Adaptación y confiabilidad inter-observador de un protocolo para evaluar la disponibilidad de alimentos en supermercados en Buenos Aires, Argentina

BACKGROUND: The International Network for Food and Obesity / Non-communicable Diseases (NCDs) Research, Monitoring and Action Support (INFORMAS) developed the Protocol to evaluate Food Availability in Supermarkets. OBJECTIVE: To describe the adaptation process of the protocol for its use in the City of Buenos Aires (CABA), to compare the indicators of the original protocol to a reference indicator and to evaluate inter-observer variability when applying it in supermarkets in the city. METHODOLOGY: The main indicator of the protocol is the relative availability of healthy (AS) versus unhealthy (ANS) foods, calculated as the ratio between the length (m) of shelves assigned to AS and ANS (length AS/ANS). The selection of foods to be included in the indicator was adapted for its use in CABA. To explore the performance of the indicator, a reference frame was built, measuring shelf length for all the foods and beverages offered in 5 supermarkets, which were classified as AS and ANS according to the Argentine dietary guidelines. The indicator was compared with the relative availability calculated from the reference. To evaluate inter-observer reliability, two observers carried out measurements in three supermarkets and the intra-class correlation coefficient (ICC) was calculated. RESULTS: According to the reference, shelf length AS/ANS varied between 0.16 and 0.61, with a mean of 0.34 (SD 0.18). The adapted indicator produced similar results with a mean difference of -0.05 (SD 0.04). The ICC between the measurements of both observers was 0.92 (95% CI 0.86-0.98). CONCLUSION: The protocol was adapted to be applied in CABA, with modifications in the foods to be evaluated and adequate inter-observer reliability.