Abstract

<h3>Research Objectives</h3> To design and implement a Chatbot that interacts with patients via instant messaging to supervise domiciliary rehabilitation and increase adherence. To conduct a feasibility study. <h3>Design</h3> Pilot randomized trial (November 2021 to March 2022). <h3>Setting</h3> Two university hospitals. Outpatient and domiciliary rehabilitation. <h3>Participants</h3> 18 individuals, below 70, who underwent primary total knee replacement, had a personal smartphone, an instant messaging application installed, familiar with its use (>3 accesses pw) and able to consent. <h3>Interventions</h3> Patients underwent surgery and followed standard inpatient (≈2 days) and outpatient (≈2 to 4 weeks after surgery) physiotherapy care. The control group received one education session with recommended home exercises. The experimental group received the same education, but the domiciliary program was supervised via Chatbot, with automated interactions including messages to inform (e.g. about the disease, importance of compliance, progression), motivate, remind training days, and instruct the exercises up to 12 weeks' follow-up. <h3>Main Outcome Measures</h3> Compliance (primary), system usability, feasibility (recruitment and retention rates) and safety. <h3>Results</h3> Compliance was 15% higher in the experimental group, with n=7 (78%) participants classified as ‘achieved adherence' (>80% of sessions). As for feasibility, recruitment rate was good but lower than expected (75% of invited agreed); retention was good (>85%). Overall, the users reported that the tool was easy to use; however, some users (n=2) reported the need to learn how to use it, and one did not perform any access or interaction. <h3>Conclusions</h3> A Chatbot that communicates via instant messaging service has been successfully developed. The preliminary assessment suggests that the tool may be useful to increase compliance with in-home rehabilitation, and warrants a randomized clinical trial to determine the clinical impact. <h3>Author(s) Disclosures</h3> This result is part of the project PID2020-115825RA-I00, funded and supported by MCIN/ AEI/10.13039/501100011033, convocatoria Proyectos I+D+i 2020 - Modalidades "Retos Investigación" y "Generación de Conocimiento" (PID2020). The funder played no role on the design or results of this work.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call