Turning Patients' Open-Ended Narratives of Chronic Pain Into Quantitative Measures: Natural Language Processing Study

Area of Science:

• Natural Language Processing (NLP)
• Computational Linguistics
• Digital Health

Background:

• Subjective patient reports are crucial for assessing chronic pain and treatment response.
• Understanding personal narratives offers accurate insights into the lived experience of illness.

Purpose of the Study:

• To apply natural language processing (NLP) tools to derive quantitative measures from patient pain narratives.
• To assess the ability of NLP to extract meaningful data from patient interviews in a pilot study of 20 chronic low back pain (CLBP) patients.

Main Methods:

• Collected patient narratives through semistructured interviews focusing on general life and CLBP experiences.
• Utilized NLP to analyze semantic relationships between patient utterances and anchor sentences from validated pain questionnaires.
• Correlated NLP-derived measures with established pain assessment tools like the numerical rating scale (NRS) and visual analog scale (VAS).

Main Results:

• Patient narratives showed semantic proximity to validated pain questionnaire concepts.
• Semantic distances correlated strongly with CLBP intensity reported on NRS and VAS.
• Significant differences in semantic measures were observed between patients with low and high pain levels.

Conclusions:

• Preliminary evidence suggests NLP can quantify the subjective experience of chronic pain.
• NLP holds potential for novel clinical applications in pain assessment.
• Further research is needed to validate these findings in larger, diverse populations.