ESTRO meets Asia 2024 - Abstract Book

S223

Interdisciplinary – Mixed sites/palliation

ESTRO meets Asia 2024

ESAS-r and EQ-5D-5L were recorded again on the last day of RT, and 1 week later. Basic demographic/socioeconomic, clinical and treatment data were collected (via REDCap). Interviews were transcribed verbatim, and content analysis performed. Descriptive statistics were generated.

Results:

Although ethics-approved in 2019, pandemic-related factors delayed study launch to 01/2021, with accrual closing (after simplified protocol) in 12/2021. Among 74 patients who agreed to learn more about the study, 39 enrolled (21 randomized to intervention-arm); only 34 completed the entire protocol. Provider-participants included 14 (of 18 affiliated) radiation oncologists (completing 38.5%, 37/96 surveys) and other providers (radiation therapists, 30.2%; pharmacist, 12.5%; registered nurse, 10.4%; trainee/other, 8.3%): 73.7% were female; most commonly (34.0%) with 11-15 years’ experience; spending most often (49.5%) 15-30 minutes face-to-face with the patient, deploying various communication strategies (verbal-only in 64.5% interactions) with no between-arm differences. Most patient-participants (56.4%, n=22) were male, aged 65-74 years (46.2%), and of European (43.2%) or non Indigenous North American ethnicity (37.8%), with median ECOG of 1 (for n=24 recorded) and median Charlson Comorbidity Index, 9 [2-14] (n=38). For 27.0% and 78.4%, this was their first visit to the cancer centre and to the Radiotherapy department, respectively. Almost all understood/spoke English (97.4%), 76.9% were accompanied (over half received help with questionnaires), 34.2% had difficulty hearing. Most (78.9%) had at least high-school education; 44.1% reported income at/below the poverty line. CHLT-6 scores were < 5/6 (“limited”) in only 15.8%, Brief-COPE patterns varied; no significant difference between study arms. Patients self-rated better understanding when providers used 4+ communication techniques. Qualitative analysis showed no between-arm differences in patient understanding of diagnosis, RT site (43.6% spine/pelvis) and purpose, side effects, or medication instructions. But nearly half understood only partially, or not at all, the RT side effects. Agreement between patient and provider Likert ratings of patient understanding was 0.45. ESAS-r scores were not statistically different between arms, but differential patterns in anxiety/pain/well-being over time (at consult, RT-end, post RT) need exploration (likewise for EQ-5D-5L). Most (88.2%) read some part of the printed materials (“easy” for 76.7%) given on RT and/or basic symptom management.

Conclusion:

HL is not assessed standardly in the care of patients with advanced cancer, including in Palliative Radiation Oncology. Other factors may mediate or moderate how HL considerations are recognized and addressed in clinical encounters. Overcoming challenges in studying HL in this population will be key to improving patient care and outcomes.

Keywords: health literacy, palliative radiotherapy

338

Digital Poster

Machine learning based mortality prediction model to assist decision-making in end-of-life care

Mi Sun Kim 1 , Jae ha Kim 2 , Tae hoon Kong 2,3

1 Department of Radiation Oncology, Yonsei University Wonju College of Medicine, Wonju, Korea, Republic of. 2 Department of Medical Informatics and Biostatistics, Yonsei University Wonju College of Medicine, Wonju, Korea,