Utilisation of codeine and propoxyphene: geographic and demographic variations in prescribing, prescriber and recipient categories

Abstract

Objectives: To assess (1) whether the utilisation of codeine or propoxyphene differs among the three major Swedish cities (Stockholm, Göteborg and Malmö) and between urban and semirural areas; (2) if so, whether it co-varies with the utilisation of other potentially dependence-promoting drugs, benzodiazepines; (3) what influence age, gender and socioeconomic factors have on the prescribing of the two narcotic analgesics; and (4) whether different codeine-prescriber categories have different prescribing habits.

Methods: In Sweden, all pharmacies are owned by one corporation, Apoteket AB. This corporation collects, stores and compiles statistics on all drug sales in Sweden, and data are available both on national, regional, county and municipal levels. The employed unit is defined daily dose (DDD) per 1000 inhabitants per day. Using the pharmacy computer system while dispensing a drug, prescription patterns can be elucidated. This system describes the number of drug items dispensed, drug amounts and age and gender of patients. Furthermore, data from another, ecological study were used to relate codeine and propoxyphene utilisation to that of benzodiazepines and to various socioeconomic data available from records of the city of Malmö.

Results: The utilisation of analgesics in Sweden has increased during a 10-year period. The withdrawal of over-the-counter combinations containing aspirin and low-dose codeine in 1990 resulted only in a transient decrease of codeine use. The utilisation of codeine in Malmö and Göteborg was considerably higher than that in Stockholm and in the rest of Sweden, including the surroundings of Malmö. In Malmö and Göteborg, codeine was most often prescribed by private physicians to middle-aged persons, particularly women. Districts in Malmö with a high utilisation of codeine were associated with unfavourable socioeconomic conditions and a high utilisation of benzodiazepines. The utilisation pattern of propoxyphene showed less or no such deviations.

Conclusion: The results suggest an inappropriate use of codeine in two major cities in Sweden.