A Lay Health Worker–Led Symptom Intervention and Acute Care Use in Older Adults With Cancer

Importance: Undertreated cancer symptoms are common among older adults, yet effective early identification and intervention remain limited. Objective: To determine whether lay health worker-led symptom assessment reduces acute care use among older adults with cancer. Design, Setting, and Participants: Multisite randomized clinical trial across 43 oncology clinics in California and Arizona conducted from November 2020 through October 2023 with 12 months of follow-up among Medicare Advantage beneficiaries aged 75 years or older with newly diagnosed, recurrent, or progressive cancer. Interventions: Participants were randomized 1:1 into a symptom assessment group (n = 200; usual care with lay health worker-led, telephone-based symptom assessments for 12 months using the Edmonton Symptom Assessment System) or a control group (n = 216; usual care alone), with planned enrollment of 200 in both groups. The lay health workers reviewed assessments with advanced practice practitioners who conducted interventions for symptoms rated 4 or greater or that increased by 2 points or more. Main Outcomes and Measures: Outcomes were determined a priori. The primary outcome was emergency department (ED) use and hospitalizations. Secondary outcomes were total costs, hospice, and, among decedents, acute care within 30 days of death and facility deaths. Results: Among 416 patients, the median age was 82 (range, 75-99) years; 219 (52.6%) were male; 171 (41.1%) had stage 4 disease; and 27 (6.4%) had recurrent disease. The mean risk adjustment factor score was 2.70 (SD, 1.77). Symptom assessment participants had 53% lower odds of ED use (61 [30.5%] vs 103 [47.7%] had ≥1 ED visit; adjusted odds ratio [OR], 0.47; 95% CI, 0.32-0.71), 68% lower odds of hospitalization (37 [18.5%] vs 86 [39.8%] had ≥1 hospitalization; OR, 0.32; 95% CI, 0.20-0.51), and lower mean total costs by $12 000 per participant (P = .01) than control group participants. Among 142 deceased participants (71 in each group), symptom assessment participants had 68% lower odds of ED use within 30 days of death (OR, 0.32; 95% CI, 0.12-0.88) and 75% lower odds of acute care facility death (OR, 0.25; 95% CI, 0.08-0.77). Conclusions and Relevance: A lay health worker-led symptom assessment intervention may be a scalable approach to reduce acute care use. Trial Registration: ClinicalTrials.gov Identifier: NCT04463992.