This study modelled the economic impact of mirtazapine, compared to amitriptyline and fluoxetine, in the management of moderate and severe depression in the UK, as well as the costs related to discontinuation of antidepressant treatment. Decision models of the management of moderate and severe depression were developed from clinical trial data, resource use obtained from interviews with general practitioners and psychiatrists, and published literature, and were used to estimate the expected direct National Health Service (NHS) costs of managing a patient with moderate or severe depression. The expected cost of healthcare resource use attributable to managing a patient suffering from moderate or severe depression who discontinues antidepressant treatment, irrespective of the initial treatment, was estimated to be pound 206 (range pound 50 to pound 504) over five months. Using miriazapine instead of amitriptyline for seven months increases the proportion of successfully treated patients by 21% (from 19.2 to 23.2%) and reduces the expected direct NHS cost by pound 35 per patient (from pound 448 to pound 413). Using mirtazapine instead of fluoxetine for six months increases the proportion of successfully treated patients by 22% (from 15.6 to 19.1%), albeit for an additional cost to the NHS of pound 27 per patient (from pound 394 to pound 420). In conclusion, this study suggests that mirtazapine is a cost-effective antidepressant compared to amitriptyline and fluoxetine in the management of maderate and severe depression in the UK. (C) 2000 Editions scientifiques et medicales Elsevier SAS.