Fatigue-related crashes account for 2.2% to 2.6% of all fatal crashes in the United States on an annual basis. These types of crashes are prevalent in rural areas, and public roadside rest areas were developed largely to alleviate motorist fatigue and reduce the opportunity for such crashes by providing safe parking areas for tired drivers. However, research about the safety effects of rest areas has been limited. Results are presented of a spatial analysis that investigates how a road segment's proximity to a rest area affects the frequency of fatigue-related crashes. Poisson and negative binomial models are estimated for freeways and two-lane highways to isolate the effects of proximity while controlling for other relevant factors, such as traffic volume. The results of these models indicate that the proximity of a road segment to the nearest rest area significantly influences crash frequencies on both freeways and two-lane highways, where traffic volumes tended to have similar effects, although the effects of proximity were slightly more pronounced on two-lane highways. The study results suggest that roadside rest areas provide a safety benefit, and the crash prediction models developed as a part of this research provide a starting point for quantifying these impacts.