This real-world observational study evaluated clinical, patient-reported (PROMIS), and wearable biosensor (Fitbit) outcomes in RA patients with moderate-to-high disease activity (CDAI >10) initiating upadacitinib or adalimumab across 28 U.S. community rheumatology clinics, followed for ≥12 weeks (enrollment 03/2021–10/2022).
Among 150 patients with paired clinical and PRO data, mean CDAI improved by 13 units and PROMIS Pain Interference, Fatigue, and Sleep Disturbance scores each improved by 4–5 units; PRO changes explained 52% of the variability in CDAI change (R²=0.52), with moderate correlations between CDAI and PROs of 0.38–0.56. Sedentary patients also showed improvement in wearable-measured activity.
- High attrition: only 150 of 299 enrolled patients (50%) provided both clinical and PRO data, raising selection bias concerns. - Non-randomized real-world design; treatment assignment was not controlled, limiting causal inference between drug type and outcomes. - Biosensor data available for only 95 patients (63% of the main cohort), further limiting generalizability of wearable findings.
PRO and wearable biosensor data captured between clinic visits correlate meaningfully with formal disease activity scores in RA patients starting new therapy — clinicians can consider app-based PRO tools to monitor treatment response remotely. Patients who are sedentary at baseline may show activity gains detectable by wearables, offering an additional signal of treatment benefit.
Explore related topics
