In this research we studied the association between commercial motor vehicle drivers' medical conditions and crash severity. Some aspects of medical condition were considered. To our knowledge, no study has ever isolated this association. The severity of a crash was measured by the total number of victims (injured and dead). We estimated nonlinear regression models (specifically, Poisson and negative binomial) which incorporated, simultaneously, information on drivers' characteristics, crash circumstances and health status, in order to isolate the association between health status and crash severity. Our results show that crashes of truck drivers with binocular vision problems and bus drivers with hypertension are more severe than those of healthy drivers. No other medical condition considered in this study was significantly associated with crash severity. Many variables describing crash circumstances were also significant.