Longitudinal Risk for Suicidal Self-Directed Violence Among Veterans With Cancer
Cancer care is not only about treating tumors. For many patients, especially veterans, it also requires attention to mental health, pain, functional decline, and social support. A large national study of U.S. veterans with cancer found that suicidal self-directed violence, or SSDV, remains a serious concern for years after diagnosis. SSDV includes both fatal suicide and nonfatal suicide attempts. The findings highlight the need for cancer care teams to screen for suicide risk over time, not just at the moment of diagnosis.
Why this study matters
People with cancer are known to have higher suicide rates than the general population. This may reflect a combination of factors, including physical pain, loss of independence, treatment burden, depression, anxiety, advanced disease, and prior mental illness. Veterans may face additional risks, such as trauma exposure, chronic psychiatric conditions, and barriers to care. Until now, less was known about how suicide-related risks change over time after a cancer diagnosis and which groups remain vulnerable in long-term survivorship.
This study aimed to measure the long-term risk of SSDV among veterans with cancer, identify the most common methods, and determine which patient characteristics were associated with higher risk. The goal was to improve prevention and screening strategies in cancer care.
How the study was done
This was a national cohort study of 292,271 veterans diagnosed with invasive solid tumors or hematologic cancers between January 2014 and December 2023. Researchers used data from oncology registries, suicide registries, and the Veterans Health Administration. They followed patients over time and analyzed outcomes from January 2025 to February 2026.
The main outcome was SSDV, measured as either a fatal or nonfatal suicide attempt. Rates were reported per 100,000 person-years, which is a standard way to compare risk across groups and follow-up durations. The investigators also used multivariable Cox proportional hazards models to estimate adjusted hazard ratios, or aHRs, which show whether a factor is associated with higher or lower risk after accounting for other variables.
Who was included
The average age of the cohort was 69 years. Most veterans were White, and about 21% were Black or African American. Women made up 2% of the group. Smaller proportions were American Indian or Alaska Native, Asian, Hispanic or Latino, or Native Hawaiian or Other Pacific Islander. This demographic mix reflects the veteran population in the study period, but it also means some subgroups were relatively small and should be interpreted with caution.
Main findings
Across the cohort, there were 2,400 SSDV events, affecting about 1% of veterans. The overall rate was 203 events per 100,000 person-years. The most common method was poisoning, including overdose with opioids, and this accounted for 617 attempts, or 26% of events. This finding is important because it suggests that access to medications, especially pain medicines, may influence prevention strategies.
Risk was not evenly distributed. Some cancer types and patient characteristics were linked to much higher SSDV rates. The highest estimated probabilities were seen in veterans with cancers of the central nervous system, pancreas, head and neck, liver and biliary system, and thyroid. These cancers can be especially difficult because they may cause neurologic symptoms, disfigurement, severe symptom burden, or difficult treatment courses.
Several clinical and social factors were associated with higher rates of SSDV compared with the overall cohort. Veterans with severe frailty had a rate of 544 events per 100,000 person-years. Those with advanced cancer had a rate of 261. Veterans with chronic mental illness had a rate of 419, and those with high pain scores had a rate of 236. These findings suggest that physical suffering and psychiatric comorbidity are major contributors to suicide risk.
Groups with particularly high risk
Some subgroups had notably elevated rates of nonfatal suicide attempts. Younger veterans aged 45 years or younger had a rate of 643 events per 100,000 person-years, which was much higher than the cohort average. Women also had a higher rate than the overall group. American Indian or Alaska Native veterans had elevated rates as well. In terms of cancer type, central nervous system and thyroid cancers stood out as having high rates of nonfatal attempts.
These findings are clinically important because they show that suicide risk is not limited to older men with late-stage cancer. Younger patients may struggle with role changes, finances, family responsibilities, and the shock of a cancer diagnosis at an early age. Women and some racial or ethnic minority groups may also face unique stressors or unmet needs.
What happened over time after diagnosis
The study looked at how risk changed after the cancer diagnosis. For many veterans, the hazard of SSDV was highest in the period soon after diagnosis and then decreased over time. However, the risk did not disappear. Five years after diagnosis, certain groups still had elevated risk, including younger veterans, unmarried veterans, veterans with central nervous system cancer, and those with advanced cancer.
For example, younger veterans aged 45 years or younger continued to have higher risk than those aged 46 to 64 years. Unmarried veterans also remained at increased risk, suggesting that social support and partnership may be protective. Veterans with central nervous system cancer and those with advanced cancer continued to show greater long-term vulnerability.
Key statistical associations
At about six months after diagnosis, several factors were associated with increased SSDV hazard. Compared with White veterans, Asian veterans had a higher adjusted hazard ratio. Unmarried veterans also had increased risk. Veterans with central nervous system cancer or head and neck cancer had higher risk than those with lung cancer, and veterans with advanced cancer had a modestly higher risk as well.
The study also observed that overall risk decreased after the early postdiagnosis period for most veterans. This suggests that the first months after diagnosis may be a critical window for intervention, though long-term follow-up is still needed for selected high-risk groups.
Why these findings matter for cancer care
This study reinforces an important point: suicide prevention should be part of routine cancer care. Screening only for depression may miss patients at risk for suicidal behavior. Likewise, focusing only on fatal suicide overlooks nonfatal attempts, which can predict future death and signal urgent distress.
The results suggest that care teams should consider a broader approach that includes:
1. Regular screening for suicidal thoughts and behaviors throughout the cancer journey.
2. Assessment of pain, frailty, and functional decline, not just tumor stage.
3. Attention to psychiatric history, especially chronic mental illness.
4. Review of medication access, including opioids and other potentially dangerous drugs.
5. More support for patients who are younger, unmarried, socially isolated, or living with high-burden cancers.
6. Coordination between oncology, primary care, mental health, palliative care, and social work.
Clinical interpretation
The study does not mean that most veterans with cancer will attempt suicide. The absolute risk remained low overall. But the relative risk was high enough in specific groups that clinicians should not wait for patients to volunteer distress. Many patients may not directly report suicidal thoughts because of stigma, fear, or a desire not to burden others. Structured screening and compassionate conversation can help identify risk earlier.
The prominence of poisoning as a method also points to practical prevention steps, such as safe prescribing, careful medication counseling, limiting unnecessary quantities of high-risk medications, and family education about secure storage and disposal.
Limitations to keep in mind
As with all observational studies, this research cannot prove cause and effect. Some important influences, such as detailed psychosocial stress, substance use, or access to community support, may not have been fully captured in the databases. Also, the cohort consisted of veterans receiving care in the VA system, so the results may not apply perfectly to non-veteran populations.
In addition, some subgroup analyses involved relatively small numbers, especially for less common racial and ethnic groups or specific cancer types. Even so, the study is one of the largest and most detailed examinations of suicide-related outcomes among cancer patients in the veteran population.
Bottom line
Veterans with cancer face a measurable risk of suicidal self-directed violence that can persist for years after diagnosis. The risk is highest in certain groups, including younger patients, unmarried veterans, patients with central nervous system or thyroid cancer, and those with severe frailty, advanced disease, chronic mental illness, or high pain burden.
The message for oncology teams is clear: suicide prevention should be built into cancer care from the beginning and continue through survivorship. Tracking both fatal and nonfatal suicidal behavior, while addressing pain, mental health, and social isolation, may help save lives and improve the overall quality of cancer care.