Lies and Damned Lies

Statistics – everyone’s favourite topic!

We have quite a problem that has existed for years and whilst I knew things were bad, I hadn’t realised the extent of it until last week.  The problem is – on sections 135 and 136 of the Mental Health Act – that we haven’t got a barking clue what is going on nationally because there are far too many problems with the data.

  • We don’t know how many times these powers are used.
  • We don’t know which buildings people are taken to as a Place of Safety under the Mental Health Act.
  • We don’t know the specific outcomes from 136.

I could go on!?

Of course, we know these things in general terms and in some local areas they do have excellent data that allows a full understanding of everything you could want to know.  Two-thirds to an NHS PoS, one-third (still!) to police custody and about a 20% admission rate after the assessment.  But if just a few significant areas of the country are unable to produce meaningful data then we will have to confess that we don’t know precisely what is going on overall.  Whilst already realising this, I learned something startling last week.


In debates over the last year or so it has been claimed that the low ‘conversion rate’ of section 136 detentions to admissions is itself evidence that the police threshold for the use of this authority is too low. Last year it was claimed that just 17% of detentions by the police resulted in a patient’s admission to hopsital and it was barely higher the previous year, at 20%.  Now there are all manner of other debates about whether the admission rate should be used to judge whether or not section 136 was used appropriately to start with and I firmly believe that it should not be used in isolation in that way.  I’ve even more reason for thinking this now!

A question was posed to me on an email this week about what the converstion rate actually was and I confidently retorted with the CQC published numbers: 17% last year and 20% the year before. A new colleague then added something to the mix which didn’t seem quite right to me so I thought I’d go back to the source documents and check – it turns out we were both wrong!

The Care Quality Commission publishes an annual report on the operation of the Mental Health Act and they rely upon data gathered by the Health and Social Care Information Centre. The HSCIC gather a ‘minimum data set’ from the NHS each year and it is then used for various reasons. When you look at the section 136 data you notice something that blew me away and which makes me realise we haven’t got a chuffing clue what is going on:

The 17% and 20% figures relate not to hospital admissions but to detained MHA patient admissions. I looked to find how many people detained under s136 MHA were then admitted as voluntary patients but it turns out we don’t know! We know how many people detained under section 136 subsequently had an ‘informal status’ but that figure was so large that it cannot realistically relate to those people who were just admitted informally – it must relate to more than that. Most likely, it indicates those who became or remained patients of the MH trust after release from s136 – both inpatients and outpatients.


We’ve known for years that the HSCIC data set includes on Hospital Place of Safety data. They described recent attempts to include what police station PoS data they can secure as ‘experimental’. More than that however, we can pretty confident that those patients who are wholly managed in A&E in areas where A&E is not ‘designated’ a PoS are not included either.

So when the HSCIC inform us that the power was used around 22,500 times in 2013/14 they issue caveats about experimental data and aren’t really sure. And against a background of not being entirely confident about how many times police officers remove someone’s liberty under this provision we don’t know what they do with them once they have.

We know that claims are made that around two-thirds of people access the NHS but we don’t always know precisely which kinds of facilities they access and we don’t know roughly how many are just not counted because they went to A&E and never reached a mental health unit PoS or police station.  We know in some areas that even if a multi-agency form exists to record all section 136 detentions, it isn’t always filled in.

Against this backdrop we are now spending millions based on the premise that the police are over-using section and applying too low a threshold for its application. But it turns out those data don’t relate to what the claimants think it relates to. And we don’t know what we don’t know about A&E and police custody.



We also know that various policy reviews are ongoing: the Mental Health Act Code of Practice is being reviewed for publication potentially in April. We know that the Government are reviewing legislation – and I’m wondering why they’d change anything if no-one can explain what the hell is happening now and therefore how things would change if the law was altered!?

Also, we know that both NHS England and the CQC are doing various types of urgent care reviews which touch on mental health crisis care to one degree or another. Don’t forget either the work that most areas should now be doing under the leadership of their health services on the Crisis Care Concordat.

How will we fully know what we need to do to make the world a better place if we haven’t got a clue what’s really going on nationally? – and in specific detail?!

I admit to wanting data, more data and even more data!

IMG_0053IMG_0052Winner of the President’s Medal from
the Royal College of Psychiatrists.

Winner of the Mind Digital Media Award.


4 thoughts on “Lies and Damned Lies

  1. I confess that I am less interested in the data than the specific detail. I dont doubt that the data it is important, It’s just from my seat, day to day other issues feel much more relevant.

    For a variety of reasons getting & co-ordinating the execution of a Sec 135(1) Warrant has become much more challengingn than it needs to be. More difficult that most reasonable people would expect. I doubt very much if the public at large or indeed those in charge of all this understand the difficulties faced by frontline practitioners.

    Did you know that during working hours AMHPs (in my part of the world) have to pay £22.00 to HMCS to get the warrant? They then have to convince a magistrate or District Judge to grant it. This has to be cash as it seems no arrangement or account can be set up between the LA & HMCS. Out of hours warrants can be gotten at no cost, simply because the on-call magistrate/judge will be at home.

    Co-ordinating the execution of the warrant can be like pulling teeth. Maybe after 30/40 phone calls & faxes & emails the bed/POS will be available, there will be a medic & the police & the ambo will have agreed to attend. You might have a lock smith & if there is a dog …….. then the medic or the police or ambo will be diverted & then the bed/POS will be used for another patient & the whole thing will start again.

  2. I’m shocked this data doesn’t exist and without such data influencing policy will be difficult to say the least. I think banging a few heads together to get some semblance of rationality regards evidence/data gathering relating to sec 136 is urgently needed.

  3. Do the conversion rates differ between urban and rural areas? Based on experience I’d expect cities, london in particular,to have high rates. If rates do differ, do you have any thoughts on why


  4. Stats should inform decisions (and opinions) but they are only advisory and no substitute for perception and what the Duke of Wellington called “being able to see the other side of the hill’.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s