Physician-Patient Agreement in the Assessment of Allergen Relevance

Abstract

The efficacy of patch testing may be enhanced by data allowing the physician to estimate the likelihood that results of a patch-test reading are relevant to a patient's dermatitis. The goal of this study was to compare the rates of agreement between the physician's assessment of relevance at the time of final reading and the patients' report 3 months to 3 years later in regard to whether avoidance of an allergen was needed to remain free of dermatitis. We hypothesize that the agreement rates between the physician and patient relevance assessments will vary based on properties both intrinsic and extrinsic to the allergen in question. We mailed 407 Institutional Review Board-approved questionnaires to patients and analyzed results for the 92 patients reporting greater than 80% improvement of their dermatitis. Cross-reacting allergens tested on the same patient were combined for analysis. Percent agreement was used to assess interrater concordance. Percent agreement regarding relevance for each allergen or group of allergens was as follows: formaldehyde and formaldehyde-releasing preservatives, 88%; neomycin sulfate, 78%; nickel sulfate hexahydrate, 71%; fragrance mix and related products, 65%; and gold sodium thiosulfate, 56%. Relevance varies between allergens. Physician assessment of relevance at the time of final reading is not the ideal method for determining allergen relevance. This has implications for when best to determine the relevance of certain allergens. For allergens with lower agreement, in particular, extended follow-up is recommended to accurately determine an allergen's contribution to a patient's allergic contact dermatitis, especially in those circumstances in which a patient's condition has not improved.