Although it has had only limited psychometric assessment in one country (the UK), the Psoriasis Disability Index (PDI) is a commonly used measure of the impact of psoriasis on patients. This study's objective was to analyze the psychometric properties of the PDI in 1196 US patients. High Cronbach's alpha coefficients suggested that the PDI's internal consistency is good. The validity of the PDI was tested using a global question on disease burden and self-assessed extent of disease and both appeared to be good predictors of the PDI. Large floor effects and the suboptimal response distribution of most items, however, suggested that the PDI is insensitive to mild functional limitation. Factor analyses indicated that the current PDI subscales are suboptimal. In the Rasch analysis, the PDI and its subscales appeared to measure multiple constructs, making the validity of deriving a single overall score questionable. The frequent presence of differential item functioning related to several patient characteristics confirmed the instrument's multidimensionality. These findings suggest that the PDI is not an optimal measure for use in US study populations. The psychometric properties of instruments designed to measure the impact of psoriasis should be tested in populations in which the instrument is to be applied.
