Desenvolvimento de um sistema inteligente de porta-comprimidos para melhorar a adesão à medicação

Introduction: The aging population and associated polypharmacy increase the risk of medication non-adherence, negatively affecting health and quality of life. Existing smart pillboxes help improve adherence but fall short in other aspects. Medication management requires training and education to be effective. Objective: This study was aimed at designing, developing and testing an original pillbox system that would automate the process of medicine preparation, dispensing and monitoring medication adherence and provide real-time feedback and cloud-based communication with caregivers and healthcare professionals. Methods: The prototype was built on a control system based on the Arduino platform, infrared sensors, GSM/Wi-Fi communication modules, and a real-time clock in the form of a modular prototype. The system was interconnected to a cloud network and an Android remote monitoring application. Laboratory validation of the prototype was done, Usability testing was conducted on the prototype with older people, and a field trial was conducted to estimate the improvement in adherence, reliability and user satisfaction. The quantitative measures of performance were dispensing accuracy, adherence rate, system-up-time and alert response-time. Results: Laboratory experiments showed the pill dispensing of 98.6%, sensor detection of 98 and continuous backup power of 5 hours in case of simulated outage. The results of usability testing on 17 participants contributed to the mean System Usability Scale (SUS) score of 84/100, which demonstrates that users are highly satisfied and find it easy to use it. Post-intervention adherence was 92.0% (SD 4.1%), compared to baseline 88.7% (SD 5.3%), absolute difference 3.3% (95% CI: 1.2–5.4). Response times of cloud synchronization were less than 5 seconds, and system uptime was 99.2%. Conclusion: The innovative pillbox system developed had potential to enhance medication adherence and reliability, decrease number of missed doses and provide the caregivers with reliable real time monitoring. Its great usability and technical stability promise to be a good assistive technology among the elderly and chronically ill patients. Further efforts in the field should be devoted to the extensive clinical validation, the long-term performance assessment, and the collaboration with electronic health records and AI-powered predictive analytics.