Return to Toolbox Main.
This excel file will help assess the potential utlity of a case-control study for a foodborne disease outbreak investigation. It calculates cumulative binomial probabilities.
During an investigation of a cluster of salmonellosis cases, four of your eight cases recall eating alfalfa sprouts during the week before onset. Is that a suspiciously high proportion? If you can estimate the "background" rate of sprout consumption in your population (e.g., from survey data or even a wild guess), you can calculate the odds of getting as many or more sprouts eaters in a sample of N independent cases by chance alone. Hint: the background rate of sprout consumption is <10%, and finding even 4/8 by chance alone is highly unlikely.