Anti-depressants: which research should we believe?

The four main SSRI antidepressants are no more clinically effective than placebo for people with mild or moderate depression, a meta-analysis in the Public Library of Science journal reported in February.

The research was front page news on four broadsheet newspapers as well as BBC and ITV headlines.

The study by psychology professor Irving Kirsch and colleagues revealed, however, nothing researchers were not familiar with. Kirsch, when employed in America and using the same data, had arrived at the same conclusions in a study he published in Prevention and Treatment journal a full six years previously, in 2002.

One key difference this time, though, was Kirsch is now also working at the University of Hull who had a media-savvy press officer. It was she who publicised the findings to her UK media contacts, resulting in the blanket coverage. The study also came at a time when the government is bidding to rid the UK of the “Prozac nation” slur.

The spiralling number of prescriptions for anti-depressants hit a record high of more than 31 million in England in 2006. There were 16.2m prescriptions for SSRIs (selective serotonin reuptake inhibitors).

Instead the government is turning to psychological therapies – and is to recruit an extra 3,600 cognitive behavioural therapists to “cure” 450,000 people with depression and anxiety.

Nevertheless, Kirsch et al’s study is highly significant. It was based on all drug firm data – published and unpublished – that the US Food and Drug Administration (FDA) had used during the eighties and nineties to first approve four SSRIs, namely fluoxetine (Prozac), venlafaxine (Seroxat), nefazodone (Effexor), and paroxetine (Serzone).

Kirsch had used freedom of information law to retrieve the data. A meta-analysis which includes unpublished trials eliminates “reporting bias” (caused by the same data being published in several journals) and “selective reporting” (where only positive outcomes of a drug are published).

Kirsch’s view is clear-cut. “These [SSRI] drugs should never have been licensed – that’s the logical response from our study,” he said. “If one can not draw positive conclusions from the data there is just no basis to conclude that antidepressants are effective interventions.”

This view is radically different to that of the National Institute of Health and Clinical Excellence. Its 2004 guidelines on depression advises against doctors prescribing SSRIs to people with mild depression, but states SSRIs are clinically useful for moderate or severe depression.

Kirsch acknowledges antidepressants may have some effect for people with severe depression “Antidepressants may still have a role but we must first think of alternative forms of treatment,” he said. “Antidepressants can be given to people with severe depression, but only if they don’t manage to recover from alternative treatment.”

Nice’s guidelines are now being revised and are to be re-published in June next year. In a statement, Nice’s clinical and public health director, Professor Peter Littlejohns, has promised Kirsch’s Plos paper will be examined. But Kirsch’s 2002 paper was scrutinised by Nice in drawing up its 2004 guidelines.

And the debate by researchers over whether Nice’s judgement on SSRIs is empirically valid continues.

In The British Medical Journal, Kirsch and Dr Joanna Moncrieff, a senior lecturer in social and community psychiatry at University College of London, have criticised the methodology used by Nice for assessing antidepressant efficacy.

In response, the guidelines’ authors accused Kirsch and Moncrieff of “an inaccurate and partial reading” of the evidence.

“We believe Moncrieff and Kirsch have failed…to undermine the evidence base for antidepressants and to establish their case for a reconsideration of the NICE recommendations for their use,” they said.

But a new book by Moncrieff, also an NHS consultant psychiatrist, is likely to up the anti. The book, entitled the The Myth of the Medical Cure, Moncrieff accuses Nice of massaging the evidence.

She writes that, when drawing up its guidelines, Nice conducted a second superfluous statistical analysis on the data to “present it in such a way that allowed them [the authors] to draw the conclusions they were comfortable with.”

A Nice spokeswoman said there was “no truth” in these accusations.

And in its 2009 guidelines, Nice is unlikely to change its mind on SSRI efficacy, some experts say.

No more so, they add, because, as well as the FDA, the UK drug regulator, the Medicine and Healthcare Products Regulatory Agency (MHRA) has licensed SSRIs for depression.

“It would be highly difficult for Nice to go against what the drug regulators have decided, says David Healy, professor in psychological medicine at Cardiff University. “They could have the drug firms on its back trying to sue.”

Nevertheless, a major concern on SSRI efficacy is whether scientific objectivity – as well as public interest and safety – is being compromised by the commercial muscle of the pharmaceutical industry.

Just days after the Kirsch headlines, GlaxoSmithKline was rapped by the MHRA for failing to submit studies which indicated its Seroxat SSRI led to suicidal tendencies in young people.

The drug, despite not being licensed for under 18s, was being widely prescribed to them by doctors. The government says it’s to tighten up the law.

In a letter to GlaxoSmithKline’s chief executive Jean-Pierre Garnier, MHRA chief executive Kent Woods said the “moral responsibility” of pharmaceutical firms’ submission of relevant trails needed to be backed up by the “unambiguous force of law.”

“There is obviously a need to tighten the law to make it absolutely clear that pharmaceutical companies have a legal responsibility to inform the regulator of any information that changes the benefit: risk profile of their products, regardless of whether the information relates to a licensed indication,” his letter read.

Health minister Dawn Primarolo says new legislation will be in place this year to make companies disclose “any information [on their drugs] they have that would have a bearing on the protection of health.”

However, Healy fears independent researchers such as himself will still be unable to scrutinise those unpublished trials not showing drug efficacy which are submitted to the MHRA.

He says its vital for this trial data to be publicly accessible without barriers being put in the way.

“In some form the drug firm data needs to available to the public generally, not just the regulators, ensuring [independent] experts can look at the data,” he says.

“There are good ethical grounds for doing this. Volunteers who sign up for a drug firm’s clinical trial would not have got involved if they thought that drug company would have put away the results.”

A MHRA spokesman said independent researchers can apply to see drug firm trial data, unless it breaches commercial or patient information. Will such exceptions be dismantled under the new law? “We will have to see what is decided in drawing up the law,” said the MHRA spokesman.

Meanwhile the debate on how best to treat the various severities of depression goes on.

Exercise, diet, life-changes, and a plethora of therapeutic interventions are all said to have effect.

Kirsch’s study also again emphasised a significant role of placebo. And although many mental health charities broadly back the government’s move to expand cognitive-behavioural therapy, some campaigners and practitioners claim the focus on CBT is short-sighted.

“If a person’s living situation is dire, CBT may help them a bit, but it is not going to improve the situation they are actually in,” says Dr Andrew McCulloch of the Mental Health Foundation.

Moncrieff says: “CBT is better than drugs. Giving drugs to people is who are depressed is damaging. Anti-depressants are psychoactive drugs – its no different from putting someone on valium.”

Added into the mix are the thousands of anecdotal accounts praising SSRIs – such as that of Andrew Solomon, author of The Noonday Demon, a former Mind Book of The Year.

“I have been on Zoloft [Sertraline]for some years. It’s my experience that is has been demonstratively life-saving,” he says.

“I also have a less dire view of the pharmaceutical industry. Without this industry and the profit motive the drugs would never have been found. I would not be here if it was not for private enterprise.”

Solomon’s view is opposite to that of other users. For example, features comments from patients on Seroxat which include claims that suicidal tendencies have been exasperated by the drug.

Anecdotal evidence is fundamentally different to scientific evidence, it is said. Yet the scientists fundamentally disagree.

* A version of this article appeared in Mental Health Today magazine.

Leave a Reply