BACKGROUND Few studies of trauma care access and quality account for prehospital injury mortality. Little is known about geographic variation in prehospital mortality or the impact of prehospital care on injury disparities. METHODS Using the Centers for Disease Control Wide-ranging Online Data for Epidemiologic Research database, we queried county-level incidence of prehospital injury mortality from 1999 to 2016. We linked mortality incidence with county-level urban-rural classifications from the National Center for Health Statistics and population data from the US Census Bureau. We used negative binomial regression to estimate the relationship between rurality and prehospital injury mortality, adjusting for county-level distribution of race, sex, age, income, and insurance coverage. Models were then stratified by injury mechanism (motor vehicle traffic [MVT] vs. penetrating) to determine if prehospital mortality rates varied by type of injury. RESULTS Prehospital injury mortality rates were elevated for all urban-rural county classes, relative to large central metro counties, with incidence rate ratios (IRR) ranging from 1.25 (95% confidence interval [CI], 1.16–1.35) for fringe metro counties to 1.69 (95% CI, 1.58–1.82) for noncore counties. For MVT injury, IRRs for urban-rural classes compared with large central metro counties ranged from 2.02 (95% CI, 1.85–2.21) for fringe metro counties to 3.02 (95% CI, 2.76–3.30) to noncore counties. Incidence of prehospital mortality from penetrating injury was 14% higher for noncore counties compared to large central metro counties (IRR, 1.14; 95% CI, 1.05–1.23). CONCLUSION There is substantial geographic variation in prehospital injury mortality in the United States, with risk of prehospital death increasing with rurality. Patterns of prehospital death associated with penetrating and MVT injuries suggest that improvements to both trauma center access, prehospital care, and primary injury prevention are essential to reduce preventable injury deaths. LEVEL OF EVIDENCE Retrospective ecological analysis, level III.