Objective Ð To assess the extent, characteristics and determinants of benzodiazepine prescription in outpatient Primary Health Care.
Methods Ð A clinical audit of a strati®ed random sample of Primary Health Care Centres in the seven islands and 1.6 million inhabitants region of Canarias' in Spain was carried out. From those centres, a random sample of 1045 clinical records was reviewed and information on diagnosis, prescription and prescribed dosages was collected in a structured questionnaire. A multivariate logistic regression analysis was performed in order to determine the risk factors' for the use of benzodiazepines.
Results Ð Benzodiazepine prescription was recorded in 23.4% of all clinical records; 87.7% of these were for benzodiazepines classi®ed as anxiolytics (N05B) and 12.3% for hypnotics (N05C2). Benzodiazepine prescription was more common for women, elderly, widowed, divorced, low educational background, housewives and retired people. Using multivariate logistic regression, the probability of benzodiazepine prescription was found to be closely related to age, gender and employment status, but not with educational level. Prescribed Daily Doses were lower than De®ned Daily Doses (DDD) in 77.1% of all anxiolytic prescriptions, but were in agreement with DDD in 90% of hypnotic prescriptions. The duration of treatment recorded in the clinical records was 25+21 months, with a range of 1 and 144 months. General Practitioners were responsible for 67% of all benzodiazepine prescription. Anxiolytics were prescribed as a single daily dose in 57% of the cases, and only `at supper' in 48.6%.
Conclusion Ð In the general population attending Primary Health Care Centres of the Canary Islands Health System the prescription of benzodiazepines is higher for women and the elderly, and the most common use is chronic, with a duration of over 2 years in most cases. Anxiolytics are prescribed in doses which are much lower than those used as DDD and were used only `at night' in almost half of the cases. This could represent an overlapping of the indications with hypnotics, and explain part of the huge dierence in the use of anxiolytics in Spain compared with other ®gures in Europe. This fact must also be taken into account when making inferences of benzodiazepine use from sales statistics, which are very imprecise measures of drug use.