Oncologists' and Patients' Perceptions of Initial, Intermediate, and Final Goals of Care Conversations

Context. Although recommendations are for goals of care (GoC) conversations to happen early and often, their timing and content over the disease course remain unclear. Objectives. To describe perceptions of the timing and content of GoC conversations among oncologists and patients with advanced cancer. Methods. Semi-structured interviews with oncologists and patients at four hospitals in New York and Connecticut. We analyzed data using interpretive description. Results. Oncologists (n = 21) were 67% male, averaged 46 years old (range 34-68 years), and had a mean 20 years (range 8-42 years) in practice. Advanced cancer patients (n = 39) were 59% female with a mean age of 58 years (range 26-88 years). Oncologists and patients viewed GoC conversations along an initial, intermediate, and final continuum. Oncologists reported having initial GoC conversations to build rapport, provide disease information, and clarify and consider treatment intent and options. Intermediate conversations focused on changes in treatment and side effects. Final conversations occurred when patients were hospice-eligible and covered the shift to comfort care, a treatment summary, and advance directives. Timing and content were influenced by cancer type, disease status, and emotional state. Patients supported beginning GoC conversations early, but had different preferences for content of initial conversations. Some wanted complete information; others found this overwhelming. Patients felt intermediate GoC conversations should occur at decision points with time for processing. Content of intermediate and final conversations was dependent on patients' prognosis, questions, and treatment decisions. Conclusion. Although GoC conversations should follow individual patient preferences, our findings offer structure for their timing and content. (C) 2017 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved.