Most hematologic oncologists accept standard quality end-of-life measures
A nationwide cohort of hematologic oncologists deemed a majority of standard oncology end-of-life measures appropriate for patients with blood cancers, according to study results published in Journal of Clinical Oncology.
Unrealistic patient expectations served as the primary barrier to quality end-of-life care, results showed.
Oreofe O. Odejide
Prior studies have shown that patients with hematologic malignancies have received suboptimal end-of-life care when assessed with standard oncology quality measures.
“Unlike many solid malignancies — where advanced disease is often incurable — many advanced hematologic cancers remain potentially curable,” Oreofe O. Odejide, MD, MPH, instructor at Harvard Medical School and Dana-Farber Cancer Institute, told HemOnc Today. “This lack of a clear distinction between the curative and end-of-life phase of disease may serve to delay the transition to appropriate end-of-life care, thus impacting quality of care.”
Odejide and colleagues sought to determine whether oncology quality measures were appropriate for patients with blood cancers, and whether these patients experienced specific barriers to quality end-of-life care.
The researchers sent surveys to 667 U.S. hematologic oncologists identified through the 2015 ASH directory. Of these, 57.3% (n = 349; median age, 52 years; 75.6% men) sent completed responses. The majority of respondents were board certified in medical oncology (86.5%), hematology (80.8%) or both (71%).
Respondents had a mean number of 10 (IQR, 8-20) deceased patients in the year preceding survey completion.
The researchers asked participants to deem eight standard end-of-life quality measures “acceptable” or “not acceptable” indicators of good quality end-of-life care for patients with blood cancers. These included:
- hospice admission more than 7 days before death;
- no chemotherapy within 14 days of death;
- no intubation within the last 30 days of life;
- no CPR in the last 30 days of life;
- no more than one hospitalization within 30 days of death;
- no more than one ED visit within 30 days of death;
- no ICU admissions within the last 30 days of life; and
- death outside of an acute facility.
The survey also included two new, hematology-specific quality measures: no red cell transfusions within 7 days of death, and no platelet transfusions within 7 days of death.
The researchers determined that any quality measure that received positive agreement from at least 55% of respondents would be deemed an acceptable indicator of good quality end-of-life care.
The only domains to fall below the researchers’ prespecified threshold were no more than one hospitalization (54.2%) and no more than one ED visit (53.6%) in the last 30 days of life.
Four domains yielded over 75% acceptance, including hospice initiation (77.9%), no chemotherapy within the last 14 days of life (79.9%), no intubation in the last month of life (80.5%) and no CPR in the last 30 days of life (85.1%).
The newly added hematology-specific quality measures had the lowest acceptance rates above 55% (red cell transfusion, 58.7%; platelet transfusion, 59.9%).
“These findings are reassuring, as they indicate that current measures are acceptable tools to assess the quality of end-of-life care for patients with blood cancers and to identify areas where interventions to improve care should be developed,” Odejide said.
A greater proportion of respondents whose practices served at least 25% of patients with solid tumors found the chemotherapy quality measure acceptable than respondents with fewer patients with solid tumors (85% vs. 74.6%; P = .02).
More hematologic oncologists who graduated from medical school within the last 15 years appeared favored the CPR measure (93.2% vs. 83.2%; P = .03) and the hospitalization measure (68.9% vs. 50.2%; P = .004).
The survey also asked respondents to identify potential barriers to the delivery of quality end-of-life care.
The most commonly identified barriers to effective end-of-life care included unrealistic patient expectations (97.3%), clinician fear of taking away hope (71.3%) and unrealistic clinician expectations (59%).
Other barriers identified by more than half of respondents included clinician uncertainty about prognosis (58.7%) and lack of clinician time (53.3%).
Respondents identified increasing access to palliative care (93.7%) and hospice facilities (92.2%), as well as policies allowing hospice enrollment during disease-directed treatment (90.1%), as interventions that would improve end-of-life care.
The researchers acknowledged the response rate was fairly low, as well as the potential for nonresponse or participation bias. The researchers further conjectured that respondents may have felt pressured to report finding commonly accepted oncology end-of-life quality measures as acceptable.
“Quality end-of-life care as defined above, by necessity, must be preceded by clear conversations between physicians and their patients regarding prognosis and preferences with respect to end-of-life care,” Odejide said. “Prognosis can be fluid. As it often shifts between diagnosis and death for patients with blood cancers, readdressing prognosis as the course of disease changes is likely to foster more realistic expectations.
“Moreover, timely and high-quality end-of-life care discussions predicated on clear prognostic disclosure between patient and physicians should increase the likelihood that patients receive quality care that is in line with their preferences,” she added. – by Cameron Kelsall
For more information:
Oreofe O. Odejide, MD, MPH, can be reached at email@example.com.
Disclosure: Odejide reports no relevant financial disclosures. One research reports a consultant/advisory role with UnitedHealthcare and royalties on patents licensed to UpToDate. An additional researcher reports stock ownership in Recap Information System.