A South Australian Salmonella Mbandaka outbreak investigation using a database to select controls

Between April and June 1996, 15 persons with Salmonella enterica serovar Mbandaka infection were reported in South Australia (population 1.6 million) compared with 12 over the previous five years. To identify a possible source for the infections a case control study was conducted.

Trained interviewers asked 15 cases and 45 controls about their consumption of 105 foods. Controls were matched to case residential location and age. They were selected from a previously constructed database of 3,014 randomly selected South Australian households.

Thirteen of the 15 cases ate 'generic' or 'retail store' brands of peanut butter produced by the same factory in another state, compared with five of the 45 controls (p < 0.01). Salmonella Mbandaka was isolated from three opened jars of peanut butter from case households, and from three unopened jars from retail outlets. Further investigation implicated roasted peanuts from a third Australian state as the source of the Salmonella contamination.

This is the first recorded outbreak of salmonellosis resulting from the consumption of peanut butter. The SA outbreak investigation comprised a matched case control study to identify possible common food sources. Such investigations need be conducted rapidly to maximise public health benefits, and the utility of selecting controls from a population based database can improve timeliness.