This paper presents a critical review of state-of-the-art macro models for road accidents. Such a review is meant to identify and establish the significance of policy and socioeconomic variables affecting the level of road accidents. The aim is to identify those variables associated with effective policies and interventions to enable decision makers to improve the level of road safety. The variables that appear to affect the number of fatalities or injuries are: vehicle miles travelled (VMT), vehicle population, income (in its various forms), percentage of young drivers, intervention policies such as speed limits, periodic vehicle inspection, and minimum alcohol-drinking age. Viewed critically, the state-of-the-art models being used to explain and predict road accidents are still deficient. One possible approach to correcting this deficiency draws from consumer utility theory, using analytical models built on a newly constructed theoretical framework. Success in estimating such models may improve predictions of road accidents, thus demonstrating the comparative cost effectiveness of alternative intervention policies.