The McGill Quality of Life Questionnaire-Revised (MQOL-R). Psychometric properties and validation of a Brazilian version on palliative care patients: a cross-sectional study

Abstract

Abstract Background To assess the psychometric properties, including internal consistency, construct validity, criterion validity, criterion-group validity, and responsiveness, the Reviewed McGill Quality of Life Questionnaire (MQOL-R), into Brazilian Portuguese-(BrP). Also, to analyze the relationship of the BrP-MQOL-R with the scores on the Karnofsky Performance Scale (KPS) and on the Numerical Pain Scale (NPS 0–10). Methods The BrP-MQOL-R was administered to a sample of 146 adults (men = 78). A team of experts translated the MQOL-R according to international guidelines. Convergent validity and Confirmatory factor analysis (CFA) was performed. Results The BrP-MQOL-R Cronbach’s alpha was 0.85. CFA supported the original four-factor structure, with the following revised model fit-indices: PCLOSE = 0.131, Tucker-Lewis Index (TLI) rho 2 = 0.918, incremental fit index (IFI) delta 2 = 0.936. The convergence validity is supported by a significant correlation between BrP-MQOL-R total scores and their subscales with KPS and with the single item related to the quality of life. And by a converse correlation with the pain scores in the NPS (0–10). Receiver operator characteristics (ROC) analysis showed subjects with KPS equal to or lower than 30% could be discriminated from those with scores on KPS higher than 30% by an area under the curve (AUC) = 0.71, sensitivity = 97%, and specificity = 92%). Conclusion The BrP-MQOL-R proves to be a reliable instrument for assessing the quality of life (QOL) in palliative care (PC), with primary evidence of validity. BrP-MQOL-R presented adequate discriminate properties to identify distinct conditions that impact the QOL in PC.