The effects of geriatric assessment on oncologist-patient communication regarding functional status and physical performance in older adults with cancer: A secondary analysis of a 541-subject nationwide URCC NCORP (NCI Community Oncology Research Program) cluster randomized trial.

Abstract

12010 Background: Despite high prevalence of functional status (FS) and physical performance (PP) impairments among older adults with cancer, standardized assessments and interventions are not routinely used in oncology care. This study characterized how oncologist knowledge of Geriatric Assessment (GA) results influenced conversations and GA-guided recommendations addressing FS and PP concerns. Methods: Data were from a NCORP funded (UG1CA189961) nationwide cluster randomized controlled trial (ClinicalTrials.gov: NCT02107443; PI: Mohile), with inclusion criteria: age ≥70, stage III/IV solid tumor or lymphoma with palliative treatment intent, and ≥1 GA domain impairment. All subjects underwent baseline GA including standardized FS ([instrumental] activities of daily living) and PP (Timed Up and Go, Short Physical Performance Battery, Older Americans Resources and Services Physical Health scale, falls in past 6 months) measures. Oncologists in Intervention arm practices received full GA results and validated recommendations for each patient, while those in the usual care (UC) arm were only notified of depression or severe cognitive impairment. One clinical encounter per patient within 4 weeks of GA was audio-recorded, transcribed and blind coded using a priori content-analysis scheme to categorize conversations and oncologist response (dismissed, acknowledged, or addressed with recommendation) by GA domain. Frequencies, raw and adjusted (for site using generalized linear mixed models) proportions were compared using the Chi square test. Results: 541 patients (mean age: 77, range 70-96) were included. More FS and PP conversations occurred in Intervention (PP=532, FS=164) than UC (PP=183, FS=87) arm (p<.0001). The adjusted proportion of all patients having one or more FS or PP conversations reached 85.8% in the Intervention arm but only 58.6% in UC (p<.0001). Intervention oncologists were more likely to address FS and PP concerns than UC oncologists (42.6% vs 16.5%, p=0.0003), and to use referrals (Intervention=23.5%, UC=5.0%, p<.0001) or information (Intervention=22.3%, UC=3.8%, p=0.0006) to address them. Conclusions: Providing oncologists a GA report with recommended interventions enhances oncologist-patient communication regarding FS and PP-related concerns in older adults with advanced cancer. FS and PP-related issues were more likely to be addressed by those oncologists receiving the GA report, demonstrating the utility of GA as a tool in creating tailored interventions for FS and PP concerns. Our findings support use of GA as an important tool in caring for patients with impairments in physical performance and function. Funding: NIH/NCI UG1CA189961, T32CA102618. Clinical trial information: NCT02107443.