A stream tube model was applied to simulate pathogen transport and fate in the subsurface at the field scale. Local-scale transport within each stream tube was described deterministically using analytic solutions for pathogen transport and fate in a uniform or dual-permeability porous medium. Important pathogen transport and fate processes that were accounted for in an individual stream tube included: advection, dispersion, reversible and irreversible retention, and decay in the liquid and solid phases. The velocity in a stream tube was related to a median grain size using the Kozeny–Carman equation, and filtration theory was used to predict the dependence of retention on physicochemical factors. The field-scale velocity distribution was described using a unimodal or bimodal lognormal probability density function (PDF). The bimodal lognormal PDF was used in conjunction with the dual-permeability model to account for exchange between slow and fast velocity domains. The mean and variance of the field-scale concentrations were calculated from local-scale stream tube information. The setback distance to achieve a selected risk of infection was determined from the modeled concentrations and a simplified risk assessment approach. Simulation results demonstrate that field-scale pathogen transport and setback distance were very sensitive to velocity distribution characteristics. Early breakthrough, higher peak concentrations, and larger setback distances were associated with faster stream tubes that had little retention, whereas the opposite trends were associated with slower stream tubes. The relative importance of faster stream tubes increased under physicochemical conditions that enhanced retention.