Hospice utilization in Veterans with newly diagnosed metastatic prostate cancer.

Abstract

11146 Background: Treating men with metastatic castration-sensitive prostate cancer (mCSPC) using androgen deprivation therapy (ADT), with or without treatment intensification, is highly palliative and leads to survival measured in years. In contrast, hospice provides end-of-life support when estimated life expectancy is ≤6 months, which rarely applies to men with newly diagnosed mCSPC. Although traditionally, patients transition to hospice when cancer treatments have been exhausted, the VA allows hospice enrollment during cancer treatment. This study investigates factors associated with hospice enrollment for Veterans within the first six months after diagnosis of mCSPC. Methods: We conducted a retrospective cohort study utilizing the VA Corporate Data Warehouse, an extensive database of medical records from 130 VA facilities, linked to Medicare data. Patients newly diagnosed with mCSPC between January 1, 2023, and June 17, 2024, were identified using a validated natural language processing tool. Veterans with prior orchiectomy, receipt of ADT the prior year, or without primary care visits the two years prior to diagnosis were excluded. Our primary outcome was hospice enrollment within six months of metastatic diagnosis. Multivariate logistic regression was used to estimate adjusted odds ratios (aOR) and 95% confidence intervals (CI), adjusting for clinical and socioeconomic variables, including neighborhood disadvantage as measured by area deprivation index (ADI). Results: We identified 3429 Veterans with mCSPC, with a mean age of 76 years, 30% of whom were Black. Overall, 11% (n = 391) of Veterans were referred to hospice within the first six months following mCSPC diagnosis. Older age and higher level of frailty were associated with greater odds of hospice enrollment within 6 months (aOR 1.84, 95% CI 1.59-2.14 per 10 years; aOR 4.79, 95% CI 3.74-6.13 for those with moderate/severe frailty versus not frail or mild frailty, respectively). Single Veterans (aOR 1.41, 95% CI 1.11-1.80) and those without additional healthcare insurance (aOR 2.22, 95% CI 1.74-2.85) also had greater odds of hospice referral. Neighborhood-level disadvantage was significantly associated with hospice enrollment: Veterans living in the most deprived neighborhoods (ADI 80–100) had higher odds of hospice referral compared to those in the least deprived neighborhoods (ADI 0–20) (aOR 2.97, 95% CI 1.90-4.73). Conclusions: Social determinants of health, including neighborhood-level disadvantage and lack of insurance coverage were strongly associated with Veteran hospice enrollment within six months of mCSPC diagnosis. However, since a unique benefit of VA cancer care allows hospice enrollment in concert with cancer treatment, more work needs to be done to understand whether hospice enrollment is being used primarily for social support or for end-of-life care in men with a highly treatable condition.