Oncology Rounds

Room for improvement: End of life care for cancer patients poor across all hospital types

Anne Taylor

New analyses from researchers at the Dartmouth Institute for Health Policy and Clinical Practice reveal that quality of care at the end of life remains poor, yet highly variable. We recently attended the Health Affairs Issues in Cancer Care conference in Washington, DC, where lead author Nancy E. Morden, Assistant Professor of Community and Family Medicine at Dartmouth Medical School, reported the findings of this research.

Measuring quality of care at the end of life

To assess the intensity of end-of-life care for poor-prognosis cancer patients, the authors examined several NQF-endorsed end-of-life quality measures for terminal cancer patients. The study cohort included 215,000 Medicare patients, cared for in over 4,400 hospitals, with advanced cancer at time of death. End-of-life care measures included: 

  • Hospitalization: Death in the hospital, hospitalized in the last month of life, days in the hospital in the last month of life
  • Hospice: Hospice use in the last month of life, days in hospice in the last month of life, hospice initiated in the last three days of life
  • ICU Use: ICU use in the last month of life, days in ICU in the last month of life
  • Other: Chemotherapy in the last 14 days of life, potentially life-prolonging procedure in the last month of life (including intubation, resuscitation, or other intense care), saw 10 or more physicians in last six months of life

Overall, end-of-life care quality is poor

Patients received care in a variety of settings, including 4,240 community hospitals, 161 academic medical centers, 21 NCCN cancer centers, and 22 non-NCCN NCI-designated hospitals (NCI-designated cancer centers that do not belong to the NCCN). 

The authors found that cancer patients received high levels of inpatient care in the last month of life, regardless of hospital type, size, and for-profit status. The results were discouraging:

  • 30% of patients died in the hospital
  • 65% of patients spent some time of the last month of life in the hospital
  • Only 54% received some hospice care in last month of life; 9% entered hospice within three days of death, presumably too late to receive benefits of hospice care

Hospital type did not predict quality of care

The authors expected to find that hospitals with a specific focus on cancer care, including members of the NCCN and those with NCI designation, would do better on the quality metrics important for poor-prognosis cancer patients. 

However, lead cancer centers did not fare better on NQF metrics. Instead, there was a broad range of intensity of care provided in all settings and no hospital characteristic could reliably predict a specific pattern of care. 

Using NCCN hospitals as a frame of reference, the authors observed relevant differences between hospital types in only four of the eleven measures. These include: 

  • Hospice initiation: Very late hospice initiation (hospice initiation within three days of death), was 13%, 19%, and 29% higher in NCI-designated, academic, and community hospitals, respectively, than in NCCN hospitals.
  • ICU care: Patients in NCI, academic, and community hospitals experienced more care in the ICU in the last month of life than in NCCN hospitals.
  • Chemotherapy: Patients cared for in the community setting were more likely to receive chemotherapy in the last 14 days of life than in NCCN, NCI, or academic hospitals.
  • Life-prolonging procedures: Potentially life-prolonging procedures in the last month of life were more common in NCI-designated and academic hospitals than in community and NCCN hospitals.

Best practices for end-of-life care may be found in any hospital setting

The authors comment that while there were some trends across hospital types and characteristics, these slight differences were overshadowed by variation within hospital groups (hospitals of similar type, size, and for-profit status). 

They acknowledge that end-of-life care decisions are some of the most complex decisions for providers and patients, and encourage all hospitals to reexamine their approach to end-of-life care to ensure accordance with patient preferences. 

Finally, the authors suggest that all hospitals, not only those with specialized cancer care, be included in the development of best practices in end-of-life care for terminal cancer patients. 

For more information, please see:

The 2012-2013 Oncology Roundtable National Meeting Series will explore strategies for improving and demonstrating high quality cancer care. To review the meeting topics and reserve a seat at your preferred location, register here.

Join the discussion

Please log in to comment.

Forgot your password?

Not an Advisory Board Member? Click here to register


Members please Log In


Forgot your password?

Not an Advisory Board Member? Click here to register