Early integrated palliative care (PC), entailing monthly visits with a PC clinician in the outpatient setting throughout the course of illness for patients with advanced care, improves both patient-reported and end-of-life (EOL) care outcomes. However, few institutions possess a sufficient number of PC trained clinicians and resources to provide monthly visits with a PC clinician for all patients with advanced cancer. Additionally, such frequent visits with a PC clinician may not be necessary for patients with minimal physical or psychological symptoms. Stepped care is a health service delivery model to increase access to and efficiency of care when there is a limited number of specialty trained clinicians. In stepped care, all patients receive care with the minimal necessary time with the trained clinician, but are monitored systematically to “step up” to more intensive treatment if they do not achieve sufficient health gain with low intensity treatment. Based upon our multidisciplinary teams' extensive experience developing and studying early integrated PC in oncology, we propose to adapt our effective outpatient PC model to a stepped care strategy to ensure all patients access PC service, while tailoring treatment delivery to patients' needs. Study Aims: The specific aims of this multi-center randomized trial are to demonstrate the non-inferiority of stepped PC to standard-of-care early integrated PC with respect to its effects on quality of life (primary outcome), as well as patient-clinician communication about EOL care preferences and length of stay in hospice. We will also compare PC resource utilization and the cost effectiveness of the two delivery models. Study Design and Methods: We will enroll and randomly assign 480 patients newly diagnosed with advanced lung cancer (240 per study group) to receive either stepped PC or early integrated PC. The trial will take place at Massachusetts General Hospital, Columbia University, and Duke University. Patients randomized to stepped PC will meet with the PC clinician at enrollment and at clinically relevant points in their illness. We will assess participants' quality of life every six weeks to detect stepped care patients whose quality of life is deteriorating to allow them to step up to monthly PC visits. Patients randomized to early integrated PC will participate in monthly visits with the PC clinician throughout their course of illness. In addition to quality of life, all participants will complete measures to assess their health status, mood, coping, prognostic understanding, and communication about their EOL care preferences every 12 weeks for one year. We will also collect data from the medical record and hospital cost accounting system on hospice use, PC resource utilization, and health care costs. Conclusions: Early integrated PC improves patient-reported outcomes, including quality of life and mood, and the delivery of EOL care. Based upon data demonstrating the efficacy of this care model, several national organizations recommend early integrated PC as standard of care. We seek to establish the non-inferiority of a stepped PC model compared with early integrated PC, as a more efficient, accessible, patient-centered, and scalable approach to provide early PC.