Hospice Enrollment in Patients With Advanced Heart Failure Decreases Acute Medical Service Utilization

Abstract

Background: Patients with advanced heart failure (HF) enroll in hospice at low rates, and data on their acute medical service utilization after hospice enrollment is limited.

Methods and results: We performed a descriptive analysis of Medicare fee-for-service beneficiaries, with at least one home health claim between July 1, 2009, and June 30, 2010, and at least 2 HF hospitalizations between July 1, 2009, and December 31, 2009, who subsequently enrolled in hospice between July 1, 2009, and December 31, 2009. We estimated panel-negative binomial models on a subset of beneficiaries to compare their acute medical service utilization before and after enrollment. Our sample size included 5073 beneficiaries: 55% were female, 45% were ≥85 years of age, 13% were non-white, and the mean comorbidity count was 2.38 (standard deviation 1.22). The median number of days between the second HF hospital discharge and hospice enrollment was 45. The median number of days enrolled in hospice was 15, and 39% of the beneficiaries died within 7 days of enrollment. During the study period, 11% of the beneficiaries disenrolled from hospice at least once. The adjusted mean number of hospital, intensive care unit, and emergency room admissions decreased from 2.56, 0.87, and 1.17 before hospice enrollment to 0.53, 0.19, and 0.76 after hospice enrollment.

Conclusions: Home health care Medicare beneficiaries with advanced HF who enrolled in hospice had lower acute medical service utilization after their enrollment. Their pattern of hospice use suggests that earlier referral and improved retention may benefit this population. Further research is necessary to understand hospice referral and palliative care needs of advanced HF patients.

Keywords: Medicare; adult; caregiver; heart failure; hospice.

Figure 1

Selection of Sample (N= 5,073).

Figure 2

Percentage of beneficiaries (N = 5,073) who enrolled in hospice by time interval (days = 0–7, 8–30, 31–60, 61–90, 91–180, 181–270, 271–365 from 2^nd hospitalization discharge).

Figure 3

Percentage of beneficiaries (N = 4,619) who died by time interval (days = 0–7, 8–30, 31–60, 61–90, 91–180, 181–270, 271–365 from hospice enrollment).

Figure 4

Kaplan-Meier survival curve of beneficiaries from hospice enrollment to death or one year.