Data from a population-based study of more than 4.7 million patients in England showed that being diagnosed with cancer conferred a 20% increased risk for suicide compared with the general population.
The study, published in JAMA Psychiatry, found that the period of highest risk was the first 6 months after diagnosis, which may represent a key window to offer psychological support, according to the researchers.
“It is important to conduct this research as an increased risk of suicide has been found in other countries, including the U.S., Sweden, Finland, Italy, Estonia, Japan and Norway,” Katherine E. Henson, MSc, DPhil, from the National Cancer Registration and Analysis Service at Public Health England, told Healio Psychiatry.
“There has not yet been a nationwide study in England to investigate whether the risk of suicide also applies to the population of patients with cancer here,” she continued. “This is important because they represent a different population in terms of cultural attitudes to suicide and health care provision, and it is important for cancer services to be aware of their needs.”
Researchers performed a nationwide, population-based study in England to quantify suicide risk in patients with cancers, analyzing the risk in terms of cancer type and stage, time since diagnosis and sociodemographic factors.
Using longitudinal data from the National Cancer Registration and Analysis Service in England, the investigators examined death certificate data of adult patients with cancer diagnosed from Jan. 1, 1995, to Dec. 31, 2015, with follow-up until Aug. 31, 2017. They measured all deaths in patients that received a verdict of suicide or an open verdict at the inquest, then calculated standardized mortality ratios (SMR) and absolute excess risks (AER).
Of 4,722,099 patients with cancer (50.3% men; 49.7% women), 74.3% were aged 60 years or older when diagnosed.
During the follow-up period, 2,491 adults with cancer died by suicide, which represented 0.08% of all deaths. The researchers reported that the overall standardized mortality ratio for suicide was 1.2 (95% CI, 1.16-1.25) and the absolute excess risk per 10,000 person-years was 0.19 (95% CI, 0.15-0.23). The findings also showed that more men with cancer died by suicide than women.
Patients with mesothelioma were at highest risk for suicide, with a 4.51-fold risk compared with the general population corresponding to 4.2 extra deaths per 10,000 person-years, followed by pancreatic cancer, esophageal, lung and stomach cancers.
The risk for suicide was greatest in the first 6 months after patients received a cancer diagnosis (SMR = 2.74; 95% CI, 2.52-2.98; AER = 1.77; 95% CI, 1.54-2.01); however, the risk remained high for the first 3 years post-diagnosis, according to the results. Henson and colleague noted that this period represents a time when the psychological health needs of the patients should be addressed.
“The take home message for clinicians is that patients diagnosed with cancer may be at theoretical risk of suicide very early on after their diagnosis — we found risk to be highest in the first 6 months,” Henson said. “This highlights the need to be vigilant for psychological distress and potential suicidality from the point of diagnosis, identifying those who may need further psychological support and screening for signs of mental illness and for suicidal thoughts. Evaluations of models of care suggest that integrating psychological support into cancer care, alongside diagnosis and treatment, has the best patient outcomes in terms of improving depression, anxiety and pain.”
In the future, suicide research should move beyond examining the role of mental disorders or small sets of nonspecific, known risk factors, Matthew K. Nock, PhD, department of psychology, Harvard University, and colleagues wrote in a related editorial.
“What is needed now are studies that provide data on novel constructs...as they unfold in real-time, during periods of known heightened risk, that can be used to test the complex and dynamic interactions that likely lead to [suicidal thoughts and behaviors],” they wrote. “Given that theory-driven general models have not gotten us very far, now is the time for us to also be more inductive and idiographic in our research.” – by Savannah Demko
Disclosure: The authors report no relevant financial disclosures. Nock reports funding from the Grant Gordon Foundation.