In a multisite randomized clinical trial of 416 Medicare Advantage beneficiaries age ≥75 with newly diagnosed, recurrent, or progressive cancer, a lay health worker–led telephone symptom-monitoring program cut any ED use at 12 months to 30.5% vs 47.7% with usual care (aOR 0.47). Hospitalizations also fell to 18.5% vs 39.8% (aOR 0.32) in the trial (ClinicalTrials.gov NCT04463992).

Why It Matters To Your Practice

• Older adults with cancer often have undertreated symptoms that escalate into ED visits and admissions.

• This intervention used lay health workers to collect Edmonton Symptom Assessment System scores by phone and escalated concerning symptoms to advanced practice practitioners—matching real-world oncology workflows.

• The study population (median age 82; 41.1% stage 4 disease) mirrors many high-risk panels seen in oncology and geriatrics.

Clinical Benefits

• Fewer ED visits: 61/200 (30.5%) vs 103/216 (47.7%) had ≥1 ED visit over 12 months (aOR 0.47; 95% CI, 0.32-0.71).

• Fewer hospitalizations: 37/200 (18.5%) vs 86/216 (39.8%) had ≥1 hospitalization (aOR 0.32; 95% CI, 0.20-0.51).

• Lower costs: mean total costs were ~$12,000 lower per participant vs usual care (P = .01).

Managing Risks

• Set clear escalation thresholds and standing orders: symptoms rated ≥4 or rising by ≥2 points triggered APP review in the study.

• Plan for access barriers (hearing impairment, cognitive decline, language needs) that can affect phone-based symptom capture; use caregiver involvement and interpreter workflows.

• Avoid over-triage: pair lay-led screening with APP-guided protocols to direct patients to self-management, urgent clinic evaluation, palliative care, or ED based on red flags.

The Bottom Line

• For older adults with cancer (malignant neoplasm), lay health worker–led symptom assessments with APP escalation were associated with materially fewer ED visits and hospitalizations at 12 months vs usual care.

• This is a scalable model for oncology practices looking to reduce acute care utilization without relying solely on clinician-led monitoring.