[Validity and reliability of the 8-item EUROHIS-QOL to assess Brazilian adults' quality of life].
Samuel Salvi RomeroTonantzin Ribeiro GonçalvesCândido Norberto Bronzoni de MattosFernanda de Souza BairrosMarcos Pascoal PattussiPublished in: Cadernos de saude publica (2022)
The study evaluated the psychometric properties of the EUROHIS-QOL 8-item scale in Brazilian adults. The EUROHIS-QOL is a measurement of quality of life developed based on the generic instruments of WHOQOL-100 and WHOQOL-BREF. Based on data from a prospective cohort of 1,100 adults, participants in the baseline in 2006 and 2007, exploratory factor analysis, internal consistency, reliability, and discriminant validity were performed via multitrait-multimethod analysis, with the calculation of the average variance extracted (AVE). Confirmatory factor analysis was conducted with 573 of these participants in 2013 to 2018. Solutions were tested with one and with two factors named functional capacity and socioeconomic conditions. Both solutions presented good internal consistency and reliability. The correlation between the items was 0.535, and the AVE was 0.397 for solution with one factor, and 0.528 and 0.341 for factors one and two, respectively, indicating good discriminant validity in the bifactorial solution. Both factors had eigenvalues greater than one and factor loadings ranging from 0.398 to 0.915. The adjustment of the unifactorial solution presented: χ2 = 186 (g.l. = 18), p < 0.001, RMSEA = 0.128 (90%CI: 0.111-0.145), CFI = 0.960, TLI = 0.938, and SMRM = 0.042; while in the bifactorial presented: χ2 = 135 (g.l. = 17), p < 0.001, RMSEA = 0.110 (90%CI: 0.093-0.128), CFI = 0.972, TLI = 0.954, and SMRM = 0.035. Overall, the results provide reasonable evidence for the construct validity and reliability of the EUROHIS-QOL 8-item. Instruments with fewer items, as in our case, provide greater practicality and are ideal for use in population-based studies with extensive samples.
Keyphrases