Interpreting Within-Patient Changes on the EORTC QLQ-C30 and EORTC QLQ-LC13

Area of Science:

• Oncology
• Psychometrics
• Clinical Trials

Background:

• Statistical significance alone may not reflect meaningful patient benefit in clinical trials.
• Patient-reported outcome (PRO) scores require established thresholds for interpreting individual changes.
• Assessing meaningful within-patient changes is crucial for understanding treatment impact.

Purpose of the Study:

• To evaluate thresholds for meaningful improvements and deteriorations using the EORTC QLQ-C30 and QLQ-LC13 instruments.
• To apply psychometric methods to phase III non-small cell lung cancer trial data.
• To determine appropriate interpretation standards for PRO score changes.

Main Methods:

• Anchor-based methods (empirical cumulative distribution functions, classification statistics) were used.
• Distribution-based methods (one-half standard deviation, standard error of measurement) were applied.
• Change scores from baseline to week 7 were analyzed using the QLQ-C30 Global Health Status item as an anchor.

Main Results:

• Correlations between domain scores and the anchor item were modest (r ≥ 0.30 for 5/24 domains).
• Triangulation of multiple methods identified patterns and proposed thresholds ranging from 11.11 to 33.33.
• Proposed thresholds yielded similar classifications to the traditional 10-point threshold for most EORTC domains.

Conclusions:

• The study confirms the appropriateness of the 10-point EORTC score threshold for interpreting within-patient changes.
• Proposed thresholds enhance interpretability by aligning with observable score scale locations.
• This research provides refined interpretation guidelines for PROs in lung cancer clinical trials.