Since 2000, Medicare beneficiaries have become less likely to die in hospitals, and more likely to die in their homes or in community health care facilities.
A review of Medicare records also determined that there was a decline in health care transitions in the final 3 days of life for these patients, Joan M. Teno, MD, and her colleagues wrote in JAMA.
It is not possible to identify a specific reason for the shift, wrote Dr. Teno, professor of medicine at the Oregon Health & Science University, Portland. Between the study years of 2000 and 2015, there were several large efforts to improve care at the end of life.
“Since 2009, programs ranging from ensuring informed patient decision making to enhanced care coordination have had the goal of improving care at the end of life. Specific interventions have included promoting conversations about the goals of care, continued growth of hospice services and palliative care, and the debate and passage of the Affordable Care Act … It is difficult to disentangle efforts such as public education, promotion of advance directives through the Patient Self- Determination Act, increased access to hospice and palliative care services, financial incentives of payment policies, and other secular changes.”
The study mined data from the Centers for Medicare & Medicaid Services, and examined end-of-life outcomes among two Medicare groups: Medicare fee-for-service recipients (1,361,870) during 2009-2015, and Medicare Advantage recipients (871,845), comparing 2011 and 2015. The mean age of both cohorts was 82 years.
Outcomes included site of death and “potentially burdensome transitions,” during the last days of life. These were defined as three or more hospitalizations in the previous 3 months, or two or more hospitalizations for pneumonia, urinary tract infection, dehydration, or sepsis during the last 120 days of life. Prolonged mechanical ventilation also was deemed potentially burdensome.
Among fee-for-service recipients, deaths in acute care hospitals declined from 32.6% to19.8%. Deaths in nursing homes remained steady, at 27.2% and 24.9%. Many of these deaths (42.9%) were preceded by a stay in an ICU. There was a transient increase in end-of-life ICU use, around 2009, but by 2015, the percentage was down to 29%, compared with 65.2% in 2000.
Transitions between a nursing home and hospital in the last 90 days of life were 0.49/person in 2000 and 0.33/person in 2015. Hospitalizations for infection or dehydration fell from 14.6% to12.2%. Hospitalization with prolonged ventilation fell from 3.1% to 2.5%.
Dying in hospice care increased from 21.6% to 50.4%, and people were taking advantage of hospice services longer: the proportion using short-term services (3 days or less) fell from 9.8% to 7.7%.
Among Medicare Advantage recipients, the numbers were somewhat different. More than 50% of recipients entered hospice care in both 2011 and 2015; in both years, 8% had services for more than 3 days. About 27% had ICU care in the last days of life, in both years. Compared to fee-for-service recipients, fewer Medicare Advantage patients were in nursing homes at the time of death, and that number declined from 2011 to 2015 (37.7% to 33.2%).
In each year, about 10% of these patients had a hospitalization for dehydration or infection, and 3% had a stay requiring prolonged mechanical ventilation in each year. The mean number of health care transitions remained steady, at 0.23 and 0.21 per person each year.
Dr. Teno had no financial disclosures.
SOURCE: Teno JM et al. JAMA. 2018 Jun 25. doi: 10.1001/jama.2018.8981.