Continuous Wearable Biometric Monitoring in Adults Undergoing Obesity Treatment (SMART-BMI)

This study addresses a specific gap in the evidence for consumer wearable devices in obesity care. Prior trials that supplied wearable devices without integrating the resulting data into clinical decision-making have generally not improved weight outcomes. This trial is therefore designed to test not the device in isolation, but whether care teams acting on continuous biometric data achieve better outcomes than standard care. The trial uses a two-phase adaptive design at a single academic medical center. Two treatment cohorts are enrolled in parallel throughout: adults undergoing metabolic and bariatric surgery, and adults initiating GLP-1 receptor agonist therapy. All participants are followed for 52 weeks. Phase 1 is a single-arm observational feasibility phase. Every participant receives an Oura Ring, and the existing clinical team (dietitians, nurse practitioners, physicians, and coordinators) can view the biometric data through a dashboard during routine care. No dedicated coach and no additional study visits are added. Phase 1 validates the end-to-end data pipeline, characterizes device wear adherence and data completeness, and estimates the variability of candidate outcomes to inform Phase 2 sizing. As a pre-specified exploratory analysis, observed Phase 1 outcomes are compared against historical electronic health record-based controls within each treatment cohort; this comparison is descriptive and hypothesis-generating rather than a formal test of efficacy. Phase 1 data will be reviewed for feasibility, safety, and outcomes, including the exploratory comparison against historical controls, before Phase 2 is activated. Phase 2 proceeds only if Phase 1 demonstrates adequate feasibility and a directional signal that biometric-informed care may improve outcomes; absent any such signal relative to historical controls, Phase 2 will not be activated. Phase 1 results are analyzed and reported regardless of whether Phase 2 proceeds. Phase 2 is a randomized controlled trial with 1:1 parallel allocation. Phase 2 will also have a health coach-in-both-arms design: every Phase 2 participant receives a study-funded health coach. In the intervention arm, the care team can view the participant's Oura dashboard and incorporates the biometric data into management and coaching. In the control arm, the coach delivers a standardized behavioral curriculum without access to any wearable data. Because both arms receive equivalent coach contact, the design isolates the specific contribution of the biometric information rather than the added attention of a coach, removing attention bias as an explanation for any observed difference. The study is open label, as participants and coaches are necessarily aware of arm assignment. The primary outcome is drawn objectively from electronic health record data, which limits the influence of unmasked assessment. Sample size follows an adaptive approach. The maximum enrollment reflects an adaptive upper bound, and Phase 1 estimates allow re-estimation of the Phase 2 sample size at the interim review. The primary analysis follows a modified intention-to-treat approach, applying a pre-specified biometric data-quality threshold symmetrically across both arms. Between-arm comparison of the primary continuous outcome uses analysis of covariance adjusting for the baseline value. The wearable data are generated by a commercially available consumer device used consistent with its general wellness purpose; no investigational device exemption applies. The sponsor supplies devices and device-use guidance and reviews manuscripts within a pre-specified window, but holds no authority over data analysis, interpretation, or the decision to publish. The principal investigator retains scientific and publication independence.