Chronic pain affects millions in the US—over 50 million adults—with intricate physical, emotional, and social components. When it endures beyond half a year, it turns chronic, shaped by diverse factors like emotions, psychology, and physical health. Among treatments, Spinal Cord Stimulation (SCS) stands out, often reducing pain and enhancing life quality. Responses vary, demanding precise adjustment for optimal results. Selecting and monitoring individuals are vital for SCS success.
Traditionally, chronic pain is assessed with patient-reported outcomes (PROs) during clinic visits. Gold standards are the Numerical Rating Scale (NRS) or Visual Analog Scale (VAS). To capture pain’s multidimensionality, validated PRO questionnaires like Pain Catastrophizing Scale (PCS) and Oswestry Disability Index (ODI) are used. Yet, subjectivity and bias limit these methods. Multiple questionnaires in short intervals are burdensome.
No established measures objectively evaluate chronic pain and SCS’s impact. Wearable tech offers “digital biomarkers,” quantifying activity, function, sleep, and more. These promise objective insight.
Wearable tech and machine learning present the chance to predict pain and outcomes. Prior work focused on symptoms and side effects, not predictive outcomes.
Wearable tech and advanced machine learning offer predictive power for subjective pain measures.
We combined smartwatch data and ML to predict common PROs for chronic pain. Though not direct pain measures, they gauge life quality and disability accurately. Our goal: predict SCS response with objective smartwatch data.
Results Subjects and Adherence Twenty participants enrolled, five withdrew. Median adherence: 88.8% for iPhone PROs, 84.7% for smartwatch wear. No significant difference. Participants averaged 52.25 years, enduring chronic pain for 12 years. Back pain dominated diagnoses (85%).
Source Nature.com
