Maximizing Online Obesity Treatment Response in the Primary Care Environment Using Clinical Decision Support

PROJECT SUMMARY/ABSTRACT With the long-term career goal of becoming a leading independent researcher using technology-driven precision medicine approaches to improve obesity outcomes in primary care, the candidate, Dr.

Hallie Espel- Huynh, PhD, proposes a mentored research project and career development plan that will prepare her to use advanced analytics to personalize routine clinical care for patients with obesity.

Despite the promise of online interventions to maximize access to effective behavioral obesity treatments in primary care, many patients do not benefit due to nonresponse.

Evidence-based rescue interventions (EBRI) can improve outcomes for these patients, however, primary care clinicians require guidance on when and how to intervene, and such a tool does not yet exist.

Clinical decision support (CDS) has the potential to fill this gap by predicting risk for nonresponse, delivering alerts about this risk to clinicians, and enabling interventions to reverse it.

The overall objective of this training application is to develop such a CDS for use in primary care and test its usability with primary care clinicians.

The proposal aims to (1) use stakeholder input to align the content and format of the CDS with primary care providers? needs, (2) build a machine learning model to predict early risk for online obesity treatment nonresponse for integration into the CDS, and (3) design the prototype CDS and test its usability with primary care clinicians, focusing on outcomes of feasibility, acceptability, and appropriateness for the target primary care setting.

This project is the first to combine precise nonresponse risk prediction with stakeholder-informed CDS to produce a tool that has the potential to maximize obesity treatment outcomes in primary care via delivery of CDS-facilitated, clinician-delivered rescue interventions.

This research will result in a complete CDS tool that is ready for future clinical testing with patients in primary care, and which could greatly enhance the potential impact of online obesity treatment in this setting.

The research plan is complemented by career development activities that include formal training in technology-assisted obesity and CVD management in primary care, stakeholder-centered CDS development, machine learning, and mixed methods for patient-oriented implementation research.

Under the guidance of an experienced mentorship team, execution of the proposed research and training plan will lead Dr.

Espel-Huynh to submission of a competitive R01 grant application to test CDS effectiveness in a pragmatic randomized clinical trial.