Risk assessment of hand washing efficacy using literature and experimental data

This study simulated factors that influence the levels of bacteria on foodservice workers' hands. Relevant data were collected from the scientific literature and from laboratory experiments. Literature information collected included: initial bacterial counts on hands and water faucet spigots, bacterial population changes during hand washing as effected by soap type, sanitizing agent, drying method, and the presence of rings. Experimental data were also collected using Enterobacter aerogenes as a surrogate for transient bacteria. Both literature and experimental data were translated into appropriate discrete or probability distribution functions. The appropriate statistical distribution for each phase of the hand washing process was determined. These distributions were: initial count on hands, beta (2.82, 2.32, 7.5); washing reduction using regular soap, beta (3.01, 1.91, - 3.00, 0.60); washing reduction using antimicrobial soap, beta (4.19, 2.99, - 4.50, 1.50); washing reduction using chlorhexidine gluconate (CHG), triangular ( - 4.75, - 1.00, 0); reductions from hot air drying, beta (3.52, 1.92, 0.20, 1.00); reduction from paper towel drying, triangular ( - 2.25, - 0.75, 0); reduction due to alcohol sanitizer, gamma ( 1.23, 4.42) - 5.8; reduction due to alcohol-free sanitizer, gamma (2.22, 5.38) - 5.00; and the effect of rings, beta (8.55, 23.35, 0.10, 0.45). Experimental data were fit to normal distributions (expressed as log percentage transfer rate): hand-to-spigot transfer, normal ( - 0.80, 1.09); spigot to hand, normal (0.36, 0.90). Soap with an antimicrobial agent (in particular, CHG) was observed to be more effective than regular soap. Hot air drying had the capacity to increase the amount of bacterial contamination on hands, while paper towel drying caused a slight decrease in contamination. There was little difference in the efficacy of alcohol and alcohol-free sanitizers. Ring wearing caused a slight decrease in the efficacy of hand washing. The experimental data validated the simulated combined effect of certain hand washing procedures based on distributions derived from reported studies. The conventional hand washing system caused a small increase in contamination on hands vs. the touch-free system. Sensitivity analysis revealed that the primary factors influencing final bacteria counts on the hand were sanitizer, soap, and drying method. This research represents an initial framework from which sound policy can be promulgated to control bacterial transmission via hand contacts. (C) 2002 Elsevier Science B.V. All rights reserved.