Specialty societies including the American Society of Clinical Oncology recommend a reduction in intensive medical services at the end of life for cancer patients, noting these services are at odds with a focus on quality of life. Yet over 25% of annual healthcare expenditures occur for persons in their last year of life, and approximately one-fifth of patients die in hospital, suggesting that some end-of-life care may be low-value. Despite its importance, the value of end-of-life care has never been evaluated.
Our objective is to evaluate the (patient-centered) value of intensive end-of-life services provided to Veterans dying from cancer (value is defined as cost relative to health outcomes). We first evaluate quality of end-of-life care and whether any differences exist for Veterans receiving care through VA versus those receiving care through Medicare (Aim 1), using well-accepted quality-of-care metrics. We then evaluate cost trajectories in the last year of life to ascertain differences between Veterans treated by VA versus those treated by Medicare (Aim 2.1). For a subset of Veterans who received care at VA, we look at the costs of care in the last month of life (Aim 2.2). We link these data to Bereaved Family Survey outcomes to investigate the value of intensive services provided at the end of life (Aim 3)..
In Aim 1, we evaluated the proportion of patients who experienced: 2+ Emergency Department (ED) visits; chemotherapy; ICU stay; hospital admission; death in hospital; and number of days spent in hospital. Poor-quality care is indicated by higher proportions of patients receiving these services. As Veterans could have received care through VA, Medicare, or both, they were allocated to the healthcare system on which they were the most reliant for outpatient care in the six months prior to the last thirty days of life (the time frame for which the outcome was measured).
Results from Aim 1 indicate Medicare-reliant Veterans were generally more likely to receive higher-intensity, lower-quality care at the end-of-life compared to VA-reliant Veterans. Medicare-reliant patients were significantly more likely to receive chemotherapy, be admitted to the hospital and ICU, spend more days in hospital, and die in hospital. Conversely, Medicare-reliant patients were significantly less likely to have two or more ED visits at the end-of-life.
In response to concerns over access to care at VA, congressional and VA leadership have recommended VA purchase more care in the community for veterans, transitioning to a role similar to that of Medicare. Aim 1 addresses whether a purchasing system with fee-for-service incentives (Medicare) can provide similar or better care than care provided by salaried clinicians (VA). Findings indicate that as VA purchases more care in the community, quality of care provided to Veterans at the end of life may suffer. Thus, VA will need to track patients' care and increase coordination with non-VA providers to ensure the best care for veterans at the end-of-life, regardless of where they are treated.
Findings from Aims 2 and 3 will inform VA policy makers about the value of intensive end-of-life services and whether VA is providing Veterans with end-of-life care that matches their desires.
External Links for this Project
Grant Number: I01HX001627-01A2
None at this time.