These Guardian / Independent stories are dodgy. Traps in data journalism.

December 30th, 2011 by Ben Goldacre in guardian, numerical context, statistics | 13 Comments »

Here’s an interesting problem with data analysis in general, and so, by extension, data journalism: you have to be careful about assuming that the numbers you’ve got access to… really do reflect the underlying phenomena you’re trying to investigate.

Today’s Guardian has a story, “Antidepressant use in England soars“. It’s much more overstated in the Independent. They identify that the number of individual prescriptions written for antidepressant drugs has risen, and then assumes this means that more people are depressed. But while that’s a tempting assumption, it’s not a safe one.

Thinking off the top of my head, it could be – for example – that doctors are writing more frequent prescriptions for the same number of patients, but with each prescription for smaller amounts (to reduce overdose risk, say). These potential alternative explanations are the sort of thing that comes up all the time in data analysis for medical research.

In fact, this specific question – what does an increase in antidepressant scripts mean? – has been researched in some detail before. I wrote about it in April 2011, the last time this rise was written up as a big story, in several major newspapers, including the Guardian. I guess nobody listens to me, and fair enough.

When journalists do primary research

…Firstly, this rise in scripts for antidepressants isn’t a new phenomenon. In 2009 the BMJ published a paper titled “Explaining the rise in antidepressant prescribing”, which looks at the period from 1993 to 2005. In the 5 year period from 2000 to 2005 – the boom before the bust these journalists are writing about – antidepressant prescribing also increased, by 36%. This isn’t very different to 43%, so it feels unlikely that the present increase in prescriptions is due to the recession.

That’s not the only problem here. It turns out that the number of prescriptions for an SSRI drug is a rubbish way of measuring how many people are being treated for depression: not just because people get prescribed SSRIs for all kinds of other things, like anxiety, PTSD, hot flushes, and more; and not just because doctors have moved away from older types of antidepressants, so would be prescribing more of the newer SSRI drugs even if the number of people with depression had stayed the same.

Excitingly, it’s a bit more complicated than that. A 2006 paper from the British Journal of General Practice looked at prescribing and diagnosis rates in Scotland. Overall, again, the number of prescriptions for antidepressants increased from 1.5 million in 1996 to 2.8.million in 2001 (that is, it almost doubled).

But they also found a mystery: looking at Scottish Health Survey, they found no increase in the prevalence of depression; and looking at the GP consultations dataset, again they found no evidence that people were presenting more frequently to their GP with depression, or that GPs were making more diagnoses of depression.

So why were antidepressant prescriptions going up? This puzzle received some kind of explanation in 2009. The BMJ paper above found the same increase in the number of prescriptions that the journalists have found this week, as I said. But they had access to more data: their analysis didn’t just look at the total number of prescriptions in the country, or even the total number of people diagnosed with depression: it also looked at the prescription records of individual patients, in a dataset of over 3 million patients’ electronic health records (with 200,000 people who experienced a first diagnosis of depression during this period).

They found that the rise in the overall number of antidepressant prescriptions was not due to increasing numbers of patients receiving antidepressants. It was almost entirely caused by one thing: a small increase in the small proportion of those patients who received treatment for longer periods of time. Numerically, people receiving treatment for long periods make up the biggest chunk of all the prescriptions written, so this small shift bumped up the overall numbers hugely.

I don’t know for certain if that phenomenon explains the increase in prescriptions from 2006-2010, as it does for the period 2000-2005 (although in the absence of work examining that question, since the increase in scripts was so similar, it does seem fairly likely). And I’m not expecting journalists to go to academic research databases to conduct large complex descriptive studies.

But if they are going to engage in primary research, and make dramatic causal claims  – as they have done in this story – to the nation, I don’t think it’s too much to ask that they familiarise themselves with proper work that’s already been done, and consider alternative explanations for the numbers they’ve found.

Incidentally, if you’re missing the column, I’m procrastinating on Twitter, and posting occasionally on posterous. I’ll stick a round-up of the most interesting things from there onto here occasionally, and there’s also a backlog of columns to pop up here too, from when I was too busy to breathe. The new book is in fighty form, thanks for asking, out in August 2012.

If you like what I do, and you want me to do more, you can: buy my books Bad Science and Bad Pharma, give them to your friends, put them on your reading list, employ me to do a talk, or tweet this article to your friends. Thanks! ++++++++++++++++++++++++++++++++++++++++++

13 Responses

  1. OrganisedPauper said,

    December 30, 2011 at 2:10 pm

    Our GP surgery went from prescribing medication to last for 3-6 months, or more, to prescribing for only 28 days, or 56 days for longer term medication. That sort of change in prescribing practice must surely have an affect on data?

  2. bencurthoys said,

    December 30, 2011 at 3:02 pm

    > a small increase in the small proportion of those
    > patients who received treatment for longer periods of time

    Does the small increase in number of people receiving longer term treatment correspond to how good the drugs are at stopping people from committing suicide, by any chance? Are they matched by a fall in death rates by the people taking them?

    That is, are the numbers actually evidence of the drugs working?

  3. skoodle-um-skoo said,

    December 30, 2011 at 6:02 pm

    anecdotally here – I suffered depression a few years ago and was treated with an SSRI. I was mainly seen by an old GP who was happy for me to stop treatment after 6 months. However I didn’t always see him and the younger GPs I saw in the practice would always encourage me to take the tablets for at least a year. I followed the older GPS advice, mainly because as I got better I didn’t like the way the drug made me feel – I guessed that serotonin was being boosted above a normal level by the drug now that I was ‘better’.

  4. SimRos1000 said,

    December 30, 2011 at 7:07 pm

    The other thing to take into account is that the media also reduced the number of scripts written for SSRIs (remember “happy pills causes suicides” etc), so the numbers would have gone down and now they are going back up again.

    I also wouldn’t be surprised by how pharma could increase the number of prescriptions — definitely ensuring more people get more drugs more of the time, irrespective of the number of diagnoses of depression.

    So you are right, the last thing the correlation would be to is to economic circumstances

  5. WilliamOfCrockham said,

    December 30, 2011 at 10:14 pm

    Well done.

    Now, how about that post on generics?

  6. slimmillipede said,

    December 31, 2011 at 5:15 pm

    I know that our local surgery has reduced the quantity in each prescription. So I now use twice as many prescriptions per year to manage my blood pressure, but exactly the same number of tablets.

    I imagine that the same is likely for other surgeries and other drugs.

  7. DonBarton said,

    January 1, 2012 at 2:41 pm

    I was recently doing some research for a gentleman who does not have access to the internet and came across a piece by the BMA called “Information and guidance on prescribing in general practice – Prescription intervals”
    I quote “Doctors are sometimes now being put under pressure to prescribe at 7 day intervals simply in order that the pharmacist can be reimbursed for ‘medidose dispensing’ when there is no other payment currently available. Pharmacists and dispensing doctors may prefer 28-day intervals for reasons of reimbursement and financial viability of a dispensary, but many other factors should be considered.” Pressure comes from many sources and these are not always for the benefit of either the patient or the GP and do not indicate an increased use of antidepressants.

  8. mikewhit said,

    January 4, 2012 at 3:15 pm

    So why is the govt./NHS not collecting total dosage prescribed, or total individuals prescribed (or somesuch), rather than number of prescriptions, which is prone to the kind of mis-interpretation seen ?

  9. Geoffrey Swenson said,

    February 1, 2012 at 4:35 pm

    It is apparently politically correct to rail about anti-depressants in the progressive media. The way this is presented reminds me more of the black and white opinions of the far right than the usual thoughtful, nuanced, fact-based opinions of progessives on most other issues.

    In this black and white world the drug companies are evil institutions creating fake research results about made-up conditions to sell their anti-depressants.

    The reality is a bit different than this. Some of these drugs do work, and I know by experience. There may be huge flaws in the testing and marketing of drugs by for-profit companies. But it isn’t as completely broken as all of the hyperbole you read on liberal-leaning publications would suggest. Some drugs are quite effective and really do save people’s lives.

  10. dr_z said,

    February 8, 2012 at 11:32 am

    I have been on anti-depressants for two years now and in that time there has been a move to not issuing more than one month’s worth of tablets when in the past I could get two or three month’s worth. Thus, my number of prescriptions has at least doubled in 2011. This clearly does not mean that I split into two depressed people.

    Nevertheless, this strikes me as good practice, especially in the case of mental health issues. My GP in particular seems committed to making sure that I am in fact doing well on the medication.

  11. DicostanzoFriesner said,

    April 16, 2012 at 8:51 pm

    Does the small increase in number of people receiving longer term treatment correspond to how good the drugs are at stopping people from committing suicide, by any chance?

  12. teabroughten said,

    June 1, 2012 at 9:34 am

    All the media exsiting with their powerful array of weapons for manipulation are successfuly washing our brains on daily basis- even the harmless presenting of information shows preliminary the point of view of only one of the involved sides- everybody claims for objectivity but there is no such thing. The specator is left on its own intuition and intelligence to decide what version to believe in.

  13. jratb said,

    August 12, 2012 at 1:45 pm

    Of course, and I’m not sure you will like this idea, there is always the GP’s knowledge that people “should be” depressed because of the recession, therefore more likely to diagnose & prescribe? It is something that should be considered. Without wishing to go all Illich on your arse, the word “iatrogenic” springs to mind.