Assessing how individuals conceptualize numeric pain ratings: validity and reliability of the Pain Schema Inventory (PSI–6) Short Form

Abstract

Background: While numeric scales to represent pain intensity have been well validated, individuals use various conceptualizations when assigning a number to pain intensity, referred to as pain rating schema. The 18-item Pain Schema Inventory (PSI-18) quantifies pain rating schema by asking for numeric values for multiple mild, moderate or severe pain conditions. This study aimed to assess the validity and reliability of a shortened form of the PSI, using only 6 items (PSI-6). Methods: A secondary analysis was performed on two existing datasets. The first (n = 641) involved a community-based population that completed the PSI-18. The second (n = 182) included participants with chronic pain who completed the PSI-6 twice, one week apart. We assessed face validity, convergent validity, offset biases, test-retest reliability, and internal consistency of the PSI-6 compared to the PSI-18. Results: Both the PSI-18 and PSI-6 demonstrated excellent face validity. The PSI-6 demonstrated excellent convergent validity relative to the PSI-18, with correlations from r = 0.88 to 0.92. Bland-Altman plots revealed offset biases near zero (< 0.22 on 0–10 scale) across all categories of mild, moderate, severe and average pain. Internal consistency was excellent, with Cronbach's Alpha = 0.91 and 0.80, for PSI-18 and PSI-6 respectively. Test-retest reliability of the PSI-6 was high with correlations from r = 0.70–0.76. Conclusion: The PSI-6 is a valid and reliable tool to assess pain rating schema with reduced subject burden, to better interpret individuals’ pain ratings and adjust for inter-individual variability.