Presentation (CTI)

OBJECTIVES: Patient-reported outcome measures (PROMs) have gained increasing importance in value-based healthcare over the past decades. However, achieving high compliance levels with PROM completion remains a persistent challenge in clinical practice. One promising strategy to enhance compliance is item reduction. For the Boston Carpal Tunnel Questionnaire (BCTQ), a shortened version using a decision tree approach based on Chi-squared Automated Interaction Detection (CHAID) has reduced the number of questions from nineteen to six, while maintaining its psychometric properties in a validation dataset. This study aimed to evaluate the inter-version reliability and agreement between the decision tree BCTQ (DT-BCTQ) and the original BCTQ in a real-world clinical environment.
METHODS: We conducted a test-retest study to evaluate the inter-version reliability and agreement between the DT-BCTQ and the original BCTQ, with a 5- to 10-day interval. We calculated the intraclass correlation coefficients (ICC) to assess the inter-version reliability. Mean differences, standard deviations, and the standard error of measurement (SEM) were calculated to determine the level of agreement. Additionally, we compared the SEM values to the minimally important change (MIC) of the original BCTQ.
RESULTS: Inter-version reliability and agreement were high across all scores: total score (ICC = 0.82, 95% CI: 0.78-0.85; SEM = 0.29), SSS (ICC = 0.75, 95% CI: 0.71-0.79; SEM = 0.35), and FSS (ICC = 0.78, 95% CI: 0.74-0.81; SEM = 0.38). All SEM values were below the MIC of the original BCTQ.
CONCLUSIONS: We found that the inter-version reliability and agreement between the original BCTQ and the DT-BCTQ are high in a real-world clinical setting. These findings suggest that the DT-BCTQ is a suitable alternative to the original BCTQ, reducing its length from nineteen to six questions while maintaining both sub-scores and lowering patients’ response burden and fatigue.