Equivalence and measurement properties of an electronic version of the Psoriasis Symptom Inventory

Abstract

To evaluate the equivalence of electronic and paper versions of the Psoriasis Symptom Inventory and to examine measurement properties of the electronic version. In a prospective, randomized, crossover, non-interventional study in adult subjects (age ≥18years) with plaque psoriasis conducted over a period of 15days, subjects were randomized to two groups, completing either the paper or electronic Psoriasis Symptom Inventory daily for 7 consecutive days followed by the alternate version. Equivalence was assessed by the intraclass correlation coefficient (ICC) between both administration modes. Differences in scores were also tested using paired Student's t test. Measurement properties included internal consistency reliability, test-retest reliability, and convergent and discriminant validity between the Psoriasis Symptom Inventory and (1) disease-specific (Dermatology Life Quality Index) and (2) general health (SF-36v2) status. Eighty subjects [74% (59/80) moderate-to-severe psoriasis; 26% (21/80) mild psoriasis receiving systemic treatment] were enrolled from 8 sites in the USA. The two modes were highly concordant for both total (ICC=0.97) and individual item scores (ICC range=0.93-0.97). Response bias testing showed no differences based on completion order with all ICC values >0.91. All mean score differences, except for one item ("flaking"), were non-significant (P>0.05). Minimum values for reliability (>0.70) and validity (convergent, r≥0.40) were exceeded for the electronic Psoriasis Symptom Inventory. Equivalence between paper and electronic versions of the Psoriasis Symptom Inventory and strong measurement properties of the electronic mode indicated a successful migration from paper to electronic format of the Psoriasis Symptom Inventory.