PS07 (P107) Updating the Assessment of the Psychological and Social Effects of Acne (APSEA) through factor analysis: developing the novel APSEA-3

Georgina Jones,Jim Morgan,Eva Lau,Moe Latt,Natasha Lee,Ibtessam El-Mansori,Alison Layton
DOI: https://doi.org/10.1093/bjd/ljae090.378
IF: 11.113
2024-06-28
British Journal of Dermatology
Abstract:Abstract Health-related quality of life is now recognized as an essential outcome to assess in acne trials and practice. Many patient-reported outcomes exist, including the Assessment of the Psychological and Social Effects of Acne (APSEA), which is an acne-specific measure (Layton AM, Seukeran D, Cunliffe WJ. Scarred for life? Dermatology 1997; 195: 15–21). APSEA was developed in 1990s and subsequently psychometric standards and methods have progressed. The purpose of this study is to conduct an updated psychometric analysis of the APSEA among a cohort of patients being treated for acne in routine clinical practice. Exploratory factor analysis (EFA), a method of classical test construction, was carried out (n = 993). To test the adequacy of the sample size for undertaking factor analysis, values for the Kaiser–Meyer–Olkin test and Bartlett’s test of sphericity were calculated, as well as Cronbach’s alpha. Following the EFA and generation of the proposed scales, the individual domains were calculated and transformed onto a scale of 0–100 by summing the items within the scale, dividing by the total possible score within that domain. A confirmatory factor analysis (CFA) was then undertaken using a new second dataset (n = 501) to test the factor conceptual framework, using MPlus 8.6 software (http://www.statmodel.com). Descriptive statistics and tests of internal consistency reliability and construct validity (concurrent) were also undertaken using a priori hypotheses and comparisons with the Dermatology Life Quality Index (DLQI) and Patient Health Questionniare-9 (PHQ-9). The EFA suggested a three-domain model (anxiety and depression, body image and self-esteem, and social functioning), which was confirmed in the CFA analysis whereby the goodness of fit indices all indicated good model fit (root mean square error of approximation = 0.038, comparative fit index = 0.96, Tucker–Lewis index = 0.96, standardized root mean squared residual = 0.048). Cronbach’s alpha values exceeded 0.70, and item-total correlations exceeded 0.40 for all three domains, indicating good internal consistency reliability. All three APSEA domains were also positively correlated with the DLQI and the PHQ-9, indicating good construct validity of the tool. This EFA study has demonstrated the validity and reliability of a novel version of APSEA, which we have termed APSEA-3 to reflect the three unique domains it captures. Preliminary results show that the tool is responsive to change and useful for measuring the psychosocial impact of acne.
dermatology
What problem does this paper attempt to address?