Exploring spatial averaging of contamination in fomite microbial transfer models and implications for dose

Background: When modeling exposures from contact with fomites, there are many choices in defining the sizes of compartments representing environmental surfaces and hands, and the portions of compartments involved in contacts. These choices impact dose estimates, yet there is limited guidance for selection of these model parameters. Objective: The study objective was to explore methods for representing environmental surface and hand contact areas in exposure models and implications for estimated doses. Methods: A simple scenario was used: an individual using their hands to contact their face and two microbially contaminated environmental surfaces. Four models were developed to explore different compartmentalization strategies: (1) hands and environmental surfaces each represented by one compartment, (2) hands represented by two compartments (fingertips vs. non-fingertip areas) while environmental surfaces were represented by one compartment, (3) hands represented by a single compartment and environmental surfaces represented by two compartments, and (4) hands and environmental surfaces each represented by two compartments. Sensitivity analyses were conducted to evaluate the influence of heterogeneous surface contact frequency, hand contact type, and hand dominance on dose. Results: Estimated doses were greatest when hand areas and environmental surfaces were each represented by two compartments, indicating that surface area “dilutes” contaminant concentration and decreases estimated dose. Significance: Model compartment designations for hands and environmental surfaces affect dose estimation, but more human behavior data are needed. Impact statement: A common problem for exposure models describing exposures via hand-to-surface contacts occurs in the way that estimated contamination across human skin (usually hands) or across environmental surfaces is spatially averaged, as opposed to accounting for concentration changes across specific parts of the hand or individual surfaces. This can lead to the dilution of estimated contaminants and biases in estimated doses in risk assessments. The magnitude of these biases and implications for the accuracy in risk assessments are unknown. We quantify differences in dose for various strategies of compartmentalizing environmental surfaces and hands to inform guidance on future exposure model development.