Background : Since antidepressants are prescribed for multiple indications, the use of an antidepressant cannot be equated with a diagnosis of depression. Objective : The objective of this study was to examine the quality of antidepressant prescribing in Belgian nursing homes, with a critical evaluation of indications and dosages, to see whether depression was appropriately treated in terms of drug choice, the indications for which antidepressants were being prescribed and whether there was underdosing. Methods : This analysis was based on data obtained in the Prescribing in Homes for the Elderly in Belgium (PHEBE) study, a cross-sectional, descriptive study of a representative, stratified, random sample of 1,730 residents from 76 Belgian nursing homes. The PHEBE study investigated overall drug utilization in Belgian nursing homes in 2006. Clinical and medication data for the present study were obtained from this study. A 28-item checklist of clinical conditions was designed ad hoc for the PHEBE study and sent to the residents' general practitioners (GPs) to collect clinical information. We copied the residents' medication charts, classified the drugs using the Anatomical Therapeutic Chemical (ATC) classification system codes and transferred the drug names and dosages into a database. Information on indications was retrospectively obtained from the GPs, so that we could link the indication to each medication. Minimum effective doses (MEDs) of antidepressants to treat major depression were obtained from the literature to assess underdosing. Results : The overall use of antidepressants in nursing homes was 39.5 % (95 % CI 37.2, 41.8). The physicians classified 34.2 % (95 % CI 32.0, 36.4) of the residents as having depression, and 80.9 To of these patients were treated with an antidepressant. Indications among the single antidepressant users (n = 551) were depression (66.2 %), insomnia (13.4 %), anxiety (6.2 %) and neuropathic pain (1.6 %). In the indication of depression, 74.8 % used a selective serotonin reuptake inhibitor (SSRI), predominantly citalopram, sertraline and escitalopram. Venlafaxine was used by 10.7 % of the residents. Dosages for these antidepressants were equal to or higher than the MED. But when trazodone, amitriptyline or mirtazapine were used to treat depression, respectively, 92.3, 55.5 and 44.5 % of prescribed dosages were below the MED. In the indication of insomnia, most of the time, trazodone (90.5 %) or mirtazapine (5.4 %) were used, and in lower dosages than those required for depression treatment (