To estimate the prevalence, characteristics and risk factors associated with the development of benzodiazepine dependence in the users of these active ingredients. A representative sample of patients currently receiving benzodiazepine treatment for 1 month or longer (mean 38.2±52 months, range 1–360 months) was studied. One thousand and forty eight (1048) consecutive patients attending 20 primary-care health centres of the Canary Islands Health Service participated in this study during 2002. The severity of dependence scale (SDS) was used as a screening test of benzodiazepine dependence among patients using benzodiazepines. Of patients using benzodiazepines for more than 1 month, 47% developed dependence to these compounds. Benzodiazepine dependence was more prevalent among women who were middle aged, separated, of low educational background, unemployed or housewives. Patients using short half-life benzodiazepines registered higher rates of dependence as well as those using higher doses or the patients with longer use. However, multivariate logistic regression analysis shows that benzodiazepine dependence was closely related only to three of the variables considered: the benzodiazepine dose used, the duration of this use and to the concomitant use of antidepressants. The distribution of dose and length of treatment shows no evidence of pharmacological tolerance, neither in the whole sample nor in those patients with addiction (SDS+), reinforcing the idea that psychological dependence (addiction) is more relevant than pharmacological or physiological dependence in benzodiazepine chronic use.
