Limitations of Using Questionnaires for Assessing the Prevalence of Psoriasis and Atopic Dermatitis Among Adults.

Importance: Questionnaire studies are important for estimating the prevalence of psoriasis and atopic dermatitis; however, validity among the adult population remains an important concern.

Objective: To examine the test-retest accuracy of questionnaires for measuring psoriasis and atopic dermatitis prevalence in an adult population.

Design, Setting, And Participants: This nationwide population-based cohort study administered questionnaires on psoriasis and atopic dermatitis to the same 2333 and 2312 randomly selected adults (≥18 years) in Denmark, respectively, at 2 different time points from May 15, 2018, to November 20, 2020. Data were analyzed from January 10 to January 28, 2021. To reduce the risk of participation bias, potential respondents were given information on the research project only after agreeing to participate.

Exposures: Participants were asked identical questions on psoriasis and atopic dermatitis in 2018 and in 2020. Responses were linked at the individual-level.

Main Outcomes And Measures: The test-retest reliability (expressed by Cohen κ).

Results: The psoriasis questionnaire was completed by 2333 (mean [SD] age, 55.1 [16.2] years; 1338 [57.4%] women) participants in 2018 and in 2020. The atopic dermatitis questionnaire was completed by 2312 (mean [SD] age, 55.0 [16.2] years; 1326 [57.4%] women) participants in 2018 and in 2020. Among participants reporting a history of psoriasis, agreement between individual responses was high (κ = 0.7558); however, among those reporting a history of atopic dermatitis, agreement was low (κ = 0.4395). For psoriasis, prevalence changed from 7.8% to 8.0%; for atopic dermatitis, from 8.2% to 7.6%. Of participants who in 2018 reported dermatologist-diagnosed atopic dermatitis, 36.9% claimed in the 2020 questionnaire that they had never had atopic dermatitis. Analyses revealed substantial agreement for psoriasis responses across all age strata; for atopic dermatitis responses, the κ declined with increasing age, to 0.2613 for participants 65 or older.

Conclusions And Relevance: This cohort study found considerable agreement between responses over time when participants were asked about a history of psoriasis. When asked about a history of atopic dermatitis, responses over time were inconsistent. This inconsistency suggests that questionnaires on a history of atopic dermatitis will confer considerable risk of bias and misclassification.