A National Study of Live Discharges from Hospice

BACKGROUND: Live discharges from hospice can occur because patients decide to resume curative care, their condition improves, or hospices may inappropriately use live discharge to avoid costly hospitalizations. OBJECTIVE: Describe the variation, outcomes, and organizational characteristics associated with live discharges. DESIGN: Retrospective cohort study. SETTING/SUBJECTS: Medicare fee-for-service hospice beneficiaries. MEASUREMENT: Overall rate, timing, and health care transitions of live discharges. RESULTS: In 2010, 182, 172 of 1, 003, 958 (18. 2%) hospice discharges were alive. Hospice rate of live discharges varied by hospice program with interquartile range of 9. 5% to 26. 4% and by geographic regions with the lowest rate in Connecticut (12. 8%) and the highest in Mississippi (40. 5%). Approximately 1 in 4 (n=43, 889; 24. 1%) beneficiaries discharged alive were hospitalized within 30 days. Nearly 8% (n=13, 770) had a pattern of hospice discharge, hospitalization, and hospice readmission. These latter cases account for 126 million in Medicare reimbursement. Not-for-profit hospice programs had a lower rate of live discharges compared to for-profit programs (14. 6% versus 22. 4%; adjusted odds ratio AOR 0. 84, 95% confidence interval CI 0. 77-0. 91). More mature hospice programs (over 21 years in operation) had lower rates of live discharge compared to programs in operation for 5 years or less (14. 2% versus 26. 7%; AOR 0. 71, 95% CI 0. 65-0. 77). Small for-profits in operation 5 years or less had a higher live discharge rate than older, for-profit programs (31. 5% versus 14. 3%, p<0. 001). CONCLUSIONS: Approximately 1 in 5 hospice patients are discharged alive with variation by geographic regions and hospice programs. Not-for-profit hospices and older hospices have lower rates of live discharge.