Effectiveness of Conversational Agents on Patient-Reported Outcomes in Chronic Pain Management: A Systematic Review and Meta-Analysis

Abstract

Background: Chronic pain remains a primary driver of global disability and impaired quality of life. While digital conversational agents (CAs) have emerged as scalable tools for symptom monitoring and self-management via patient-reported outcome measures, their clinical efficacy remains poorly synthesized. This systematic review and meta-analysis aimed to evaluate the impact of CA-based interventions on PROMs in adults with chronic pain. Methods: A systematic review and meta-analysis was conducted following PRISMA 2020 guidelines. PubMed, Scopus, and Web of Science were searched from inception to 22 October 2025. Eligible studies were RCTs including adults with chronic pain and evaluating fully automated CA interventions, such as digital coaching or messaging programs. PROMs related to pain, well-being, disability, and work-related outcomes were extracted. Continuous outcomes were synthesized using standardized mean differences (SMDs) with 95% confidence intervals (CIs). Results: Five RCTs involving 572 participants were included. Interventions were self-guided, digitally delivered, and lasted 4 to 12 weeks. The overall pooled analysis suggested a potential benefit of CA-based interventions on PROMs (SMD = −0.43; 95% CI −0.55 to −0.31; p < 0.00001), although heterogeneity and risk of bias across studies warrant cautious interpretation. Improvements were observed particularly in pain intensity, although evidence for other outcomes was less consistent, with some studies reporting benefits in quality of life, fear of movement, and well-being. Conclusions: CA-based interventions may have potential as adjuncts in chronic pain management; however, the current evidence is limited and should be interpreted with caution due to heterogeneity and risk of bias across studies. These tools may represent a scalable solution for supporting remote symptom monitoring and self-management within digital health frameworks, although further high-quality evidence is required.