In the study of the beneficial and adverse effects of psychotropic drugs, in addition to experimental studies (Barbui et al., 2007), epidemiological studies may provide interesting insights. A key methodological aspect in drug utilization studies is how drug consumption should be measured. Specifically, the need is to end up with a single variable that may be reliably used to indicate the overall consumption, for each patient included in a certain survey, of one specific drug (for example one antipsychotic, say haloperidol) or a specific drug class (for example second-generation antipsychotics) or a group of drugs (for example psychotropic drugs, including antipsychotics plus antidepressants plus benzodiazepines). Not only such a variable has to accommodate the problem that actual doses cannot be directly compared, but also that many patients often receive more than one drug, belonging to different or to the same pharmacological class. A methodology is therefore needed to convert each medication dosage into a standardised measure that allows to calculate, for each patient, a cumulative index of drug consumption. Standardised measures of drug consumption are also beneficial for national and international comparisons, for the evaluation of time trends in drug use, and for assessing the public health impact of specific events (for example a change in reimbursement status).
