LAUSR

174Background: Patient navigation programs in cancer care have historically focused on assisting persons to overcome barriers to accessing care. Evidence is emerging to support the impact of navigation interventions across the cancer continuum. However, navigation programs have varied designs, resulting in a lack of clarity about the optimal approach to delivering services to patients, and a lack of evidence linking program design to outcomes. Methods: A planned retrospective analysis of Medicare administrative claims for a population of older beneficiaries diagnosed with cancer: The main exposure was the number of contacts in person or over the phone with PCCP navigators in the 6 month period starting from the quarter in which patients enrolled in the PCCP. Repeated measures generalized linear models with normal distribution were used to evaluate trends in total cost over time based on: number of contacts, quarters post-enrollment (TIME), and the interaction between number of contacts and TIME. Intra-cor...