Fewer but heavier caffeine consumers in schizophrenia: a case-control study.

According to the literature, there is an association between schizophrenia and caffeine consumption, but it is not clear whether schizophrenia is associated with either higher prevalence of daily caffeine intake or the amount consumed. In this study we compared our previously published schizophrenia patients (n=250) with a control sample (n=290) after controlling for demographic variables and tobacco and alcohol consumption. Current caffeine intake was less frequent in schizophrenia patients (59%, 147/250) than in controls (70%, 204/290). In the multivariate analyses, caffeine intake was less frequent at an older age and in schizophrenia patients, and more frequent in smokers and alcohol users. Among caffeine consumers, heavy caffeine intake (> or =200 mg/day) was significantly associated with schizophrenia (64%, 94/147 in schizophrenia versus 36%, 73/204 in controls), as well as older age and smoking. Daily amount of caffeine intake and smoked cigarettes correlated significantly in the schizophrenia group but not in the control group; the correlation of caffeine intake with nicotine dependence was low and non-significant in both groups. The association between current smoking and heavy caffeine intake may be partly explained by a pharmacokinetic effect: tobacco smoke compounds induce caffeine metabolism by the cytochrome P450 1A2. Although schizophrenia by itself may be associated with heavy caffeine intake in caffeine users, part of this association was explained by the association between schizophrenia and smoking. The relationship between caffeine and alcohol intake appeared to be more complex; alcohol and caffeine use were significantly associated, but within caffeine users alcohol was associated with less frequent heavy caffeine consumption among smokers. In future studies, the measurement of plasma caffeine levels will help both to better define heavy caffeine intake and to control for smoking pharmacokinetic effects.