Comprehensive Digital Metric for Chronic Back and Leg Pain Health Status

Overview

This study developed and validated a single, multidimensional digital health metric for individuals with chronic lower back and leg pain using over 190,000 daily-reported data points from 498 patients. The metric identified five distinct symptom clusters that correlated with clinical assessments and patient-reported text messages, providing a more holistic and actionable measure of chronic pain status beyond pain intensity alone.

Background

Chronic pain affects at least 20% of Americans and significantly impacts physical, mental, and social well-being. Traditional clinical trials often rely on infrequent self-reported pain intensity measures, which inadequately capture the complex, multidimensional nature of chronic pain influenced by factors such as sleep, mood, and activity. Advances in digital health and artificial intelligence offer opportunities to integrate diverse symptom data collected longitudinally to better understand and manage chronic pain. This study focused on chronic lower back and leg pain, the most common pain locations, to develop a comprehensive digital health metric using multidimensional data including clinical assessments, patient-reported symptoms, text responses, and smartwatch actigraphy.

Data Highlights

Key Findings

• Unsupervised clustering of multidimensional digital data identified five novel symptom clusters representing ordinal health states from best to worst.
• Clusters correlated significantly with standard clinical assessments of function and quality of life (correlation coefficients ranging from 0.34 to -0.51, p < 0.001).
• Clusters differentiated patient states even when pain intensity was similar, indicating additional dimensions beyond pain magnitude.
• Patient text message content about their status aligned better with cluster assignments than pain reports alone, highlighting the value of qualitative data.
• Incorporation of smartwatch-based actigraphy added objective activity intensity measures to the model, enhancing the multidimensional assessment.

Clinical Implications

This multidimensional digital health metric offers clinicians a more comprehensive and actionable tool to assess chronic back and leg pain beyond traditional pain intensity scales. It enables better monitoring of symptom clusters that impact function and quality of life, potentially guiding personalized treatment strategies. Integration of patient-generated text and objective activity data may improve patient-clinician communication and outcome prediction.

Conclusion

The study successfully established a validated, comprehensive digital measure that captures the complex symptomatology of chronic lower back and leg pain. This metric advances chronic pain assessment by integrating multidimensional data, offering a promising approach for improved clinical management.

References

1. NCT01719055/NCT03240588 -- Clinical trial data source