Network analysis of the RASFF database: a mycotoxin perspective

Abstract

Analysis of the Rapid Alert System for Food and Feed (RASFF) provides useful trend analyses such as identification of transgressor and detector nations, and determination of seasonal variations in contamination patterns. This approach may inform a nation's testing regimes along with trade intelligence to secure safe food supplies. The aim of this investigation was to explore the use of descriptive statistics coupled to network analysis to provide a user friendly approach to allow interrogation of the RASFF database. Categorisation of notifications by contaminant type reveals that some 30% of all notifications are generated owing to mycotoxins with aflatoxins being the major contributor. In the period between January 2008 and October 2010, the order of nations reported for mycotoxin contamination appeared in the order Turkey, China, Iran and USA. Network analysis allows rapid determination of trends in the RASFF database as each contaminant type can be filtered to focus on nations acting as transgressors and detectors. In addition, the impact of each nation can be assigned as the network tool incorporates a consideration of frequency counts as well as number of countries involved. This approach rapidly identifies the key detectors and transgressors, confirming Turkey, China and Iran as key transgressor nations for mycotoxins. Changes over time, during the assessment period, reveal that Iran is improving in the longer term in contrast to Turkey and China where further remedial action is warranted. In summary, a network tool has several advantages over descriptive statistics. It can rapidly identify trends in detector and transgressor nations for each category of contaminant. It can also provide an impact score for each nation providing weekly updates to help identify emerging issues either as nations and/or contaminant type.