Abstract Background Effective postdischarge management is essential for maintaining disease control and improving long-term outcomes in rheumatoid arthritis (RA). Digital health technologies, particularly intelligent management platforms, provide new opportunities for continuous care and self-management in real-world settings. Objective This study aimed to develop a nurse-led artificial intelligence (AI)–assisted interactive management platform for postdischarge RA care and examine its association with patient outcomes in a real-world clinical setting. Methods A single-center, multicampus, real-world observational study with a quasi-experimental framework was conducted at Tongji Hospital between November 2024 and October 2025. Participants were allocated to either a platform-based management group (campus 1; n=184) or a comparison group receiving routine postdischarge care (campus 2; n=157). The platform enabled remote monitoring through a smartphone app, which collected information on joint symptoms, fatigue, medication adherence, laboratory results, and emotional status, with data uploaded to a secure cloud server. The AI system analyzed data in real time and alerted health care providers to abnormalities. Nurses and health coaches delivered personalized education and interventions. Primary outcomes were changes in Disease Activity Score in 28 Joints and Health Assessment Questionnaire–II scores at discharge and the 6-month follow-up. Secondary outcomes included medication adherence measured as the proportion of prescribed doses taken and patient satisfaction assessed via a 5-point Likert-scale questionnaire at 6 months. Results A total of 341 patients with RA completed the 6-month follow-up. Disease Activity Score in 28 Joints values decreased significantly in the platform-based management group at 6 months (mean 5.04, SD 1.38 at discharge to mean 4.27, SD 1.20 at 6 months; P <.001), showing a greater reduction than in the comparison group (mean 4.86, SD 1.19 at discharge to mean 4.59, SD 1.36 at 6 months; within-group P =.03; between-group P =.02). Health Assessment Questionnaire–II scores improved significantly in the platform-based management group (mean 2.45, SD 0.75 at discharge to mean 1.67, SD 0.56 at 6 months; P <.001), while the comparison group showed no significant change ( P =.12; between-group P <.001). Medication adherence was higher in the platform-based group (166/184, 90.2% vs 125/157, 79.6%; χ 1 2 =7.6; P =.006). Overall health care satisfaction was also higher in the platform-based group (176/184, 95.7% vs 120/157, 76.4%; χ 1 2 =27.8; P <.001), with a greater proportion reporting to be “very satisfied” in this group (150/184, 81.5% vs 59/157, 37.6%). Conclusions The intelligent interactive management platform markedly enhanced disease control, functional status, medication adherence, and satisfaction among patients with RA after discharge. This nurse-led, AI-assisted system represents a promising model for improving outpatient RA management.
Zhang et al. (Thu,) studied this question.