Use of hospice services among Medicare beneficiaries with acute and chronic leukemias.

Abstract

68 Background: Studies show lower rates of hospice use among patients (pts) with hematologic malignancies. Our objective was to describe trends in hospice use and in quality measures for end-of-life (EOL) care among Medicare beneficiaries with leukemias. Methods: From the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database, we selected pts with acute (myeloid or lymphoblastic) or chronic (lymphocytic, myeloid, or myelomonocytic) leukemias diagnosed in 1996-2011 who died in 2001-2011. We identified hospice enrollment at death, duration of hospice enrollment, inpatient deaths, intensive care unit (ICU) admissions within 30 days of death, and chemotherapy use in the last 14 days of life. We summarized linearized trends by year of death using log-binomial regression, reporting average annual percent change (APC). Results: Among 38,038 leukemia pts (41% acute, median age 78 years) 81% died, after median 2.8 months from diagnosis (95%CI, 2.7-2.9) for acute and 49.1 months (95%CI, 47.9-50.1) for chronic leukemias. Among pts who died in 2001-2011 ( N= 23,941), 42% were enrolled in hospice at the time of death. This proportion significantly increased between 2001 and 2011, from 33% to 48% (APC +4.1%, P< .001), both for acute and chronic leukemias ( Pinteraction= .25). Median time on hospice was 8 days, and the proportion of pts with < 3 days on hospice increased from 20% to 24% between 2001 and 2001 (APC +1.2%, P= .05). Inpatient deaths significantly decreased from 54% to 39%, respectively (APC -3.2%, P< .001), but ICU use at EOL increased from 34% to 41% (APC +2.4%, P< .001). Chemotherapy use at EOL was more frequent in acute (17%) than chronic (6%) leukemia, and decreased for both (overall from 15% to 10%, APC -3.2%, P< .001). Conclusions: The use of hospice services among older pts with leukemia has increased, suggesting its wider acceptance over time. However, the increasing proportion of brief, terminal hospice admissions, and increasing rate of ICU use at EOL reflect persistent barriers to early enrollment in this population. Some measures of aggressiveness of care (inpatient deaths, chemotherapy at EOL) are lower in the community than previously reported from academic centers (El-Jawahri, Cancer 2015).