Rasch analysis of the WHOQOL-BREF in post polio syndrome.

Abstract

OBJECTIVE: The World Health Organisation quality of life abbreviated scale (WHOQOL-BREF) was developed as a measure of quality of life across 4 separate health domains; physical health, psychological, social relationships and environment. This study evaluated the validity of the WHOQOL-BREF in post-polio syndrome by testing it for fit against the Rasch model. RESULTS: The scale was posted to 319 volunteers, 271 (85%) completed the scale with a mean age of 66.7 years (standard deviation 8.15); 64% were female. The social relationships domain fitted the Rasch model (_2 p_=_0.19) but reliability was low (__=_0.69) and there were insufficient items to test the assumption of unidimensionality. Solutions were derived for physical health (p_=_0.45, t-test_=_1.5%, __=_0.67), psychological (p_=_0.19, t-test_=_4.9%, __=_0.78) and environment domains (p_=_0.48, t-test_=_6.0% - lower confidence interval 3.4%, __=_0.80) by accounting for local dependence and to cancel out differential item functioning. An overall measure of quality of life, which combined all 4 domains was validated (p_=_0.80, t-test_=_4.6%, __=_0.81). A transformation table for this total score is provided. CONCLUSION: The 4 domains of the WHOQOL-BREF provide valid measures of quality of life in post-polio syndrome. The summed score was more reliable and better targeted and can be used as an ordinal estimate of quality of life.