This research was undertaken to determine the effectiveness of an alcohol-based hand sanitizer on hands contaminated with a nonpathogen surrogate for Escherichia coli O157:H7, where the source of the contamination was frozen hamburger patties. A nonpathogenic nalidixic acid-resistant food-grade strain of Enterobacter aerogenes was used to inoculate frozen hamburger patties composed of 76% lean beef and 24% fat. Thirty-two individuals participated to produce the data used in this study. Each participant handled nine patties at least three times, a sample for microbiological analysis was collected from the surface of one hand, the participant sanitized both hands, and a sample was collected from the other hand. Burger handling created perceptible and visible food debris on the hands of most participants. Computer simulations also were used to perform a variety of risk calculations. The average reduction in bacteria from the use of sanitizer on hands contaminated by frozen burgers containing E. aerogenes was 2.6 +/- 0.7 log CFU per hand. An experiment designed to simultaneously test the effect of sanitizer on E. aerogenes and E. coli O157:H7 also revealed no significant difference in sanitizer effectiveness against the two organisms. The results of the real-world risk estimation calculations (using the actual prevalence and concentration of E. coli O157:H7 in ground beef) predict that once in 1 million trials, a single pathogen cell will be transferred to a single lettuce piece. The effectiveness of this sanitizer intervention was similar to that for hand washing and glove use previously reported. The person-to-person microbial reduction variability from sanitizer use is similar to published data for glove use and was less variable than published data on hand washing effectiveness.