Patient Engagement With an Automated Telephone Symptom Management Intervention: Predictors and Outcomes.

BACKGROUND: Self-management of symptoms related to cancer and its treatment is important for maintaining treatment regimens and improving outcomes. PURPOSE: To determine factors associated with engagement in a symptom self-management intervention among patients initiating oral anticancer treatment. METHODS: This secondary analysis included 127 patients randomized to the medication adherence reminder and symptom management intervention in a recently completed trial. Patients were recruited from six Comprehensive Cancer Centers, interviewed at intake, and mailed a Symptom Management Toolkit (Toolkit) with self-care management strategies for 18 symptoms. During eight automated telephone weekly calls, patients were asked to use the Toolkit to manage elevated symptoms. Toolkit use and symptoms were tracked weekly, and generalized linear mixed-effects models were used to determine factors predictive of Toolkit use. General linear modeling was used to relate the Toolkit use during intervention to postintervention symptom severity. RESULTS: Better cognitive function at intake into the trial and higher symptom burden were predictive of the patients' initial decision to try the Toolkit during Week 1. In subsequent weeks, Toolkit use in the previous week and worsening of symptoms were associated with greater odds of Toolkit use. The extent of Toolkit use modified the relationship between intake and 8 week symptom severity: among patients with higher levels of severity at intake, use of the Toolkit conferred greater benefit at 8 weeks. CONCLUSIONS: Patients make realistic decisions regarding when to use a self-directed approach to self-management and are likely to use strategies when their symptoms are higher and to forego use once symptoms subside. CLINICAL TRIAL REGISTRATION: NCT02043184.