Alcohol consumption regression models for distinguishing between beverage type effects and beverage preference effects.

In relating health outcomes to alcohol consumption, several investigators have evaluated differences among beverage types, but there is no consistency with respect to models used for this purpose. Furthermore, beverage type effects and beverage preference effects have not been evaluated simultaneously. In this report, the authors propose regression models which permit the simultaneous evaluation of beverage type (congener dose response) effects and beverage preference (sociobehavioral) effects. The presence of sociobehavioral effects can be established even if the variables responsible for them have not been measured or identified. The models are applied to a data set from 589 women who participated in an oral contraceptive study at the Johns Hopkins University (Baltimore, Maryland) in 1988-1989.