Essential Components of an Electronic Patient-Reported Symptom Monitoring and Management System: A Randomized Clinical Trial

Abstract

Importance: Multicomponent electronic patient-reported outcome cancer symptom management systems reduce symptom burden. Whether all components contribute to symptom reduction is unknown. Objective: To deconstruct intervention components of the Symptom Care at Home (SCH) system, a digital symptom monitoring and management intervention that has demonstrated efficacy, to determine which component or combination of components results in the lowest symptom burden. Design, Setting, and Participants: This randomized clinical trial included participants who were older than 18 years, had been diagnosed with cancer, had a life expectancy of 3 months or greater, were beginning a chemotherapy course planned for at least 3 cycles, spoke English, and had daily access and ability to use a telephone. Eligible participants were identified from the Huntsman Cancer Institute, University of Utah (Salt Lake City), and from Emory University Winship Cancer Institute, including Grady Memorial Hospital (Atlanta, Georgia), from August 7, 2017, to January 17, 2020. Patients receiving concurrent radiation therapy were excluded. Dates of analysis were from February 1, 2020, to December 22, 2023. Interventions: Participants reported symptoms daily during a course of chemotherapy and received automated self-management coaching with an activity tracker without (group 1) and with (group 2) visualization, nurse practitioner (NP) follow-up for moderate-to-severe symptoms without (group 3) and with (group 4) decision support, or the complete SCH intervention (group 5). Main Outcomes and Measures: The primary outcome, symptom burden, was assessed as the summed severity of 11 chemotherapy-related symptoms rated on a scale of 1 to 10 (with higher scores indicating greater severity), if present. Results: The 757 participants (mean [SD] age, 59.2 [12.9] years) from 2 cancer centers were primarily female (61.2%). The most common cancer diagnoses were breast (132 [17.4%]), lung (107 [14.1%]), and colorectal (99 [13.1%]) cancers; 369 patients (48.7%) had metastatic disease. The complete SCH intervention including automated self-management coaching and NP follow-up with decision support (group 5) was superior in reducing symptom burden to either of the self-management coaching groups, as shown by the mean group differences in area under the curve (group 1, 1.86 [95% CI, 1.30-2.41] and group 2, 2.38 [95% CI, 1.84-2.92]; both P