On July 19, 1992, we were notified by the director of the Tabuk Health Region about an outbreak of food poisoning among workers in an agricultural company in the northwestern part of Saudi Arabia. The company employs 1,100 workers of many nationalities who are provided with full accommodation, including meals native to their home countries. Similar outbreaks of apparent food poisoning have occurred at this camp twice before.
Methodology
We reviewed the clinic register and visited the kitchen, work sites and housing facilities. We defined a case of food poisoning as any person who had eaten the Indian supper on July 19, 1992, and had abdominal cramps or diarrhea from four to 48 hours after eating. Because only workers who had eaten the Indian meal were affected, we asked all Indian workers which of 17 food items served they had eaten. For each item we calculated the relative risk (RR) and the 95% confidence intervals (95% CI). Available leftovers were cultured for Staphylococcus aureus, Salmonella and Clostridium perfringens.
Results
Of 221 Indian workers, 76% complained of diarrhea and 75% of abdominal cramps. Only 2% had fever and all recovered within 24 hours. The median incubation period for the onset of illness was 13 hours (mean + SD=17.7 + 11.1 hours), but the distribution was bimodal, with the second peak 36 hours after the meal. Yogurt showed the highest relative risk (RR=9.6; 95% CI 4.5, 20.4). C. perfringens was isolated from leftover yogurt and from the table on which yogurt was made. Yogurt was prepared in 82-liter pots by mixing 50-kilogram. sacks of dried milk in water with three tins of yogurt concentrate and leftover yogurt. The mixture is prepared at 9 p.m. and is left at 40 degrees Centigrade until 5 p.m. the next day, when supper is served.
Conclusion
Although yogurt is an unusual source of C. perfringens food poisoning, the typical illness incubation period and the isolation of C. perfringens from leftover yogurt support this conclusion. Food poisoning due to anaerobic bacteria is likely to recur since large cooking pots are used and the storage area is inadequately refrigerated. The second wave of the epidemic curve could have occurred because some contaminated yogurt was used as a starter.