Antidepressant drug choice for first users in two regions in the Netherlands

Objective: To map the prescribing habits of physicians in two different regions in the Netherlands with respect to antidepressant drug choice for first users. Design: Retrospective follow-up study. Methods: All persons who received an antidepressant drug for the first time during October 1994 to September 1995 were identified from 29 community pharmacies situated in two regions in the Netherlands. Age, gender, type of prescriber, region and the concomitant use of certain groups of drugs as marker for certain diseases were evaluated as determinants for prescribing either a "classic" or a "second generation" antidepressant drug to first users using logistic regression analysis. Results: We identified 4,637 first users during the study period corresponding with an overall incidence density of 16 per 1,000 person-years. For the majority of first users of both regions, the same five antidepressants were prescribed. However, the two regions differed markedly with regard to the frequency of prescription of the individual antidepressants. Region of living was identified as the most important determinant of prescribing a certain category of antidepressant drugs to first users (OR 2.9 [95%CI 2.5-3.3]). Just a few patient characteristics were associated with antidepressant drug choice. The elderly were more likely to receive a classic antidepressant, as were patients concomitantly using anti-epileptics (OR 0.4 [95%CI 0.3-0.6]). Patients concomitantly using cardiac glycosides were more likely to receive a second generation antidepressant (OR 1.8 [95%CI 1.2-2.1]). With respect to age and the other patient characteristics studied, the same pattern was observed in both regions. Conclusion: Regional differences contributed more to differences in prescribing habits than individual patient characteristics. In order to improve pharmacotherapy with antidepressant drugs relevant patient characteristics should more be taken into account.