Ed Sykes from the Science Media Centre explains why we must rely on research to inform investigative journalism in light of Panorama's recent episode on the link between antidepressants and homicidal behaviours.
Imagine a pill, taken by millions of people every day, that unwittingly turns some of us into murderers. That was the scenario laid out in the recent Panorama on antidepressants, it was the message trailed in the media in advance of the show and it was the line used in the press release to drum up interest. That’s a very powerful and scary message, and when you’re talking about a drug that saves thousands of lives every year - and that already has a bad public image - then we need to be extra careful. Putting people off antidepressants costs lives.
Antidepressants are not perfect, they have many side-effects, it can take a long time for people to find one that works for them and they don’t solve the problems that drive people into depression in the first place. People should be aware of these problems and we should discuss them openly. At the same time, we should be celebrating investigative journalism because it can be a massive force for good and I’d love to see more of it, but with any topic it needs to be done well, and when there’s a risk of people stopping a life-saving medication then that’s especially true.
So why was I concerned by the programme? What riled so many other people? I fully believe that antidepressants can change someone’s behaviour, and that we should be asking whether the negative side-effects are really outweighed by the benefits. But to do that, we need to see evidence, not simply be given a case study.
The programme focused almost exclusively on the case of James Holmes, known as the Batman killer after opening fire in a cinema showing the Dark Knight Rises. We were shown the timeline of events leading up to the massacre, we heard from his parents and from a couple of psychiatrists involved in the case. But where was the evidence that the murders were linked to the medication?
I spend my days working with researchers on all aspects of mental health and neuroscience, these are people who dedicate their lives to sifting through the data to see whether we are imagining patterns or are blind to what the stats are telling us. Many of them are investigating exactly these questions around antidepressants.
Professor Seena Fazel is known around the world for his work investigating whether antidepressants lead to an increase in violence. The research doesn’t show any evidence for an increase in people over 25. In younger people, there is evidence of a slight increase but we still can’t say if this is due to antidepressants. It’s complex as young people are taking more alcohol and other recreational drugs, and some of these people are actually suffering from severe personality problems that make them violent in the first place. It’s fascinating work using data from tens of thousands of people and I would have sat engrossed for an hour hearing about what the figures show. But even if Seena’s work did show antidepressants caused people to be more violent, that would still be a big step away from turning someone into a murderer.
It’s not just Seena and his datasets who don’t see the evidence for saying antidepressants turn people into murderers. I contacted around 100 researchers who work on various aspects of depression and antidepressants and the only message that kept coming back to me was that there was just no good evidence to suggest that antidepressants turn people into murderers. Yet in the show we were presented with one psychiatrist who said he thought the drugs weren’t to blame, one who thought they definitely were and one who thought we should be looking into it. It seemed as though the field is split evenly on this and that the evidence is up for grabs – this isn’t representative of the views of the mainstream science community; it may be ‘balanced’ but it isn’t measured. This kind of false balance has been blogged about many times before.
The main evidence the programme seemed to be giving us, alongside the case studies, was the result of a Freedom of Information request to the Medicines & Healthcare products Regulatory Authority (MHRA). The MHRA is responsible for many aspects of healthcare, including a Yellow Card system that records all the bad side-effects linked to any drugs. The Panorama team informed us that their investigation had unearthed 28 cases linking antidepressants to murder and 32 to murderous thoughts. The show did point out that these were just reports, not conclusions that the drugs had caused murder. But what they failed to mention was that these reports to the MHRA can be made by anyone. You or I could go on their website right now and make a report, it doesn’t necessarily mean it was a medical professional who reported it or that the report was supported by strong evidence. Maybe each of these reports was made by a healthcare professional and was supported by good evidence, but Panorama didn’t show it to us.
The programme did well in reminding viewers not to stop taking medication without speaking to a medical professional first, but it also kept repeating that in rare circumstances these drugs might be turning people into murderers. The EU defines a rare disease as affecting fewer than 5 in 10,000 people, but when there are a lot of people involved then even a rare event should start to appear quite a lot. Around 1 in 10 people in the UK experience some form of depression during their life, there were 61 million prescriptions for antidepressants in 2015 alone, probably more than 500 million during the last three decades. So even if all 60 of the Yellow Card reports to the MHRA had all definitely been a case of the drug turning someone into a murderer that would be around 1 per 800,000 prescriptions. There are only 600,000 people living in Glasgow. These circumstances aren’t rare, they’re vanishingly small.
What viewers of this Panorama didn’t see was a measured, proportionate account of the huge wealth of evidence on this topic. Through false balance, by not showing the evidence, and by not giving some much-needed context what could have been an insightful programme left people misinformed. Antidepressants would benefit from investigative journalism – there are good questions about over-prescribing and prescribing for the wrong reasons, but there are also concerns about under-prescribing and how a really useful drug is still being shunned by many. With a thorough investigation we could learn a lot about how much the drugs do change people’s personalities, whether using them stops us from investing in psychological therapies, if they result in social problems being ignored and even whether they do lead to violence or murder – but to have that debate we need to see the evidence, not just rely on one case study, no matter how distressing.
Ed Sykes is Head of Mental Health and Neuroscience at the Science Media Centre.
Last updated: 28 July 2017