The effect of real-time electronic monitoring of patient-reported symptoms and clinical syndromes in outpatient workflow of medical oncologists: E-MOSAIC, a multicenter cluster-randomized phase III study (SAKK 95/06)

Background: Patients with advanced, incurable cancer receiving anticancer treatment often experience multidimensional symptoms. We hypothesize that real-time monitoring of both symptoms and clinical syndromes will improve symptom management by oncologists and patient outcomes.

Patients and methods: In this prospective multicenter cluster-randomized phase-III trial, patients with incurable, symptomatic, solid tumors, who received new outpatient chemotherapy with palliative intention, were eligible. Immediately before the weekly oncologists' visit, patients completed the palm-based E-MOSAIC assessment (Edmonton-Symptom-Assessment-Scale, ≤3 additional symptoms, estimated nutritional intake, body weight change, Karnofsky Performance Status, medications for pain, fatigue, nutrition). A cumulative, longitudinal monitoring sheet (LoMoS) was printed immediately. Eligible experienced oncologists were defined as one cluster each and randomized to receive the immediate print-out LoMoS (intervention) or not (control). Primary analysis limited to patients having uninterrupted (>4/6 visits with same oncologist) patient-oncologist sequences was a mixed model for the difference in patients global quality of life (G-QoL; items 29/30 of EORTC-QlQ-c30) between baseline (BL) and week 6. Intention-to-treat (ITT) analysis included all eligible patients.

Results: In 8 centers, 82 oncologists treated 264 patients (median 66 years; overall survival intervention 6.3, control 5.4 months) with various tumors. The between-arm difference in G-QoL of 102 uninterrupted patients (intervention: 55; control: 47) was 6.8 (P = 0.11) in favor of the intervention; in a sensitivity analysis (oncologists treating ≥2 patients; 50, 39), it was 9.0 (P = 0.07). ITT analysis revealed improvement in symptoms (difference last study visit-BL: intervention -5.4 versus control 2.1, P = 0.003) and favored the intervention for communication and coping. More patients with high symptom load received immediate symptom management (chart review, nurse-patient interview) by oncologists getting the LoMoS.

Conclusion: Monitoring of patient symptoms, clinical syndromes and their management clearly reduced patients' symptoms, but not QoL. Our results encourage the implementation of real-time monitoring in the routine workflow of oncologist with a computer solution.