The choice of treatment after failure of selective serotonin re-uptake inhibitors in the treatment of depression largely depends on the preferences of the physician. The STAR*D (Sequenced Treatment Alternatives to Relieve Depression) trials set out to rationalise the treatment of depression after the failure of citalopram. When subjects who were unresponsive or intolerant to citalopram were switched to sustained-release bupropion, sertaline or extended-release venlafaxine, remission rates were similar (21, 18 and 25%, respectively). When subjects who were unresponsive to citalopram had sustained-release bupropion or buspirone added to citalopram, there were similar remission rates of 29.7 and 30.1%, respectively. The interpretation of this data are complicated by the lack of placebo groups. One interpretation is that sustained-release bupropion, sertaline or extended-release venlafaxine are equally effective in subjects with major depression after the failure of citalopram, and that sustained-release bupropion or buspirone are equally effective when added to citalopram in subjects with major depression after the failure of citalopram. An alternative explanation is that none of the drugs are effective in citalopram failure, and the responses are the placebo effect.