Predicting treatment response in patients with acute or chronic pain using a composite scale derived from multiple patient–reported outcome measures

抄録

The aim of this study was to investigate the most important aspects of patient–reported outcome measures (PROMs) for pain assessment and their predictive value in treatment response. We analyzed 367 patients with acute–to–chronic pain (age: ≥ 20 years) who visited the pain clinic of a university hospital for the first time and responded to nine PROMs related to pain symptoms. Principal component analysis (PCA) was applied to the PROMs data, deriving principal components with their component scores (i.e., a composite scale). Data from 175 patients who completed the same PROMs three months after their initial visit were then analyzed. Multiple regression was used to examine whether these principal components, along with age, sex, and pain chronicity, could predict changes in pain intensity, as measured by a numerical rating scale (NRS) over three months. In addition, we stratified the patients based on pain chronicity and categorized pain as primary, neuropathic, or musculoskeletal. We also examined the relative risk associated with the principal components derived from PCA and other variables for clinically significant improvement in pain intensity, defined as a reduction of at least two points on the NRS. We identified four principal components (pain intensity, pain quality, disability, and cognition ⁄ affect for pain) for the clinical assessment of patients with pain. At the initial visit, pain intensity, pain–related cognition ⁄ affect, and pain chronicity were predictors of pain management difficulties. Pain quality was a specific predictor for primary and musculo­skeletal conditions. Our findings provide evidence for the use of PROMs in pain management.