A Rasch model analysis of two interpretations of ‘not relevant’ responses on the Dermatology Life Quality Index (DLQI)

Abstract Purpose Eight of the ten items Dermatology Life Quality Index (DLQI) have a ‘not relevant’ response (NRR) option. There are two possible ways to interpret NRRs: they may be considered at all’ or missing responses. We aim compare measurement performance DLQI in psoriasis patients when NRRs scored as ‘0’ (hereafter zero-scoring) and ‘missing’ missing-scoring) using Rasch model analysis. Methods Data 425 with from earlier cross-sectional surveys were re-analysed. All completed paper-based Hungarian version DLQI. A partial credit was applied. The following assumptions properties tested: dimensionality, item fit, person reliability, order options differential functioning (DIF). Results Principal component analysis residuals confirmed unidimensional structure Person separation reliability indices similar zero-scoring (0.910) missing-scoring (0.914) NRRs. With zero-scoring, 6 (sport), 7 (working/studying) 9 (sexual difficulties) suffered misfit item-level disordering. missing-scoring, no observed only illogically ordered. Six three showed DIF for gender age, respectively, that reduced four by missing-scoring. Conclusions Missing-scoring resulted an improved scale. scores least one vs. cannot directly compared. Our findings provide further empirical support DLQI-R scoring modification treats replaces them average score relevant items.