Eur J Clin Pharmacol (2004) 60:221–224

It has come to our attention that an error was made in one calculation in respect of the above paper. As a result the likelihood ratio published is incorrect. The correct likelihood ratios for admissions involving venlafaxine shown in Table 1 (and also in the abstract) is 0.97 (0.81-1.16), not 1.33 (1.13-1.55).

This error does not alter the main findings of the manuscript or any of the overall conclusions that we reached.