Measurement equivalence across gender and education in the WHOQOL-BREF for community-dwelling elderly Taiwanese

Background: The WHOQOL-BREF, a generic quality of life (QoL) instrument, has been widely used clinically and for research on older populations. However, its measurement equivalence/invariance (ME/I) has not been well examined for the elderly (≥ 65 years) across some different demographics. Methods: The data were derived from a cross-sectional study with a convenience sampling design in Taiwan. We enrolled 244 elderly participants: men = 143 (58.6%); educational level ≤ primary school = 121 (49.6%). The ME/I was examined using multiple group confirmatory factor analysis (MGCFA) across gender and educational level. Results: The fit indices were satisfactory for the configural models of gender and educational level (standardized root mean square residual [SRMR] = 0.0742 and 0.0770; root mean square error of approximation [RMSEA] = 0.0655 and 0.0686; comparative fit index [CFI] = 0.953). In addition, MGCFAs showed that ME/I was supported across gender (ΔSRMR = 0.001 to 0.019; ΔRMSEA = -0.003 to 0.001; ΔCFI = -0.003 to 0.000) and educational level (ΔSRMR = 0.002 to 0.006; ΔRMSEA = -0.002 to 0.004; ΔCFI = -0.007 to 0.000). Conclusion: The WHOQOL-BREF Taiwan version is appropriate for combined use and for comparisons in older people across gender and different educational levels.