Wildlife-vehicle collisions (WVCs) negatively impact wildlife populations and create dangerous driving situations for motorists. In Utah, USA, mule deer (Odocoileus hemionus) encounter a variety of hazards as they attempt to cross highways and interstates, some of which are 8 lanes wide. Agencies have sought to mitigate the risks posed to drivers and mule deer by building crossing structures for wildlife. The objectives of this study were to evaluate the effectiveness of crossing structures in Utah to safely pass mule deer under highways and to determine the variables that best explain mule deer passage use. From 2008 - 2011 we used 26 camera traps to measure levels of mule deer use of 9 culverts and 4 bridges in Utah. We tested for relationships between mule deer structure use and a variety of structural and landscape attributes at each site, including 2 time variables: time since the structure was built and time each structure was monitored by our camera traps. We also developed and tested a new equation (window ratio) that measured culvert openness to approaching mule deer. In the single variable regression models, mule deer structure use was positively correlated with short culverts and coarse scale shrub cover, and negatively correlated with fine scale grass cover. In the multivariate model, structure use was positively correlated with days monitored and elevation and short culverts. Although the new window ratio did not emerge as the most important predictor for mule deer crossing use, it was more effective at predicting mule deer culvert use than the often referenced openness factor. Our results indicated that 12 of the 13 crossing structures studied effectively facilitate the movement of mule deer in Utah; however some were used far more than others. We suggest that older crossing structures built with the shortest dimensions possible, with attached wildlife-exclusion fencing, and in shrubby habitat will be most effective at passing a high volume of mule deer under Utah highways.