Highlights
- Veterans with cancer exhibit an overall Suicidal Self-Directed Violence (SSDV) rate of 203 per 100,000 person-years, significantly higher than general population benchmarks.
- Temporal risk peaks within the first 6 months post-diagnosis but remains elevated for more than 5 years in specific cohorts, including younger and unmarried veterans.
- Central Nervous System (CNS), head and neck, and pancreatic cancers carry the highest hazard ratios for SSDV.
- Nonfatal attempts are disproportionately frequent in female veterans and those with thyroid or CNS malignancies, necessitating a shift from mortality tracking to comprehensive behavioral surveillance.
Background
Patients diagnosed with cancer face psychological burdens that transcend the physiological challenges of the disease. Among military veterans, these burdens are compounded by service-related stressors, higher baseline rates of mental health conditions, and unique sociodemographic vulnerabilities. Recent large-scale epidemiological data indicates that cancer-related distress contributes to significantly higher rates of suicidal self-directed violence (SSDV)—a composite measure including both fatal and nonfatal suicide attempts. Despite advancements in oncology, suicide remains a preventable cause of mortality in this population. Understanding the longitudinal trajectory of this risk is paramount for developing integrated oncology-psychiatry screening protocols and targeted prevention strategies within the Veterans Health Administration (VHA).
Key Content
Temporal Development of Risk Evidence
Clinical understanding of suicide risk in cancer has evolved from static cross-sectional observations to longitudinal cohort analyses. A landmark national cohort study (2014–2023) of 292,271 veterans identified 2,400 SSDV events, revealing that risk is not uniformly distributed over time (PMID: 42207508). While the immediate post-diagnostic period (first 3–6 months) represents a critical window of vulnerability with an 86% higher hazard rate compared to non-cancer cohorts (PMID: 40623636), the risk remains elevated deep into survivorship. Specifically, for younger veterans (≤45 years) and those with advanced disease, the hazard for SSDV remains significantly higher (aHR 1.58 and 1.30, respectively) even five years after the initial diagnosis (PMID: 42207508).
High-Risk Cancer Subtypes and Stages
Evidence suggests that site-specific cancers dictate varying levels of psychological distress and SSDV risk:
- CNS and Head/Neck Cancers: Veterans with CNS malignancies face the highest hazards (aHR 2.07), likely due to neurocognitive changes and the profound impact on physical identity. Head and neck cancers follow closely (aHR 1.67), associated with significant functional impairment in speech and swallowing (PMID: 42207508, 36029038).
- Esophageal and Pancreatic Cancers: Esophageal cancer has been associated with a staggering 6.01-fold increase in suicide risk in early periods, while pancreatic cancer shows consistently high probabilities across multiple studies (PMID: 36029038, 40623636).
- Thyroid and Testicular Cancers: Although often considered to have better prognoses, thyroid cancer patients show high rates of nonfatal attempts (359 per 100,000 person-years). Similarly, testicular cancer survivors exhibit a massive hazard ratio for suicidality (HR 22.99) compared to controls, with chemotherapy further increasing psychosocial morbidity (PMID: 41668134).
Advanced disease stages (Stage 3 and 4) at diagnosis are positively associated with suicide risk, with aHRs ranging from 2.36 to 3.53 (PMID: 36029038).
Sociodemographic and Clinical Risk Determinants
Beyond the cancer diagnosis itself, several comorbid factors amplify SSDV risk:
- Frailty and Pain: Severe frailty is one of the strongest predictors, with SSDV rates reaching 544 per 100,000 person-years. High pain scores are also significantly associated with elevated rates (236 per 100,000) (PMID: 42207508).
- Mental Health Comorbidities: Chronic mental illness (aHR 419 per 100,000) and specific multimorbidity phenotypes—such as anxiety, PTSD, and substance use disorders—significantly heighten risk. In older veterans with schizophrenia, those in the “high anxiety/PTSD” class showed the highest risk of suicide death (HR 1.37) (PMID: 40610254).
- Demographics: Younger age (≤45 years), female sex, Asian race, and being unmarried or divorced are identified as independent risk factors for increased SSDV hazards (PMID: 42207508, 41668134).
Methods of Self-Directed Violence
Identifying the methods used is crucial for lethal means safety interventions. Longitudinal data indicates that poisoning (including opioid overdose) is the most common method for all SSDV events (26%), particularly in nonfatal attempts (PMID: 42207508). However, when focusing on *fatal* suicides, firearms remain the dominant method (79.37%), especially among older survivors (89.9% in the 70-84 age group) (PMID: 40623636).
Technological and Precision Health Advances
To address these risks, the VA has leveraged the MVP-CHAMPION collaboration, utilizing artificial intelligence and high-performance computing to refine suicide risk screening (MVP 062). These tools aim to identify novel genetic and clinical variants that predict vulnerability, allowing for more discriminating models that can be applied in near-term clinical management (PMID: 42035288, 39507405).
Expert Commentary
The findings from the 2026 Sullivan et al. study and related VHA literature represent a paradigm shift in how we approach oncological survivorship. Historically, suicide prevention in cancer has focused on the terminal or advanced patient. However, the data reveals high nonfatal SSDV rates in survivors of “low-mortality” cancers like thyroid and testicular cancer. This suggests that the psychological impact of diagnosis, treatment side effects (e.g., erectile dysfunction, which correlates with 1.91 odds of suicidal ideation), and the fear of recurrence are potent drivers of self-harm (PMID: 35792506).
Clinicians must recognize that the highest risk window is the first six months, but a “one-and-done” screening approach at diagnosis is insufficient. The persistence of risk five years post-diagnosis among younger and unmarried veterans underscores the need for integration between oncology and mental health services. Furthermore, the role of clinical pharmacist practitioners in managing high-dose opioids is vital, as poisoning remains a primary method of SSDV attempts (PMID: 36044721).
Conclusion
SSDV among veterans with cancer is a complex, longitudinal challenge. While aggressive and advanced-stage cancers carry the highest mortality-related risk, psychological distress persists across all cancer types and stages, often manifesting as nonfatal attempts in younger and female cohorts. Future research and clinical practice must prioritize: 1) Universal, longitudinal screening that continues years into survivorship; 2) Targeted interventions for high-risk subtypes (CNS, pancreas, head and neck); and 3) Enhanced lethal means safety education focusing on both firearm storage and medication (opioid) management. The transition toward precision health and AI-driven screening offers a promising path toward proactive, individualized suicide prevention in the veteran oncological population.
References
- Sullivan DR, et al. Longitudinal Risk for Suicidal Self-Directed Violence Among Veterans With Cancer. JAMA Oncol. 2026. PMID: 42207508.
- Sullivan DR, et al. Suicide rates and methods among veteran cancer survivors within the Veterans Health Administration (VHA), 2010-2020. J Affect Disord. 2025. PMID: 40623636.
- Guerrettaz J, et al. Suicide risk following a new cancer diagnosis among Veterans in Veterans Health Administration care. Cancer Med. 2023. PMID: 36029038.
- Basourakos SP, et al. Anxiety, Depression, and Suicidality Among Testicular Cancer Survivors. Cancer Med. 2026. PMID: 41668134.
- Lynch J, et al. Increasing value in the Veterans Affairs Healthcare System (VA) with precision health: a continuing landmark collaboration with the Department of Energy. J Am Med Inform Assoc. 2026. PMID: 42035288.
- Thomas ML, et al. Erectile Dysfunction in a U.S. National Sample of Male Military Veterans. Mil Med. 2023. PMID: 35792506.