By Shira Klapper, Senior Writer/Editor

A new study in the Journal of Clinical Oncology (JCO) found that patients who were admitted to hospice in the last three days of life were significantly more likely to have a hematologic malignancy, to be male, married, and younger than age 65. The study, “Hospice Admission for Cancer Within the Last Three Days of Life: Independent Predictors and Implications for Quality Measures,” was published online, ahead of print, August 25. According to Cancer.Net, “Hospice care is for people with cancer that cannot be cured who are expected to live six months or less. It is high-quality, compassionate care that focuses on relieving symptoms and helping patients and families cope with death and dying.”

The article comes at a time when organizations such as the American Society of Clinical Oncology (ASCO) and the National Quality Forum (NQF) have stated that end-of-life-care should be measured as part of quality assessment programs. One way of measuring end-of-life-care is to determine the proportion of patients, in any given practice or hospital, who are enrolled in hospice in the last three days of life (ASCO’s Quality Oncology Practice Initiative [QOPI®] currently includes this proportion as one of its measure. Click here for an article in JCO explaining the rational for end-of-life quality measures in general). The ASCO and NQF initiatives are based on data showing that many patients are not admitted to hospice until nearly the very end of their lives; studies have shown that the median length of stay in hospice is only 18.7 days, that a third of patients are enrolled in the last week of life, and that 14.3% of patients with cancer enroll in hospice in the last three days of life.

Rating doctors by the timeliness of their patient’s admission to hospice might be a good idea, but there are limitations to applying the measure to every practice. After all, certain characteristics—such as race, gender, insurance type, and cancer type—are outside of the doctors’ influence and may have a large influence on whether a patient enrolls in hospice.

Nina R. O’Connor, MD

Which is where the JCO study comes in to play. Nina R. O’Connor, MD, first author of the study and Director of the Palliative Care Program at the Hospital of the University of Pennsylvania, shed more light on the motivation behind the study.

“ASCO and the NQF have endorsed the idea of using hospice admissions as a quality indicator, and so, increasingly, oncologists are going to be getting scores and measurements of their timing of hospice referrals,” said Dr. O’Connor. “But that doesn’t take into consideration that every oncologist’s practice is very different. For example, an oncologist who mostly sees hematological malignancies would have more late referrals, whereas an oncologist who treats an older population might have more early referrals.”

At risk: Younger, male, and married

To tease out the patient characteristics associated with admission to hospice in the last three days of life, the study analyzed data from the Coalition of Hospices Organized to Investigate Comparative Effectiveness (CHOICE), a database comprised of the electronic health records of 12 hospices. The researchers extracted data on the following variables: age, gender, race, marital status, payer type, time in hospice until death, and patient’s pre-hospice site, such as home, long-term care facility, hospice, or inpatient.

Upon analysis, the researchers observed that patients with hematologic cancers were more likely to be admitted to hospice in the last three days of life. This was not a surprising finding since previous research had reported similar results. What was surprising was the finding that lower rates of late admissions were seen among patients with Medicaid as well as patients who did not have formal insurance—two groups who often do not receive optimal care.

In explaining these findings, Dr. O’Connor stated that patients with Medicaid and those without formal insurance might face other challenges that lead them to enter hospice earlier.

“It might be harder for these patients to shoulder the out-of-pocket expenses of care, or they might be less likely to pursue experimental treatments,” said Dr. O’Connor. “Or, sometimes those patients also have fewer formal caregiving networks, so they might not have many people who can take care of them later on and they therefore enroll in hospice earlier.”

Another interesting finding within the study was that being married was associated with enrolling in hospice within the last three days of life. The study authors theorized that married patients might have the emotional support to continue aggressive treatment toward the end of life, or that they might continue treatment for the sake of their spouses.

Informing patients about their risk of late-hospice admission

While the study provides important information for organizations that are designing quality assessment programs, it also provides an immediate take-to-the-clinic message for doctors.

“This study gives clinicians some ideas about patients to be especially aware in terms of having a hospice conversation,” said Dr. O’Connor. “For patients who fit the categories in this paper, consider earlier conversations or increased conversation efforts about the potential benefits of hospice.”

“I think for clinicians just to have these particular characteristics in mind when they’re seeing patients with advanced cancer might help them be more proactive about hospice referrals.”

Nina R. O’Connor, MD, is a palliative care doctor, Assistant Professor of Clinical Medicine, and Director of the Palliative Care Program at the Hospital of the University of Pennsylvania, in Philadelphia.

 Source

Click here to read the abstract.

Click here to read the PDF.

O’Connor NR, Hu R, Harris PS, et al. Hospice admission for cancer within the last three days of life: independent predictors and implications for quality measures. J Clin Oncol. Epub 2014 Aug 25.

The Exclusive Coverage series on ASCO.org highlights selected research from JCO and JOP with additional perspective provided by the lead or corresponding author.

@ 2014 American Society of Clinical Oncology