The influence of personality on nicotine craving: A hierarchical multivariate statistical prediction model

The present study proposes a hierarchical multivariate statistical predicti on model which enables to determine the most prominent variables (physiolog ical, biochemical and personality factors) related to nicotine craving and dopaminergic activation. Based on animal studies reporting a reduction of t he rewarding effects of psychotropic drugs after blockade or destruction of the mesolimbic dopamine (DA) system, changes in nicotine craving after pha rmacological manipulation by means of a DA agonist (lisuride 0.2 mg) and a DA antagonist (fluphenazine 2 mg) were assessed in 36 healthy male heavy sm okers. The major aim was the development of a multivariate prediction model which is applicable in samples lacking variance homogeneity or the prerequ isite of a multivariate normal distribution, The model proposed is a combin ation of multivariate parametric and nonparametric methods taking advantage of their individual merits. Especially personality variables, such as sens ation seeking, impulsivity, and neuroticism showed to be important predicto rs of craving in this responder approach. Copyright (C) 2001 S. Karger AG, Basel.