Abstract

Introduction: Minimal important change estimates (MIC) are useful for interpreting results of clinical research with quality of life (QoL) as an endpoint. For the European Organisation for Research and Treatment of Cancer head and neck cancer module, the EORTC QLQ-HN43, no such thresholds are established. Methods: Head and neck cancer patients under active treatment (n = 503) from 15 countries completed the EORTC QLQ-HN43 three times (t1: before treatment, t2: three months after t1, t3: six months after t1). A subgroup completed a Subjective Significance Questionnaire (SSQ), indicating experienced change from the previous time point in four QoL domains. QoL was assumed to deteriorate after t1 and improve again until t3. The MIC was established using the average of mean differences in SSQ groups (MICmean) and estimates based on logistic regressions (MICpredict). Additionally, minimal detectable changes (MDC) were computed using 0.5 standard deviation and standard error of the mean. Results: For swallowing, speech, dry mouth, and global QoL, the MIC for deterioration were 13, 14, 26, and 10 respectively. The MIC for improvement were 8 (swallowing), 6 (dry mouth), and 5 (global QoL); no MIC for speech improvement can be presented because of insufficient correlation between change score and anchor. The MDC estimates for deterioration were 15, 14, 15, and 11. For improvement, the MDC estimates were 13, 14, 14, and 11. Conclusions: Our results underline that no single MIC or MDC can be applied to all EORTC QLQ-HN43 scales, and that the MIC for deterioration seems larger than those for improvement. © 2024 The Authors