The present study was conducted to evaluate the effect of potential risk factors--such as driving without a license, alcohol use, speed, seat belt, and helmet--use on fatality in motor vehicle traffic accidents. Unconditional multiple logistic regression analysis was employed to take these factors and age into account, simultaneously. The effect of driving without a license was not significant after controlling for other factors. The deleterious effect of alcohol use remained significant for male motorcar drivers after controlling for speed and seat belt use. Magnitude of the risk due to speed was slightly reduced after controlling for alcohol use and seat belt use, but the striking effect remained highly significant. Speed was the strongest risk factor of fatality for both motorcycles and motorcars and for both sexes and seemed to be more critical for motorcyclists than motorcar drivers. The protective effect of seat belt use was unchanged after adjustment for alcohol and speed, and the effectiveness of seat belt use was demonstrated for motorcar drivers. The effectiveness of helmet use for male motorcyclists was dependent upon speed at the time of the accidents, suggesting an interaction between helmet use and speed. Helmet use was definitely protective at a low speed of < or = 50 km/h, but ineffective at high speeds of over 50 km/h.