Abstract
Purpose:
To estimate glaucoma utility values from a Discrete Choice Experimental (DCE) questionnaire set and test the theoretical and convergent validity of this new glaucoma utility index developed in Singapore (GUI-S).
Methods:
The GUI-S had seven validated dimensions namely central and near vision; lighting and glare; mobility; activities of daily living; eye discomfort; other glaucoma effects; and psychosocial effects. A preference based efficiently constructed experimental design for the 7 dimensions with three levels of difficulty (no difficulty; some difficulty and severe difficulty), was used to elicit responses from glaucoma patients at a tertiary eye hospital in Singapore. A conditional logistic regression was used to estimate the coefficients, and utility weights (GUI-S) were derived for a given glaucoma profile. A-priori coefficient movements were hypothesized to correlate with disease severity and a generic VAS (Visual Analogue Scale) was used to determine the theoretical and convergent validity parameters of the GUI-S, respectively.
Results:
70 primary glaucoma patients (72.9% with primary open angle) responses were collected: mean age 68.2 years (±9.1), female (27.1%), glaucoma severity; (48.6% mild, 34.3% moderate, 17.1% severe); and the majority were Chinese (94.3%). The significant estimated utility values comparing ‘no difficulty’ with ‘some/severe difficulty’ in each of the dimensions were: central near vision (0.13, p=0.001), lighting and glare (0.20, p<0.001), mobility (0.32, p<0.001), psychosocial dimensions (0.19, p<0.001). GUI-S values were negatively related with glaucoma severity (p= 0.04): 0.52 (± 0.28) for mild; 0.40 (±0.21) for moderate; 0.31 (±0.30) for severe. VAS values were also negatively associated with glaucoma severity (ANOVA, p=0.003): 0.78 (±0.18) for mild; 0.75 (± 0.24) for moderate; 0.52 (± 0.27) for severe.
Conclusions:
GUI-S, based on responses elicited from an efficiently designed DCE questionnaire, showed sufficient convergent and theoretical validity with a generic health outcome (VAS) and glaucoma disease severity, respectively.
Keywords: 460 clinical (human) or epidemiologic studies: health care delivery/economics/manpower •
468 clinical research methodology