The PIVOT Trial: Project on EHR-Integrated Lifestyle Interventions for Adults Aged Fifty and Older

This is a multisite clinical trial of healthy lifestyle programs for middle-to-older-aged adults (50 -74 years) to improve their weight and, therefore, reduce risk for chronic disease. This study will investigate whether a proven, self-directed video-based lifestyle program can be significantly enhanced with the assistance of a coach via videoconference or phone to help people as they adopt healthy eating and exercise behaviors. Additionally, the study will examine what factors might explain why some people achieve better outcomes than others. Understanding this can help to tailor the program to an individual for personalized care in the future. Importantly, this study aims to demonstrate how readily a digital lifestyle program, with or without remote coaching, can be seamlessly delivered to patients at home via the patient portal of their electronic health record. This practical use of existing telehealth tools could be a feasible and effective means to offer behavioral treatments during routine medical care.

This multisite clinical trial uses a 2-stage sequential randomization design to test the adaptive and nonadaptive augmentation of a validated Group Lifestyle Balance (GLB) video program (base intervention) using problem solving treatment (PST), a proven behavior therapy. The GLB video program will be delivered via the patient portal of the electronic health record system. Trained coaches will deliver PST remotely via videoconference or phone. English-speaking adults (N=1029), 50-74 years of age with a body mass index ≥27 and ≥1 cardiometabolic conditions, will be randomized at baseline to base intervention or waitlist control. Participants initially randomized to the waitlist control will be re-randomized after a 12-week control period to receive the base (Group A) or the augmented intervention (Group B), without tailoring based on early weight loss. Among participants initially randomized to the base intervention, responders defined by ≥3% weight loss at 6 weeks will continue the base intervention (Group C); nonresponders who are participants with \<3% weight loss or missing weight data by 6 weeks will be re-randomized to continue the base intervention alone (Group D) or augmented with PST coaching via videoconference (Group E). We hypothesize: (1) the augmented intervention will be more effective than the base intervention both among early nonresponders to the base intervention (adaptive) and among participants in the waitlist control condition (nonadaptive) at 52 weeks; (2) the adaptive augmented intervention will be more efficacious than the base intervention and more efficacious than the waitlist control condition at 12 weeks. Aim 2 is to identify predictors of clinically significant (5%) weight loss for individual patients, using sociodemographic, clinical and behavioral engagement characteristics.