This study determined the crash modification factors (CMFs) associated with horizontal curve realignment using the beforeafter empirical Bayes method and compared the results from published CMFs from cross-sectional studies. This evaluation used data from rural, two-lane roads in California, North Carolina, and Ohio. The evaluation revealed a 68% reduction in total crashes, a 74% reduction in injury and fatal crashes, a 78% reduction in run-off-road and fixed object crashes, a 42% reduction in crashes during dark conditions, and an 80% reduction in wet-road crashes, all of which were statistically significant at the 95% confidence level. The results pertain to a range of site characteristics, the most important of which is the range of before and after degree of curve. The average degrees of curve in the before and after periods were 18.1 (with a minimum of 3.2 and a maximum of 52.1) and 6.9 (with a minimum of 0.0 and a maximum of 16.3), respectively. The average central angle of the curves was approximately 42 (with a minimum of 1 and a maximum of 117). The CMFs from this beforeafter evaluation are lower compared to CMFs estimated from two previous cross-sectional studies. There is a need for further research with a larger sample of sites to assess the reliability of the CMFs obtained from this beforeafter evaluation.