Item Reduction and Psychometric Validation of the Oily Skin Self Assessment Scale (OSSAS) and the Oily Skin Impact Scale (OSIS)

Introduction: Developed using focus groups, the Oily Skin Self Assessment Scale (OSSAS) and Oily Skin Impact Scale (OSIS) are patient-reported outcome measures of oily facial skin. Objective: The aim of this study was to finalize the item-scale structure of the instruments and perform psychometric validation in adults with self-reported oily facial skin. Methods: The OSSAS and OSIS were administered to 202 adult subjects with oily facial skin in the United States. A subgroup of 152 subjects returned, 4 to 10 days later, for test-retest reliability evaluation. Results: Of the 202 participants, 72.8% were female; 64.4% had self-reported nonsevere acne. Item reduction resulted in a 14-item OSSAS with Sensation (five items), Tactile (four items) and Visual (four items) domains, a single blotting item, and an overall oiliness item. The OSIS was reduced to two three-item domains assessing Annoyance and Self-Image. Confirmatory factor analysis supported the construct validity of the final item-scale structures. The OSSAS and OSIS scales had acceptable item convergent validity (item-scale correlations > 0.40) and floor and ceiling effects (< 20%). Cronbach's alpha coefficients ranged from 0.83 to 0.89 for the OSSAS and 0.82 to 0.87 for the OSIS, demonstrating excellent internal consistency. The a priori test-retest reliability criterion (intraclass correlation [ICC] >= 0.7) was met for one of the three OSSAS domains and one of the two OSIS domains. OSSAS and OSIS domains distinguished among groups that differed in patient-reported facial oily skin severity (P < 0.0001), and bother associated with oily skin (P < 0.0001). Conclusions: The OSSAS and OSIS versions tested in this study have been found to have strong psychometric properties in this patient sample (adults with self-reported oily facial skin), as assessments of self-reported oily facial skin severity and its emotional impact, respectively.