Correlation and agreement of self-assessed and objective skin disease severity in a cross-sectional study of patients with acne, psoriasis, and atopic eczema.

Background: Previous studies have shown variable correlation of patients' self-assessed skin severity measures and clinician-assessed objective measures of severity. But, generally, correlation has not been as good as might be expected for conditions in which the objective physical extent of skin disease is apparent to the sufferer to an extent that is not applicable in many other diseases.

Methods: This paper reports agreement and correlation of self-assessed and objective severity measures in a study of 108 subjects with acne, psoriasis, or atopic eczema. The study was a cross-sectional study examining psychological associations of these skin diseases. Objective severity was assessed with the Leeds technique (acne), the Psoriasis Area and Severity Index, and Six Area Six Sign Atopic Dermatitis instruments. Agreement is a more appropriate measure than correlation in this situation and was measured with weighted kappa, while correlation was measured with Spearman's rank correlation.

Results: There was a modest correlation of ρ = 0.46 and similarly very modest agreement of 0.35 (weighted kappa) of self-assessed and clinician-assessed disease severity. Furthermore, self-assessed (but not clinician-assessed) severity was statistically associated with psychological morbidity in this study; i.e. - depression, anxiety, and overall psychological morbidity.

Conclusions: Clinicians should consider psychological sequelae of skin disease, not only in those with objectively more severe disease but in patients across the severity spectrum. Both observational and interventional studies of skin disease should include both clinician-assessed and self-assessed measures of severity among assessed variables.