A Cluster Randomized Trial of a Primary Palliative Care Intervention (CONNECT) for Patients With Advanced Cancer

Study investigators will conduct a cluster randomized controlled trial comparing the CONNECT intervention to usual care in 16 oncology clinics. CONNECT (Care Management by Oncology Nurses to address supportive care needs) is a care management intervention using existing oncology nurses to improve provision of primary palliative care within outpatient oncology practices.

The study will enroll 672 patients with advanced cancer and their caregivers. The study will assess the intervention's impact on patient and caregiver quality-of-life and psychological outcomes at three months and on healthcare utilization through one year of follow up. Investigators will follow a rigorous plan to establish and maintain intervention fidelity. Detailed process information will be collected at all sites and the study will evaluate how CONNECT impacts mortality and costs.

Following best practices for complex system-level interventions, the study will use a cluster randomized trial design. The unit of randomization is the oncology practice, defined as a unique location and provider group for outpatient oncology care. The unit of analysis is the individual patient and caregiver.

Designated staff at each clinic will review oncologists' schedules on a weekly basis to identify and track potentially eligible patients with an upcoming appointment. Research staff will review tracking systems weekly to maintain recruitment fidelity. Identified patients will receive a 1-page study information sheet at their next clinic visit, followed by a detailed in-person explanation of the study from a trained staff member. This staff member will obtain informed consent from all participants. To minimize potential selection bias, participants will be told in advance of both groups and not randomized if they decline to be exposed to one of them.

To enhance retention, blinded research assistants will conduct monthly follow-up calls to assess healthcare utilization for participants in both groups. The data collection strategy is designed to minimize participant burden by using parsimonious measures previously pilot-tested with this population.