Skip to Main Content


Skip Nav Destination

Examining Hospice Use Among Transfusion-Dependent Patients With Leukemia

December 30, 2021

Since 2001, more patients with leukemia are receiving hospice services, which has been associated with improved quality of life, according to an analysis published in Blood. However, transfusion dependence is a substantial barrier to patients receiving these services, as many hospice organizations disallow life-extending transfusion support. This translates to less time in hospice care and potentially higher risk of inpatient death and higher Medicare spending, lead author Thomas W. LeBlanc, MD, and researchers reported.

"Leukemia clinicians have long expressed frustrations about hospice-care programs not providing transfusion support to their patients at end of life; however, there wasn't actually a study robustly demonstrating any association between transfusion dependence and use of hospice-care services," Dr. LeBlanc, from Duke University School of Medicine in Durham, North Carolina, told ASH Clinical News. "These findings, which demonstrate an association between transfusion dependence and use of hospice-care services, highlight a need for policy solutions allowing for transfusion support among patients with leukemia who are enrolled in hospice."

The researchers analyzed the Surveillance, Epidemiology, and End Results (SEER)-Medicare database to identify 21,033 Medicare beneficiaries (median age = 79 years; range = 73-84 years) who were diagnosed with acute and chronic leukemias and died between 2001 and 2011.

All patients were 66 years or older at the time of diagnosis and had been continuously enrolled in Medicare Parts A and B for at least one year prior to diagnosis. Patients who died within 30 days of diagnosis were excluded from the analyses.

Investigators then examined the association between transfusion dependence (defined as having ≥2 transfusion events ≥5 days apart within a 30-day period prior to death or hospice admission) and two primary endpoints: use of hospice services at time of death and duration of hospice length of stay.

Secondary endpoints included:

  • death in the inpatient setting
  • admission to an intensive care unit (ICU) within 30 days of death
  • receipt of chemotherapy within 14 days of death
  • hospice enrollment <3 days prior to death
  • outpatient referral to hospice (defined as hospice admission >2 days after discharge from any preceding hospitalization)
  • Medicare spending within a 30-day period prior to death

A total of 4,141 patients (20%) were transfusion dependent before death or hospice enrollment. Nearly half of the participants (n=9,230; 44%) were receiving hospice care at end of life. The authors noted that this proportion increased significantly during the study period, from 35 percent to 49 percent (p<0.001).

Surprisingly, hospice enrollment was higher among patients who were transfusion dependent than those who were not (48% and 43%; relative risk [RR]= 1.08; 95% CI 1.04-1.12; p<0.001).

Among patients who received end-of-life hospice care, the researchers observed a significant decrease in inpatient deaths and chemotherapy use (p<0.001), but no change in the proportion of hospice stays lasting more than three days (p<0.11).

The median duration of hospice stay, regardless of transfusion-dependent status, was nine days (range = 3-28 days). However, patients who were transfusion-dependent had a 51-percent shorter hospice stay than those who did not receive transfusions (6 vs. 11 days, respectively; p<0.001).

Multivariable analyses revealed that, compared with patients without transfusion dependence, those with transfusion dependence appeared to be more likely to:

  • receive hospice services for <3 days (RR = 1.37; 95% CI 1.25-1.51; p value not reported)
  • die in the inpatient setting (RR=1.04; 95% CI 1.00-1.08; p value not reported)
  • be admitted to the ICU prior to death (RR=1.05; 95% CI 1.00-1.10; p value not reported)

However, transfusion-dependent patients appeared to be less likely to receive outpatient hospice referral (RR=0.89; 95% CI 0.84-0.94; p value not reported).

The authors concluded that transfusion dependence represents "a barrier to timely hospice referral" but added that the associations observed in their study may not be causal. Reviewing transfusion claims to the date of death, rather than the date of hospice enrollment, may have introduced a "guarantee-time bias," which may explain the higher-than-expected enrollment among transfusion-dependent patients.

Also, because the analysis included only Medicare beneficiaries with leukemias, the findings may not be generalizable to patients enrolled in managed-care plans or with other cancers or to younger patients.

Future steps, Dr. LeBlanc said, could include pilot-testing an open-access program that would allow patients with leukemia to enroll in hospice care while also receiving transfusion support for palliative purposes.

"This will require some funding and also partnership with payers, such as the Centers for Medicare and Medicaid Services," he commented, "but this change could have a dramatic impact on the well-being of patients and families who are struggling with a terminal leukemia diagnosis."

Dr. LeBlanc added that further research is necessary to quantify the real-world advantages and optimal delivery of palliative transfusion support for patients with leukemia at the end of life. "Clinicians in both the hematology and palliative-care communities have raised questions about the appropriateness of palliative transfusion support, asking what it really accomplishes for patients and families, and wondering whether it just prolongs the dying process," he concluded. "While many of us have seen what we think to be substantial benefits from palliative-intent transfusions in our practices, there is a need for larger, high-quality studies to answer these important questions more definitively."

The authors reported no conflicts of interest.


LeBlanc TW, Egan PC, Olszewski AJ. Transfusion dependence, use of hospice services, and quality of end-of-life care in leukemia. Blood. 2018;132:717-26.


Connect with us:

August 2022


Close Modal

or Create an Account

Close Modal
Close Modal