A Wholly Inadequate response from Ben Goldacre (and my reply)

With great reluctance, I publish the following 'response' from Dr Ben Goldacre to my posting yesterday. I do not believe it even addresses the points I made. It dissipates the respect I formerly had for him. That is why I begged him, on Twitter last night, to try again. The response has been silence.

 

'Hi Peter,

 

I’m working on the ward today so not able to reply immediately to your dozen tweets in response to mine. Big fan of your writing, as I’ve said before, tho clearly we disagree on a lot.

 

I don’t think my tweet can be construed as misleading anyone about your words, since the entirety of your text that worried me was in the screengrab posted in my tweet, and your entire column was linked to in the tweet. I tweeted it (yesterday?) because that’s when someone tweeted it at me.

 

I think your suggestion that terrorist atrocities can be prevented by widespread pre-emptive incarceration of people with mental health problems is disproportionate, offensive to those with mental health problems (who make a tiny contribution to the total amount of violence in society), and unjust.

 

I also think it would be unlikely to work, for the reasons I set out in the piece linked below, which is on the difficulties of accurately predicting the risk of rare events (such as individual acts of terrorist violence) in large populations:

 

http://www.badscience.net/2006/12/crystal-balls-and-positive-predictive-values/

 

The key quote – last paragraph - is pasted below, but the whole piece is needed as background to the maths really:

 

Now let’s look at violence. The best predictive tool for psychiatric violence has a “sensitivity” of 0.75, and a “specificity” of 0.75. Accuracy is tougher, predicting an event in humans, with human minds, and changing human lives. Let’s say 5% of patients seen by a community mental health team will be involved in a violent event in a year. Using the same maths as we did for the HIV tests, your “0.75” predictive tool would be wrong 86 times out of 100. For serious violence, occurring at 1% a year, with our best “0.75” tool, you inaccurately finger your potential perpetrator 97 times out of a hundred. Will you preventively detain 97 people to prevent three events? And for murder, the extremely rare crime in question, occurring at one in 10,000 a year among patients with psychosis? The false positive rate is so high that the best test is almost entirely useless. I’m just giving you the maths on rare events. What you do with it is a matter for you.

 

I didn’t know about the Rigby killer’s mental health history, thanks for that, was just discussing it with a forensic psychiatry colleague at lunch, will have a read around if I get a chance. But this fact doesn’t change the concerns set out above with your proposals for massive pre-emptive incarceration of mental health patients on a national scale.

 

I hope we get to meet one day.

 

Ben'

 

I replied to this as follows:

 

Dear Ben,

 

I do urge you to rethink this rather complacent  response.  It has been ‘misconstrued’ or perhaps correctly construed as the damaging smear it was from the start. Did you (or your correspondent) not know that the Paris murders had happened at the time you posted that tweet?  You claim an interest in my writing, but did not respond to this at the time it was written, or for weeks afterwards. Did it not occur to you that immediately after the Paris atrocity  it would be seen as a comment on that event and its implications? Others did, and have posted  ‘Peter Hitchens blames care in the community for terrorist attacks’. Another has posted ‘Some days I think Peter Hitchens is a reasonably intelligent guy with whom I happen to disagree. Then I see this’, ‘This’ being a reproduction of your extract from my December article, which of course most people will read without following the link which provides the rather different context. (as it happens, for what it’s worth, we now know that one of the Paris suspects is a known heavy cannabis user).  Another has posted ‘A thoroughly deranged assessment of modern terrorism from Peter Hitchens’ (also reproducing your extract).

 

Your summary ‘Peter Hitchens believes terrorists are on drugs and in the 1980s would’ve been in asylums. Jesus’ is (as I have demonstrated) misleading and inaccurate. To try to turn an attack on securocrat spending and a call for higher health spending into some sort of attack on the mentally ill is ridiculous, and I think you know it. I didn’t say anything about the number of mentally ill people who commit violent crimes, though I did examine this topic here in May 2013 http://hitchensblog.mailonsunday.co.uk/2013/05/dont-care-in-the-community.html

 

I repeatedly make the point in these and other articles that we may not know about these cases, because nobody is actually asking the questions which would enable us to. I make no claims of knowledge that I don’t possess. It doesn’t fit the conformist narrative,  which was why you ‘didn’t know’ about the Rigby killer’s mental health history. You hadn’t looked. It’s not hard to find if you do.

 

I don’t care if you disagree with me. But I think you of all people should do it in a civil, thoughtful fashion.  If you’re not even slightly abashed at your behaviour over this, then my opinion of you has fallen sharply.

 

 

Sincerely,

 

Peter Hitchens

 

I'd add here, on reflection, that to characterise my article as a 'suggestion that terrorist atrocities can be prevented by widespread pre-emptive incarceration of people with mental health problems' is a ludicrous caricature of what I actually said.

 

This (I remind readers) came after a description of the plainly irrational behaviour of several violent murderers, whose crimes are undoubted and who exhibited multiple examoples of irrayional behaviour long before they committed these crimes. It was as follows: '...in the days before ‘care in the community’ they would not have been able to kill because they would have been in mental hospitals. Such hospitals would be a much better use for all the money we currently pour into grandiose ‘security services’.

 

1.It's mainly a statement about the past. There is no 'call for widespread pre-emptive incarceration of people with mental health problems'. the word 'widespread' is nt used, nor anything resembling it, nor anyting that could truthfully be interpreted to mean 'widespread' or 'incarceration'.  

 

Does Dr Goldacre object to the use of sectioning and the detntion of somke menatlly ill people in locked wards today, as is regrettably common? Does he think that such actions are in some way a defamatory reflection all people classifed as 'mentally ill', and does he argue against thenm on those grounds.

 

The article expresses a specific concern that violently unhinged people who should be hospitalised (and would have been before the cheapskate modern policy of closing hospitals took hold) are left free to roam about. The article to which I link above demonstrates that for several British people the outcome of this policy has been horribly fatal.

It also objects to something which I believe Dr Goldacre objects to, the ever-tightening surveillance society created on the pretext of fighting terrorism. 

 

I believe a serious person of any political persuasion could find a lot to agree with in this article, and might write to me personally to express any disgareements. Instead, I received a public smear adorned with a blasphemy. 

 

Dr Goldacre's behaviour, and his unwillingness to retreat or apologise when shown up, contrast sharply with his image as the Knight Errant of rigorous truth.