Mapping the Evolving AI Preferences and Care Needs in Orthopedic Transitional Care From Hospitals to Home: Cross-Sectional Study.

Background: Enhanced recovery after surgery protocols have shortened orthopedic hospital stays but have shifted rehabilitation and safety-monitoring tasks to patients and families after discharge. In this study, AI refers to patient-facing digital systems for orthopedic transitional care, including large language model chatbots, computer vision or platform-based monitoring tools, and wearable sensor-enabled systems for education, rehabilitation guidance, motion correction, and risk alerts. However, patient-reported preferences for different AI-supported functions across the hospital-to-home transition remain underexplored. Objective: This study aimed to map the evolution of care needs from hospital to home recovery and to identify specific preferences and factors associated with the willingness to use AI systems in orthopedic transitional care. Methods: We conducted a multicenter cross-sectional survey among patients recovering from orthopedic surgery in 33 Guangdong hospitals, China. Of the 860 submitted questionnaires, 752 responses were included after prespecified quality control, including exclusion of responses completed in 180 seconds or less based on pilot-informed screening. Participants rated standardized function-based AI descriptions rather than a specific prototype or live tool. The data covered demographic and clinical characteristics, task priorities across care phases, perceived transitional care challenges, and stated willingness to use AI. The survey was informed by the technology acceptance model, although perceived usefulness, perceived ease of use, and attitude toward use were not directly measured. Exploratory factor analysis was used to examine perceived challenges. Descriptive mapping summarized care needs and AI function preferences, and multivariable logistic regression explored factors associated with willingness. Results: Most respondents reported a willingness to use AI (604/752, 80.3%). Care priorities varied by phase: inpatient priorities were more often related to information acquisition and care instruction, whereas home-stage priorities more often involved functional safety, rehabilitation guidance, motion correction, and risk alerts. Exploratory factor analysis identified 3 perceived challenge dimensions: home rehabilitation self-management barriers, lack of professional support, and symptom uncertainty. In adjusted exploratory analysis, willingness to use AI was associated with older age (adjusted odds ratio [aOR] 1.02, 95% CI 1.00-1.03), comorbidities (aOR 1.72, 95% CI 1.09-2.69), later rehabilitation stage (aOR 1.28, 95% CI 1.01-1.62), and urban residence (aOR 1.85, 95% CI 1.14-3.01). Unmarried, divorced, or widowed status was associated with lower willingness than married status (aOR 0.59, 95% CI 0.39-0.89). Physical disability and self-care ability were not independently associated with willingness after adjustment. Conclusions: In this hospital-based convenience sample, most respondents were willing to use AI, and their stated priorities shifted from information support during hospitalization to functional safety and rehabilitation support after discharge. The associated factors should be interpreted as exploratory associations rather than causal determinants. Because participants evaluated function-based AI descriptions rather than actual AI tools, these findings can inform future prototype development and real-world evaluation, particularly around usability, trust, privacy, digital accessibility, and clinician oversight.