College members (Craig Whittington, Tim Kendall, Peter Fonagy, David Cottrell, Andy Cotgrove and Ellen Boddington) were authors of the Lancet Paper of the Year, awarded in December 2004 and referred to by the Editor as ‘one of the most stunning and influential systematic reviews in recent years’.

The paper was selected by the Lancet’s International Advisory Board and Editors from among any paper worldwide published in a scientific or medical journal between October 2003 and October 2004.

The paper compared the results of two meta-analyses: one using only published randomised controlled trials; the second combining the published and unpublished trials, all obtained by the Medicines and Healthcare Products Regulatory Agency (MHRA) from drug companies. The published trials suggested that for the four selective serotonin reuptake inhibitors (SSRIs) examined (fluoxetine, sertraline, paroxetine and venlafaxine) there was a weak but positive risk-benefit ratio. When these data were combined with the unpublished data the risk-benefit ratios were reversed for all except fluoxetine. This supported the decision of the MHRA to contraindicate the use of SSRIs (except fluoxetine) for children and adolescents with depression.

The importance of the paper lies in the fact that it presented concrete meta-analytical evidence of the potentially dangerous effects of publication bias and/or the failure to publish negative results by drug companies (selective publishing).