Palliative Care

PD 02 - Palliative 1 - Poster Discussion

1017 - Association Between Accuracy of Survival Predictions and Quality of End of Life Care Among Metastatic Cancer Patients Receiving Radiation Therapy

Sunday, October 21
2:03 PM - 2:09 PM
Location: Room 217 C/D

Association Between Accuracy of Survival Predictions and Quality of End of Life Care Among Metastatic Cancer Patients Receiving Radiation Therapy
K. Sborov, S. Giaretta, A. Koong, S. Aggarwal, R. Von Eyben, D. T. Chang, M. F. Gensheimer, and E. Pollom; Department of Radiation Oncology, Stanford Cancer Institute, Stanford, CA

Purpose/Objective(s): Intense end of life cancer care may be influenced by poor prognostication by oncologists. We examined the association between intense end of life care and radiation oncologists’ ability to predict survival among patients with metastatic cancer.

Materials/Methods: We used data from a prospective study of ability of radiation oncologists to predict survival of metastatic cancer patients receiving palliative radiation in our department. We identified patients who died within 12 months (mos) of study enrollment and excluded those who died within 1 mo. Patients could be enrolled on the study multiple times if they had multiple radiation courses. Intense end of life care was defined by National Quality Forum (NQF) and ASCO Quality Oncology Practice Initiative (QOPI) metrics: receipt of chemotherapy in the last 14 days of life, intensive care unit stay, 2+ emergency room visits, 2+ hospital admissions, or use of advanced radiation technique in the last 30 days of life, in-hospital death, and referral to hospice 3 days before death. Survival predictions were categorized as follows: correct (survival predicted <12 mos), 12-18 mos, 18-24 mos, and >24 mos. The association between survival prediction category and intense end of life care was assessed using repeated measures logistic regression to account for within patient correlation, as each patient could be in the dataset for multiple encounters and oncologist predictions.

Results: We identified 375 patients with 468 encounters with survival estimates. Overall, 130 (27.8%) encounters were associated with at least 1 NQF/QOPI metric of intense end of life care. Oncologists accurately predicted survival (<12 mos) in 363 (77.7%) encounters, while 54 (11.6%) encounters were predicted to live 12-18 mos, 27 (5.8%) encounters were predicted to live 18-24 mos, and 23 (4.9%) encounters were predicted to live >24 mos. Correct survival prediction was associated with poor performance status (85% correct among those ECOG 2-4 versus 72% among ECOG<2, p= 0.001)and primary cancer site (82% correct among lung primaries, 68% breast, 65% prostate/genitourinary, 83% GI, 70% head and neck/skin, 84% other, p=0.02), and was not associated with age, race, marital status, insurance status, or presence of brain metastases. Compared to encounters with correct survival predictions, those who were predicted to live >24 mos were more likely to meet at least 1 NQF/QOPI metric of intense end of life care (OR 2.55, p=0.03). Those who were predicted to live 12-18 mos and 18-24 mos were not significantly more likely than those with accurate survival predictions to meet at least 1 metric of intense end of life care (OR 0.97, p=0.94 and OR 1.39, p=0.45).

Conclusion: Oncologists’ inaccurate survival predictions are associated with more intense end of life care for patients with advanced cancer. Additional tools to aid with survival prediction may improve quality of end of life cancer care.

Author Disclosure: K. Sborov: None. S. Giaretta: None. A. Koong: None. R. Von Eyben: None. D.T. Chang: Research Grant; Varian Medical Systems. Stock; ViewRay. M.F. Gensheimer: Research Grant; Varian Medical Systems, Philips Healthcare.

Kate Sborov, BS

Disclosure:
Employment
Stanford University Cancer Center: Assistant Clinical Research Coordinator: Employee

Presentation(s):

Send Email for Kate Sborov


Assets

1017 - Association Between Accuracy of Survival Predictions and Quality of End of Life Care Among Metastatic Cancer Patients Receiving Radiation Therapy



Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Association Between Accuracy of Survival Predictions and Quality of End of Life Care Among Metastatic Cancer Patients Receiving Radiation Therapy