Exploring the Effect of an 8-Week AI-Composed Exercise Program on Pain Intensity and Well-Being in Patients With Spinal Pain: Retrospective Cohort Analysis

• DOI: 10.2196/57826
• Publication Date: 2025-02-18T21:31:36Z
• Authors: Annika Griefahn; Florian Avermann; Christoff Zalpour; Robert Percy Mars...; Inés Cordon Morillas; Kerstin Luedtke
• Keywords: McGill Pain Questionnaire; Intensity; Oswestry Disability Index

Abstract Background Spinal pain, one of the most common musculoskeletal disorders (MSDs), significantly impacts quality life due to chronic pain and disability. Physical activity has shown promise in managing spinal although optimizing adherence exercise remains a challenge. The digital development artificial intelligence (AI)-driven applications offers possibility for guiding supporting patients with MSDs their daily lives. Objective trial aimed investigate effect an 8-week AI-composed program on intensity well-being pain. It also examined relationship between frequency, intensity, well-being. In addition, app usage frequency was as proxy engagement. Methods Data from users who met inclusion criteria were collected retrospectively medicalmotion January 1, 2020, June 30, 2023. intervention involved use app, which provides 3‐5 personalized exercises each session based individual user data. primary outcomes assessed using numeric rating scale (NRS) Likert scale. at baseline (t0), 4 weeks (t1), 8 (t2). correlation analyzed secondary outcome. average length measured determine Statistical analysis included ANOVA Spearman analysis. Results study 379 participants mean age 50.96 (SD 12.22) years. At t2, there significant reduction 1.78 points NRS ( P <.001). score improved by 3.11 after weeks. Pain showed negative number performed t1 t2. Well-being had small per day. day 3.58. approximately 10 minutes, interaction 49.2% (n=27.6 days) 56 available days. Conclusions Overall, demonstrates that app-based can substantially reduce increase This retrospective digitizes multidisciplinary rehabilitation self-management reduced user-reported preselected population users.