This study will investigate how PATHWEIGH, a weight management tool built into the EPIC electronic health record, in combination with clinician and staff training on the use of PATHWEIGH as a tool, education on effective weight management practices, and practice facilitation to support implementation can facilitate weight loss and maintenance for primary care patients. The study examines patient weight loss and maintenance, associated predictors (patient health characteristics and demographics), and contextual factors in the practice environment that influence the intervention's adoption, implementation, and sustainability. This will be achieved through a mixed methods implementation study including collection and analysis of de-identified clinical data, surveys, observations, and interviews.
This study is being conducted because despite obesity being the leading cause of preventable death in the U.S. and a major risk factor for disease, people with any degree of overweight or obesity are rarely offered anything other than lifestyle advice, rather than more rigorous and evidence-based weight management strategies. Poor reimbursement for weight-related visits, competing priorities, lack of appropriate tools, and limited time in visits also prevent more widespread use of weight management interventions. PATHWEIGH helps primary care clinicians and teams prioritize weight management with their patients when appropriate and ensures that they have the tools and knowledge handy to do so effectively. As part of the assessment of the PATHWEIGH intervention, the study will identify predictors of patient weight loss and weight loss maintenance using mediator and moderator analysis. Relevant factors (patient demographics, health metrics (e.g. vital signs, anthropometrics, comorbidities), behaviors and goals, provider age, sex, years and % time practicing, and clinic information (e.g. size, location, culture scale)) will be evaluated by including interaction terms between the treatment variable and possible moderating variables in outcome regression models .
