This paper presents a new approach to estimating the relationship between consumption of specific foods and foodborne illness in the United States. We apply this approach to the case of foodborne sporadic campylobacteriosis illness. Most foodborne illness is sporadic and not part of a widespread outbreak. Foodborne Campylobacter infections are widely thought to be linked to chicken and are highly seasonal, primarily driven by temperature. We find that chicken purchased for consumption at home is not associated with sporadic Campylobacter infection in the United States, while ground beef and berries purchased for consumption at home are. The association between seasonality and the rate of Campylobacter infections is stronger than the association with temperature.