A Conversational Interface to Improve Medication Adherence: Towards AI Support in Patients Treatment

Medication adherence is of utmost importance for many chronic conditions, regardless of the disease type. Engaging patients in self-tracking their medication is a big challenge. One way to potentially reduce this burden is to use reminders to promote wellness throughout all stages of life and improve medication adherence. Chatbots have proven effectiveness in triggering users to engage in certain activity, such as medication adherence. In this paper, we discuss “Roborto”, a chatbot to create an engaging interactive and intelligent environment for patients and assist in positive lifestyle modification. We introduce a way for healthcare providers to track patients adherence and intervene whenever necessary. We describe the health, technical and behavioural approaches to the problem of medication non-adherence and propose a diagnostic and decision support tool. The proposed study will be implemented and validated through a pilot experiment with users to measure the efficacy of the proposed approach.

💡 Research Summary

The paper presents “Roborto,” a conversational AI chatbot designed to improve medication adherence among patients with chronic conditions. Recognizing that traditional one‑way reminder systems often fail to sustain long‑term compliance, the authors adopt an interdisciplinary framework that integrates health, technology, and behavioral science. From the health perspective, they reference the WHO’s taxonomy of adherence barriers and construct patient profiles that capture socioeconomic status, disease complexity, and treatment regimens. Technologically, Roborto combines a BERT‑based natural language understanding engine, slot‑filling with conditional random fields, and sentiment analysis to capture dosage, timing, and side‑effect reports in real time. The system architecture comprises a mobile/web front‑end, a dialogue management module, a secure cloud‑based data layer, and a clinician dashboard that visualizes adherence metrics, risk signals (e.g., consecutive missed doses), and enables immediate intervention.

Behaviorally, the design is grounded in the COM‑B model and the Behaviour Change Wheel, aiming to boost Capability, Opportunity, and Motivation through personalized prompts, educational content, and feedback loops. A reinforcement‑learning scheduler dynamically adjusts reminder frequency and motivational messaging based on each user’s response pattern, thereby tailoring the intervention.

To evaluate efficacy, the authors propose an eight‑week pilot trial with 60 chronic‑illness patients randomly assigned to either the Roborto group or a control group receiving standard SMS reminders. Primary outcomes include adherence rate (accuracy and persistence), user experience (UEQ, SUS), and behavioral change indicators such as self‑efficacy scores. Mixed‑effects statistical models will assess time‑by‑group interactions. Preliminary expectations suggest a 15‑percentage‑point increase in adherence and higher satisfaction for the chatbot cohort.

The discussion acknowledges limitations: initial script rigidity, lack of multilingual support, and regulatory constraints on data privacy (GDPR, HIPAA). Future work envisions multimodal interfaces (voice, video), integration with electronic health records, and scaling the model across diverse linguistic and cultural contexts.

In conclusion, the study offers a comprehensive blueprint for a patient‑centric, AI‑driven medication adherence tool, demonstrates its potential through a structured pilot, and outlines a roadmap for broader clinical adoption and continuous improvement.