Related Experiment Video
Updated: May 16, 2026
Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
Mapping EORTC QLQ-C30 onto EQ-5D for the assessment of cancer patients
Seon Ha Kim1, Min-Woo Jo, Hwa-Jung Kim
1Department of Preventive Medicine, University of Ulsan College of Medicine, Songpa-gu, Seoul, Korea.
Background:
The European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) is the instrument most frequently used to measure quality of life in cancer patients, whereas the EQ-5D is widely used to measure and evaluate general health status. Although the EORTC QLQ-C30 has been mapped to EQ-5D utilities, those studies were limited to patients with a single type of cancer. The present study aimed to develop a mapping relationship between the EORTC QLQ-C30 and EQ-5D-based utility values at the individual level.
Methods:
The model was derived using patients with different types of cancer who were receiving chemotherapy. The external validation set comprised outpatients with colon cancer. Ordinary least squares regression was used to estimate the EQ-5D index from the EORTC QLQ-C30 results. The predictability, goodness of fit, and signs of the estimated coefficients of the model were assessed. Predictive ability was determined by calculating the mean absolute error, the estimated proportions with absolute errors > 0.05 and > 0.1, and the root-mean-squared error (RMSE).
Results:
A model that included global health, physical, role, emotional functions, and pain was optimal, with a mean absolute error of 0.069 and an RMSE of 0.095 (normalized RMSE, 8.1%). The explanatory power of this model was 51.6%. The mean absolute error was higher for modeled patients in poor health.
Conclusions:
This mapping algorithm enabled the EORTC QLQ-C30 to be converted to the EQ-5D utility index to assess cancer patients in Korea.
