Rasch analysis holds no brief for the use of the Dermatology Life Quality Index (DLQI) in Chinese neurodermatitis patients

Abstract

Background: The Dermatology Life Quality Index (DLQI) is the most widely used measure of health-related quality of life (HRQoL) associated with skin disease. Recently, the psychometric properties of the DLQI have caused some controversy because the instrument appears not to meet the requirements of modern test theory. The purpose of this study was to assess whether these psychometric issues also occur in Chinese patients with neurodermatitis. Methods: One hundred fifty consecutive outpatients (83 males and 67 females) seeking treatment for neurodermatitis were assessed for eligibility for this prospective study between July 1, 2011 and September 30, 2011. The DLQI and a demographic questionnaire were completed. One female participant who incompletely answered the DLQI was excluded. Data were analyzed using the Rasch model in order to obtain meaningful scores for the DLQI. Scale assessment included analysis of rating scale function, item fit to the Rasch model, aspects of person-response validity, unidimensionality, person-separation reliability, and differential item function. Results: The rating scale advanced monotonically for all items in the DLQI, but item 9 did not demonstrate acceptable goodness-of-fit (Infit MnSq values >1.3) to the Rasch model. The 10 items of the DLQI met the criteria for person-separation reliability (PSI=2.38) and the first latent dimension (general QoL) accounted for 50.8% of the variance; but the variance explained by the second dimension (7.1%) exceeded the criterion of 5%. There were also limitations related to person-response validity, because>5% (18.1%) of cases demonstrated unacceptable fit. There was no uniform differential item functioning. Conclusions: For neurodermatitis patients, the DLQI seems to have poor fit to the Rasch model; therefore, we recommend against using this instrument with neurodermatitis patients.