12 months for women.¹⁹

In summary, multiple studies have identified the first months after cancer diagnosis as higher risk for suicide. More data exploring specific cancer diagnosis and extent of disease may be helpful in further defining this risk. Interventions identifying early detection of depression or suicidal ideation may help prevent suicide.

SUICIDE RISK SCREENING TOOLS

Depressed cancer patients are more likely to have a high DHD compared with patients without depression (47% vs 12%).⁸ Consequently, systematic approaches to screening for suicide ideation with validated instruments hold promise for improved mental healthcare for patients with cancer. Eight studies from a general oncology population explored various screening tools in an effort to identify suicide risk. Tools used to assess depression in these articles included the Beck Hopelessness Scale (BHS); Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria; Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, Revised Third Edition (SCID); Endicott criteria using the Hamilton Depressive Rating Scale; Edinburgh Postnatal Depression Scale; Hospital Anxiety and Depression Scale; Schedule of Attitudes towards Hastened Death; Patient Health Questionnaire (PHQ-9); 13-item structured interview of symptoms and concerns; and simply asking the question “are you depressed.”

Clinical signs and symptoms of major depression are often also present as signs and symptoms in cancer patients making it difficult to differentiate them.⁹ Anorexia, weight loss, low energy, and sleep disturbances are common as a result of both disease and treatment processes. Some study authors have proposed to drop these somatic symptoms from assessment screening tools that evaluate depression in this population. The prevalence of major depression in cancer patients has been recorded as low as 5% to 6% in 1 study to as high as 40% in others.⁹ This variability is most likely related to differences in how depression was measured in these studies. The gold standard for the diagnosis of depression is a clinical examination administered by someone trained to use the SCID; instead, many studies used brief screening with varying psychometric qualities, cancer type, and disease stage.⁹ A study by Ciaramella and Poli⁹ using 2-structured methods for assessing major depression found that 49% of patients were considered depressed when evaluated using the SCID screening tool; however, when somatic symptoms were eliminated using the Endicott criteria, only 29% were depressed. Using both depression tools, the SCID and Endicott criteria, 28% were found to have a current depressive episode. These results suggested that the prevalence of major depression in cancer patients was 28%, which is similar to depression in other medical illnesses.⁹ The sensitivity and specificity of self-reported depression screening have been studied for both cancer patients and those with mental illness. Symptoms, age, and gender were found to bias scores.⁹

Hopelessness is a powerful predictor of suicidal ideation and completed suicides.⁵ Abbey et al⁵ found that the BHS could be improved when used with the terminally ill, by removing problematic items like the question “I can’t imagine what my life will be like in ten years” or “I have enough time to accomplish the things I want to.” This study examined abbreviated versions of the BHS for use in the terminally ill population. All 3 versions reported adequate reliability and validity as measures of hopelessness for the terminally ill. When compared with the original 20-item scale, the 7- and 13-item versions had slightly more variance at end of life despair measures (ideation and hastened death). The 3-item version performed well accounting for as much variance as did the longer original. In summary, multiple tools have been used to help identify those patients who may be at high risk for suicide because of depression, helplessness, or DHD.

Desire for hastened death, as defined by Abbey and colleagues,⁵ in the context of advanced cancer is a multidimensional construct, not necessarily pathological, and may have multiple meanings unrelated to taking one’s life. Three distinct experiences of the meaning include DHD as a hypothetical exit plan, an expression of despair, and as a manifestation of letting go. It often served as a adaptive purpose of managing distress.¹⁶

The PHQ-9 is a validated self-report measure that is used to screen for mental disorders.^21,22 One item of the PHQ-9 asks, “In the last 2 weeks how often have you been bothered by the following problem: thoughts that you would be better off dead or hurting yourself in some way?”. In a study of 330 cancer patients, an association was found for those who endorsed this question and suicidality; those with higher scores were more likely to be suicidal on interview.

Discussion

Cancer patients have a higher rate of suicide than the general population.^23,24 A number of factors place the cancer patient at higher risk for suicide including having prostate, lung, pancreatic, or head and neck cancer and being male, older, depressed, and recently (within the first year) diagnosed with cancer. Suicidality was also higher in adult survivors of childhood cancers.²⁵ In general, cancer patients who have significant physical, psychological, and social impairments may be at greater risk for suicide.²⁶

Depression is a well-documented risk factor for suicide in cancer patients.²⁶ Depression and hopelessness are the strongest predictors of a desire for death in terminally ill cancer patients.⁸ It has been established that depression in cancer patients adversely affects quality of life, including length of survival, adherence with care, and perception of pain. Depression is often undetected and underdiagnosed in cancer patients, and failure to treat is a major concern. Based on this research, use of the shorter screening tool (eg, the 3-item BHS, which includes [1] in the future, I expect to succeed in what concerns me the most; [2] all I can see ahead of me is unpleasantness rather than pleasantness; [3] it is very unlikely that I will get any real satisfaction in the future) to assess for depression and suicidal ideation in cancer patients is recommended. There is enough evidence to support use of this tool in current clinical practice. Alternatively, simply asking “are you depressed” may ease the use of suicide risk tools. Implementation of screening tools, however, will require clinical consultation with psychiatric experts, for instance, referral to a psychiatrist, a psychiatric-mental health nurse practitioner or clinical nurse specialist, social worker, psychologist, or other mental health professional.

Special populations were identified who may be at especially high risk of suicide. Men 65 years or older with lung, pancreatic, head and neck, or prostate cancer were identified in multiple studies to be at particularly high risk. Further studies, which focus on these patients, are needed in an effort to decrease the high rate of suicide in special oncology populations. However, current studies are limited by their focus on white populations and exclusion of minority groups. Future studies should focus on minority populations with cancer. Published population studies also failed to explore underlying mental illness, such as anxiety and major depression, and will likely bias results. The most glaring omission in current studies is the lack of screening tools to help prevent suicide in cancer populations. Further studies should be initiated to explore suicide prevention strategies in high-risk populations. Given these studies, one may speculate that cancer site does make an important difference in suicidality and therefore suggest a differential clinical approach to the identification and management of suicide.

Secondary prevention of suicide, defined as decreasing the likelihood of a suicide attempt in high-risk patients, is an important goal in the care of individuals with cancer.²⁵ Increased awareness among healthcare providers for cancer patients at greater risk may be the key to help decrease preventable deaths in this population. Early identification of and intervention with cancer patients at high risks of suicide should impact rates of death. A secondary benefit of early detection is to identify those risk factors, which are known to contribute to suicidality including depression, distress, and pain and to make appropriate interventions. These interventions include pharmacological and psychological interventions, follow-up care, and reduced access to lethal means.²⁷

Implication for Future Research

Current research is limited, and prospective trials should be funded to further refine suicide risk using detection tools and to develop best practices for prevention. Special populations, such as older men with prostate cancer, should have interventions developed and tested to minimize suicide risk. Early identification with valid and reliable screening tools combined with these interventions should lead to decreased morbidity and mortality. Further research should also explore other factors such as marital status, socioeconomic factors, and ethnicity.

Implications for Healthcare Practice

Although a relatively rare event, healthcare providers should be aware that having a diagnosis of cancer increases the risk for suicide. More specifically, prostate, lung, pancreatic, and head and neck cancer; male sex; older adult; recent diagnosis; and depression have all been associated with an increased incidence of suicide. All of these variables except depression are easily identified by²⁸ the clinical approach. Incorporating the 3-item Beck assessment tool (16) into the admission flow sheet and simply asking the question “are you depressed” are simple, efficient methods to help identify the depressed cancer patient and alert the provider to either refer or treat these symptoms and arrange for follow-up. The tool should be in place on all intake follow-up questionnaires for all cancer patients as it may be hard to differentiate on specific cancer patients at risk. Educating providers of these increased risk factors especially among those with significant physical, psychological, and social impairments could be accomplished during annual continuing cancer updates.

Again, identifying and treating depression in the cancer patients could not only decrease the risk of suicide but also improve quality of life.