This site is intended for health professionals only

At the heart of general practice since 1960

NICE to use unpublished data in depression guidance

By Lilian Anekwe

NICE will consider unpublished data on the efficacy of antidepressants before new guidance on the management of depression is published later this year, Pulse has learned.

The institute has been criticised for failing to look at such data after a meta-analysis of published and unpublished data suggested antidepressants were no more effective than placebo.

The meta-analysis of 47 clinical trials, published in the open-access journal PLoS Medicine, reviewed data on patients with moderate and severe depression treated with the four most commonly prescribed antidepressants in the US– fluoxetine, venlaflaxine, nefadozone and paroxetine. Nefadozone is not licensed in the UK.

Depression severity scores in patients taking the antidepressants were no better than placebo (see box).

Professor Irving Kirsch, professor of psychology at the University of Hull, claimed NICE had been unable to access unpublished data on the efficacy of antidepressants.

‘NICE did want to get all of the data but they were not able to determine which was published and which was unpublished. They should have had access to the information from the pharmaceutical companies.'

Mr Stephen Pilling, co-director of the National Collaborating Centre for Mental Health, which oversees the development of mental health guidelines for NICE, told Pulse previously unpublished data from the US Food and Drug Administration would be considered when NICE published updated its depression guidance in December.

‘We will be formally looking at the meta-analysis when we come to update the guidelines, and looking very carefully at both published and unpublished data. When we come to look at some of the new drugs we will be looking at unpublished data on efficacy as well.'

Mr Pilling said greater clarity was needed on what data NICE included in their reviews of clinical trial evidence.

‘It's not clear to me for example if US pharmaceutical companies are included, and maybe they should be. The fuller datasets we get the more confident we can be about the guidance we produce.'

Dr Tim Kendall, consultant psychiatrist at Sheffield Care Trust and the college's deputy director, said: ‘We have called on the government to change this and it is extraordinary that we did not know about this legal barrier before.'

But NICE may have to stop short of mandating pharmaceutical companies to provide unpublished or negative data, as under EU law such data could be commercially sensitive.

A spokesperson from the DH said: ‘We planned to support the principle of mandatory registration of clinical trials but legal advice stated this would be illegal under EU law. For that reason we have adopted a voluntary approach in the UK.'

What the new meta-analysis showed

What the new meta-analysis showed

What the new meta-analysis showed

Data on all clinical trials submitted to the FDA were obtained under the US Freedom of Information Act
The effect of initial severity on improvement was assessed for both active and placebo arms
There were no differences between drug and placebo outcomes in moderate depression
There was a small and statistically significant difference for patients with the most severe depression

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say