Introduction of the Reliable Estimation of Atopic Dermatitis in ChildHood: Novel, Diagnostic Criteria for Childhood Atopic Dermatitis

PURPOSE: Questionnaire-based diagnostic criteria for atopic dermatitis (AD) have been proposed to detect the major group of AD with flexural dermatitis. We aimed to develop novel, questionnaire-based diagnostic criteria for childhood AD, which can detect more comprehensive AD including non-flexural type. METHODS: The draft version of questionnaire to detect childhood AD was prepared to be used for preliminary hospital- (n=1,756) and community-based (n=1,320) surveys. From analysis, the Reliable Estimation of Atopic dermatitis of ChildHood (REACH) was derived and verified in derivation (n=1,129) and validation (n=1,191) sets by community-based surveys. RESULTS: The REACH consists of 11 questions including 2 major and 9 minor criteria. AD is diagnosed as the major group of 'eczema on the antecubital or popliteal fossa' to fulfill the 2 major criteria (2M), and the minor group of 'eczema on the non-antecubital or popliteal fossa' to fulfill the 1 major plus 4 or more minor criteria (1M+4m). In the validation set, the overall 1-year AD prevalence by the REACH was estimated as 12.3% (95% CI, 10.5%-14.2%), and the REACH showed a sensitivity of 75.2%, a specificity of 96.1%, and an error rate of 6.4%. The REACH demonstrated better diagnostic performance than the ISAAC in terms of the number of misclassification (10.0%). CONCLUSIONS: We propose the REACH as new full, questionnaire-based diagnostic criteria for childhood AD in epidemiological surveys. Further studies are warranted to validate the REACH in different populations or countries in the context of large-scale, epidemiological surveys.

Desenvolvimento e pré-teste de um questionário de frequência alimentar para graduandos / Development and pretesting of a food frequency questionnaire for undergraduate students

Objetivo Desenvolver e aplicar em um projeto-piloto um questionário de frequência alimentar quantitativo de autopreenchimento destinado a graduandos da área da saúde. Métodos Trata-se de um estudo transversal realizado em 151 universitários de ambos os sexos, usuários do ambulatório de Nutrição do Corpo Discente da Universidade Federal de São Paulo, Brasil. O questionário inicial foi composto a partir dos alimentos e preparações informados no Registro Alimentar de Três Dias. As informações em medidas caseiras foram transformadas em gramas ou mililitros com o auxílio de tabelas, e os alimentos foram ordenados segundo a porcentagem de contribuição para o valor energético total informado. Foram selecionados 198 ali-mentos responsáveis por 95 por cento do consumo energético e agrupados em 77 itens alimentares de acordo com a similaridade nutricional. O tamanho das porções alimentares foi classificado conforme o valor do percentil 50 da distribuição dos pesos correspondentes às medidas caseiras referidas. Definiu-se como porção pequena, aquela cujo valor foi igual ou inferior ao percentil 25; como média, o percentil 50 e grande, o valor igual ou superior ao percentil 75. Resultados Após o pré-teste do questionário inicial, o questionário final resultou em uma lista com 89 alimentos, agrupados em 70 itens alimentares. As instruções para o autopreenchimento foram refeitas, objetivando-se um melhor preenchimento. Conclusão Após as modificações realizadas, o questionário de frequência alimentar quantitativo encontra-se pronto para o estudo de validação e calibração.

Objective The objective of this study was to develop and pretest a self-administered, quantitative food frequency questionnaire for undergraduate students of the health sciences. Methods This is a cross-sectional study with 151 undergraduate students of both genders from the Students' Healthcare Service of the Universidade Federal de São Paulo, Brazil. The first version of the food frequency questionnaire was based on the foods and preparations listed in the 3-day food records. Tables were used to convert cooking units into grams or milliliters and the foods were ranked according to their percentage contribution to the total calorie intake. The 198 foods responsible for 95 percent of the energy intake were selected and grouped into 77 food items according to nutritional resemblance. Serving size was classified according to the 50th percentile of the weight distribution corresponding to the cooking units. Serving size was defined as small (S) when £25th percentile; average (A) when equal to the 50th percentile and large (L) when ≥75th percentile. Results The pretest of the first version of the questionnaire resulted in a final version with a list of 89 foods, grouped into 70 food groups. The instructions for filling out the questionnaire were rewritten to get better answers from the students. Conclusion After these changes, the quantitative food frequency questionnaire is ready for the validation and calibration study.