Feasibility of applying the 24HR method for dietary data collection by computerized and standardized platform - GloboDiet in epidemiological studies

Author(s)

Gustavo Rosa Gentil Andrade

• Document type: Master's Dissertation
• Press: São Paulo.
• Institution: Universidade de São Paulo (USP). Faculdade de Saúde Pública (FSP/CIR)
• Defense date: 2020-08-12

Examining board members

Dirce Maria Lobo Marchioni; Paulo Rogério Melo Rodrigues

• Advisor: Dirce Maria Lobo Marchioni

Abstract

Introduction. Technological advances have enabled the development of software that computerized and standardized the instruments for collecting dietary data, such as the 24-hour recall. Objective. Evaluate the feasibility and applicability of the Brazilian version of the GloboDiet software, in two measurements, in the routine of a national epidemiological study. Methods. A Longitudinal Study of Adult Health (ELSA-Brasil) evaluated at the Research Center of São Paulo-SP was applied to participants, 24HR guided by the GloboDiet software, in person (n = 2,093) and, in a subsample (n = 1,084), a second measure predominantly by telephone. The adherence rates were captured from the interview control spreadsheets completed by the interviewers and the Chi-square test was used to study the differences between the groups (gender and age group) and the final adherence status of the participants. For the duration of the interviews, data were presented as median and interquartile range (IQR) Q1 - Q3 (25-75%), and from the Mann-Whitney, Kruskal-Wallis and Wilcoxon tests, the differences between the modalities in person and by telephone and between the categories of sex, age and BMI for the total duration and stages of the GloboDiet interview. To assess the applicability of GloboDiet by the interviewers (n = 8), the focus group technique combined with Content Analysis was applied. To assess the applicability of the software by the interviewees, a questionnaire was sent by e-mail and employees used descriptive statistics procedures and the Chi-square and Fisher\'s Exact tests to study the differences between sex and age group and the answers provided. Results. adherence rates were 71% and 50% in face-to-face and predominantly telephone interviews, respectively. The total duration of face-to-face and telephone interviews was 30.7 (interquartile range [IQR], 23.3-40.7) and 35.3 (IQR, 25.3-49.7) minutes (p <0.001). As for the focus group, five main themes emerged during the discussion, pointing out that GloboDiet has potentialities related to its level of detail and ability to guide the interviewer during the collection, and limitations in the training stage and in the recipe database. In the questionnaire sent to the interviewees, a response rate of 18.6% (n = 540) was obtained, and the results indicated a good acceptance of the software, the interview time and the photographic manual by the interviewees, and the preference for face-to-face interview format. Conclusion. From the results found in this study, it was possible to show that the GloboDiet software is viable to the routine of an epidemiological study and acceptable to the interviewees and interviewers. (AU)

• FAPESP's process: 18/14868-6 - Feasibility of the application of the 24HR method for the collection of dietary data by computerized and standardized platform: GloboDiet in a cohort study
• Grantee: Gustavo Rosa Gentil Andrade
• Support Opportunities: Scholarships in Brazil - Master