Understanding the impact of an EHR-integrated hereditary cancer risk assessment application on patient-provider communication

This application is being submitted in response to the Notice of Special Interest (NOSI) identified as NOT-CA- 23-041. Our current approach to cancer care remains reactive, with late presentation of many patients. Though innovation is rapidly advancing, effective interventions are often not perpetuated into practice. The Program in

Cancer Risk, Prevention, and Early Detection (CaRPED) of the Dana-Farber/Harvard Cancer Center (DF/HCC) arose out of growing recognition of the importance of bridging technology with DF/HCC expertise in cancer science to detect cancer at earlier stages, when interventions are most effective. The Cancer Care

Delivery Research (CCDR) program was designed to investigate strategies to improve care delivery to ensure that interventions that theoretically can work based on clinical trials, do work in the context of oncology practice, and that vulnerable populations are not left behind. The Principal Investigator is co-leader of the

CaRPED program and has led the development of the PREMM models (funded by RO1CA132829, 2008- present). The PREMM5 algorithm assesses for one of the most common hereditary cancer syndromes, Lynch syndrome (LS), which affects 1 in 279 people and causes high lifetime cancer risk. PREMM5 is now the guideline-recommended standard for assessing LS risk. We have recently developed a literacy-adapted,

patient-facing hereditary cancer risk assessment app based on PREMM5. When embedded in the electronic health record (EHR), PREMM5 was able to identify a large number of at-risk patients, however, three quarters of at-risk patients eligible did not receive genetics referral or testing. To optimize the benefit of risk

assessment, it is critical to understand the role of the PREMM5 app as a clinical decision support tool and patient-provider communication support tool in clinical practice. This proposed supplemental project represents a collaboration between the DF/HCC CaRPED and CCDR programs and the Survey and Data Management

Core and will investigate the role of PREMM5 in patient-provider communication through evaluation of patient outcomes in order to identify key areas for improvement. We will used a mixed-methods approach – in Aim 1 we will quantitatively assess whether there are associations between demographic factors and patient

outcomes, including discussion of PREMM5 during the clinical visit, referral receipt, referral uptake, and genetic testing uptake in individuals who chose to proceed with genetics referral versus those who didn’t. In Aim 2, we will use in-depth semi-structured interviews with at-risk patients per their PREMM5 scores and their

providers to elucidate key barriers and facilitators to use of PREMM5 as a communication support tool to facilitate downstream care. The result will be a more equitable and clinically useful version of PREMM5 that can be implemented DF/HCC-wide to improve patient-provider communication about risk for LS.

Further, we plan to use insights gained in this study to guide development of a multi-gene panel patient-facing risk assessment tool (PREMMplus) to assess risk for 19 hereditary cancer genes.