Food safety incidents in Beijing: occurrence patterns, causes and wider social implications

Food safety incidents have the potential to result in a range of adverse health effects, including diarrhoeal diseases and various forms of cancer. Furthermore, they also have a disruptive effect on the economy, trade and tourism. As Beijing strives to become a global centre of politics, economy and culture, serious food safety incidents continue to be frequently reported in the city. It is clear that a large number of food safety incidents, such as those that have occurred in China, and in some other parts of the world, cannot be fully tackled by means of conventional analysis, which focuses on biological and chemical factors, but which often ignores potential human factors (which may be intentional in nature). In this article, we dynamically examine the time intervals between successive distinct food safety incidents that were mainly caused by human factors. Our intention is to identify information that could be of use to governmental and other bodies in efforts to curb the occurrence of food safety incidents. We analyse data spanning a 10-year period from 2004 to 2013, during which time 295 food safety incidents occurred in Beijing. We find that the occurrence of food safety incidents was drastically different from Poisson processes, suggesting that their causes may have been systemic in nature. We further found that the sequence of time intervals had persistent long-range correlations, characterized by a Hurst parameter of H=0.65, suggesting that food safety incidents occurred in bursts. Finally, we propose scenarios that may be responsible for these long-range correlations. Our quantitative findings may not only hold in Beijing and elsewhere in China, but also in other parts of the world where human factors are an important cause of food safety incidents. In all those situations, our proposed scenarios for long-range correlations may be used as part of efforts to devise strategies to curb or prevent future food safety incidents.