Using Modifiable Factors to Predict Risk of non-Guideline Concordant Treatment in Foregut Cancers: A Novel Implementation Science-based approach

PROJECT SUMMARY Foregut (esophageal, gastric, liver, biliary and pancreatic) cancers account for approximately 20% of new cancer deaths in the United States each year, with an incidence that is projected to increase by almost 70% over the next ten years. These are aggressive malignancies, with complex therapies that require coordinated, specialized multidisciplinary care. However, up to 70% of patients with foregut cancers do not receive guideline-concordant treatment (GCT). Patient demographic characteristics and geographic location serve as proxies for multiple, potentially modifiable factors along the cancer care continuum, that threaten delivery and receipt of care. Identifying these factors is critical to improve the delivery of high quality cancer care. Accurate identification of patients at risk for non-GCT is a vital first step to intervene. However, there is no screening tool to identify patients at risk for non-GCT. This project aims to use mixed methods research to develop a clinic-based risk prediction tool that utilizes modifiable risk factors to identify patients at risk for non-receipt of GCT. Together, this clinically applicable and statistically valid model and the high-yield screening instrument will inform an emerging model that identifies patients at risk for non-receipt of care, and facilitates the administration of personalized, solution-focused interventions that can redirect the course of care. My long-term objective is to become an independent researcher focused on understanding and developing systems-based approaches to improve the delivery of cancer care. This career development award will help achieve my objective by filling gaps in knowledge and skills through a personalized training program that will include didactics, mini-sabbaticals, experiential training and mentorship in three areas: (1) mixed methods with a focus on patient-oriented research (2) statistical modeling and measure development and (3) implementation science. This training plan will support my proposed project and enable my transition to an independent surgeon-scientist. The specific aims are (1) to identify challenges in access to GCT in patients with foregut cancers through a patient-centric and stakeholder-informed approach (2) to examine the extent to which modifiable factors predict risk for non-GCT using quantitative modeling and (3) to develop and pilot test a clinic-based screening tool to prospectively identify patients at risk for non-GCT. This proposal is novel in its exploration of underlying modifiable determinants to inform the development of a model and clinic-based screening tool that predicts risk for non-GCT. While there are currently a range of successful interventions such as patient navigation that have shown promise in improving the delivery of cancer care, there is no streamlined way for clinicians to identify patients at risk for non-GCT. This proposal will allow for the identification of patients at risk for non-GCT in a consistent way across providers and embedded in health systems. The results of this study will be used to apply for an R01 to prospectively evaluate