AB035. Validation of EORTC brain cancer module (EORTC QLQ-BN20) for assessment of health-related quality of life in glioma patients in Singapore.

Background: Existing international data has shown that glioma patients suffer from poorer health-related quality of life (HRQoL). The European Organization for Research and Treatment of Cancer (EORTC) brain cancer-specific Quality of Life Questionnaire (QLQ-BN20) was developed to be together with EORTC Core Quality of Life Questionnaire (QLQ-C30) for cancer patients, highlighting issues particularly relevant to brain tumor patients. It has since been translated and validated across numerous cohorts. However, its psychometric properties have yet to be examined in Singapore. This study aimed to validate the use of QLQ-BN20 in a nationally representative sample of glioma patients in Singapore.

Methods: Eighty-seven patients who had undergone neurosurgery for glioma from six hospitals in Singapore completed three self-reported measures of HRQoL (the EuroQol EQ-5D-5L, EORTC QLQ-C30, and EORTC QLQ-BN20). Descriptive statistics summarized their characteristics and scores on the questionnaires. Psychometric properties of QLQ-BN20 examined included convergent and discriminant validity, internal consistency (Cronbach's alpha), and construct validity (Spearman's correlation). Clinical validity of QLQ-BN20 was determined based on whether QLQ-BN20 scores could differentiate patients with good and poor functional status as measured by Karnofsky Performance Scale and Barthel's Index.

Results: The QLQ-BN20 was demonstrated to have good convergent validity (item-own scale correlation >0.70) and discriminant validity (item-own scale correlation higher than item-other scale correlation). There is high internal consistency, both overall (α=0.88) and within multi-item subscales (α=0.74-0.88). Conceptually similar subscales between different tools were more strongly correlated. For instance, the QLQ-C30 physical functioning subscale and the QLQ-BN20 motor dysfunction subscale (r=-0.65, P<0.001), and the QLQ-C30 cognitive functioning subscale and the QLQ-BN20 cognitive deficits subscale (r=-0.51, P<0.001). QLQ-BN20 was also able to distinguish between functional statuses of patients (P<0.05).

Conclusions: This study supports the validity and reliability of the EORTC QLQ-BN20 among patients with glioma in Singapore. There is good convergent and discriminant validity, internal consistency, construct validity, and clinical validity. The QLQ-BN20 is a valuable supplement to the QLQ-C30. Hence, we recommend expanding its use for all glioma patients and possibly brain cancer patients in Singapore.