Data From Unpublished Trials Suggest That Most SSRI Antidepressant Drugs Unsuitable for Children

Lancet 2004; 363: 1335, 1341-45

A UK study in this week's issue of THE LANCET suggests that reliance on published studies alone to guide the treatment of childhood depression could be inappropriate. A systematic review including published and unpublished data shows that only one drug, fluoxetine (Prozac), was not associated with negative outcomes for children with depression.

There is a clear need for effective treatment for childhood depressive disorders; up to 6% of children and adolescents are thought to suffer from depression, and suicide is the third largest cause of death among 10-19 year-olds in developed countries. The class of drugs known as selective serotonin reuptake inhibitors (SSRIs) has previously been considered to offer the best drug-treatment option for childhood depression, although recommendations for various SSRIs including fluoxetine and paroxetine (Paxil) are based on limited evidence from published studies. Questions concerning the safety of SSRIs led Craig Whittington and Tim Kendall from the National Collaborating Centre for Mental Health, London, UK, and colleagues to compare and contrast published and unpublished data on the
risks and benefits of these drugs.

The investigators reanalysed data from randomised controlled trials that evaluated an SSRI compared with placebo in children aged 5-18 years that were published in a peer-reviewed journal. The analysis was extended to include data from unpublished trials (made available from the UK Committee for the Safety of Medicines).

From the published studies alone, all the SSRIs appeared to have a favourable risk-benefit ratio. With the exception of fluoxetine, the wider analysis of including data from unpublished trials suggested that the risk of SSRI treatment exceeded the benefits; for example, the inclusion of unpublished data on the effects of paroxetine showed it to be associated with a small increased risk in suicidal thoughts or attempted suicide.

The investigators comment that increased access to unpublished data from the pharmaceutical industry is crucial to enable a clear picture of antidepressant drug safety and efficacy to emerge. Tim Kendall comments: "The clinical guideline programme developed by NICE [UK National Institute for Clinical Excellence] is underpinned by an evidence-base published in peer-reviewed journals. Although NICE accept submissions of evidence from stakeholders, which might not be published, this acceptance is only done on the understanding that data are made publicly available. Drug sponsors who withhold trial data (or do not make full trial reports available) undermine the guideline programme, which can ultimately lead to recommendations for treatments that are ineffective, cause harm, or both. Others have suggested that the pharmaceutical industry needs greater regulation, and in particular that all trial data-whether published or unpublished-should be fully accessible. In any event, greater cooperation and openness between the pharmaceutical industry and guideline developers, including gaining access to unpublished full trial reports, is clearly a matter of some urgency; this access would allow critical appraisal of study methodology and inclusion of unpublished data that meet recognised standards of quality. The fact that the drugs reviewed here have previously been recommended for use in children on the basis of a very restricted published evidence base can only serve to increase that sense of urgency".

This week's Lancet editorial (p 1335) comments: 'The story of research into SSRI use in childhood depression is one of confusion, manipulation, and institutional failure. Although published evidence was inconsistent at best, use of SSRIs to treat childhood depression has been encouraged by pharmaceutical companies and clinicians worldwide. Last month, the Canadian Medical Association Journal revealed excerpts from an internal GlaxoSmithKline memorandum demonstrating how the company sought to manipulate the results of published research. Concerning a study of paroxetine use in children, the memorandum states "It would be unacceptable to include a statement that efficacy had not been demonstrated, as this would undermine the profile of paroxetine". Last year the UK Committee on Safety of Medicines prohibited the treatment of childhood depression with any SSRI except fluoxetine. Despite this, the Food and Drug Administration in the US appears last week to have failed to act appropriately on information provided to them that these drugs were both ineffective and harmful'.

'In a global medical culture where evidence-based practice is seen as the gold standard for care, these failings are a disaster. Meta-analysis of published data supports an increasing number of clinical decisions and guidelines, which in turn dictate the use of vast levels of health-care resources. This process is made entirely redundant if its results are so easily manipulated by those with potentially massive financial gains. The global sales of the GlaxoSmithKline SSRI paroxetine, for example, amounted to US $4*97 billion last year alone. Moreover, the utility of organisations such as the National Institute for Clinical Evidence (NICE) is significantly undermined in circumstances where they are only able to access data on health-care products that are seen as advantageous to the product's manufacturers'.

Contact: Dr Tim Kendalll, National Collaborating Centre for Mental Health; tim.kendalll@sct.nhs.uk or tim2.kendall@virgin.net.