Did The Guardian exaggerate its comparison of prison versus hospital safety?

Data on preventable deaths in police custody, prisons and psychiatric care

Photo:  Wellcome Trust via Wikimedia Commons

Photo: Wellcome Trust via Wikimedia Commons


In February this year the Equality and Human Rights Commission (EHRC) published a report assessing the scale of preventable deaths of mentally ill people in custody, prisons and psychiatric wards.

Whilst most people reported the total number of deaths, The Guardian went a step further in its analysis. It concluded that based on the data in the report:

Patients detained in hospital psychiatric wards are up to five times more likely to suffer a preventable death than mentally ill prisoners in the prisons of England and Wales”.

Was this comparison a step too far? Prison Watch investigates.


What does the report say?

Firstly, the report defined preventable or ‘non-natural’ deaths as:

“One of the following categories: self-inflicted/suicide, deaths caused by another person including homicide, other non-natural deaths including overdose and accidental deaths and deaths the cause of which is unknown.”

It also stated that prisons are still not recording the mental health of their inmates. The report cited the following as the most reasonably accurate statistic on the number of inamtes suffering from mental health issues:

“The most recent data relates to 1997, where 92 per cent of male prisoners were reported to have one of the following five conditions: psychosis, neurosis, personality disorder, alcohol misuse and drug dependence. Seventy per cent had at least two of these.”

It also said that in hospitals – unlike prisons and police settings – there is no body charged to carry out investigations into the non-natural deaths of detained patients, leaving many questions unanswered for the loved ones of the deceased.


The Data

The report found that across police custody, prisons and psychiatric care  there was a total of 662 preventable deaths. While deaths in police custody have been the topic of much controversy in the media, psychiatric wards and prisons suffer from a significantly higher number of preventable deaths; something the EHRC wished to emphasise.


Data table taken from the EHRC report

Data table taken from the EHRC report


The Independent and NHS news website covered this report focusing on the headline figures of 367 preventable deaths in psychiatric wards and police custody, and 295 preventable deaths in prisons.


deaths mental health


This is how the preventable death figures look as a percentage of total deaths. Proportionally they are roughly the same, with prisons having a greater proportion of preventable deaths than hospitals, about 35 per cent and 30 per cent respectively.

However, this comparison is not fair. As the report highlighted, the mental health of prisoners is not properly recorded. In order to make a fair comparison one would have to calculate the percentage of prisoners who suffered a preventable death that also suffered from equivalent mental health issues, to those who needlessly died in hospitals.

In an attempt to make this cross comparison the Guardian reported on the data saying:


Guardian Headline for thier article on the ECHR report.

Guardian Headline for their article on the ECHR report.


Their coverage led with the following statistic:

Patients detained in hospital psychiatric wards are up to five times more likely to suffer a preventable death than mentally ill prisoners in the prisons of England and Wales”.

A statistic like this instantly seems to bring the whole philosophy of providing care for mentally ill prisoners rather than criminal detention, instantly into question.


How did they get this figure?


The author of The Guardian article, Randeep Ramesh, said he based his calculations roughly on the following assumptions:

  • The number of people detained under the mental health act in 2012/13 – 17,000 (according to the Care Quality Commission)
  • Total number of people in prison for that year 84,392 (for last week in December 2013 – Howard League)
  • Preventable deaths for the year 2013.

Assuming that one uses the lower figure of 70 per cent of the prison population suffering from at least two mental health conditions (see beginning of article) the calculations are as follows:


Psychiatric Care: 87/17,000 = 0.5%

Prisons: 84/(70% of 84,392 (59,074.4) ) = 0.01%


This calculation says 50 times more likely making the ‘five times’ figure seem very conservative.

But would all 59,074 prisoners, if not in prison, be sectioned under the Mental Health Act or be in a similar condition? Surely not.

Mr. Ramesh claimed that even if you use the lower estimates of the prison population said to suffer from the most extreme mental health issues you still end up with a preventable death ratio higher than that found in hospitals.

Using a lower bound estimate is likely to come from the Prison Reform Trust’s statistics and analysis. They claim that:

  • 10 per cent of men and 30 per cent of women have had a previous psychiatric admission before they entered prison. Another study found that 25 per cent of women and 15 per cent of men in prison reported symptoms indicative of psychosis. The rate among the general public is about four per cent.
  • Personality disorders 62 per cent of male and 57 per cent of female sentenced prisoners have a personality disorder.
  • Anxiety and Depression49 per cent of women and 23 per cent of male prisoners in a Ministry of Justice study were assessed as suffering from anxiety and depression.


Are there any plausible reasons why death rates might be higher in hospitals than in prisons?

Despite a potentially higher rate of preventable deaths, the following reasons could provide possible explanations to this statistical difference.

  • More people detained under the Mental Health Act are being placed in hospitals rather than in police custody. Of the 23,036 Place of Safety orders made using Section 136 of the Mental Health Act, the proportion where the individual went to a hospital rather than a police custody increased from 64 per cent (14,053) during 2012/13 to 74 per cent (17,008).


Data Suggesting

Data from the Health & Social Care Information Centre shows an increasing number of Section 136 Place of Safety orders are being used to detain people in psychiatric care rather than police custody.


  • Most recent figures suggest that there has been a slight overall increase in the number of prisoners moved to hospitals with more severe mental health illnesses.


Is a comparison even possible?

Richard Colwill from SANE, a mental Health Charity, thinks that a comparison like this is just not that easy to make.

“I think you need to be extremely careful because the entire foundation and the comparison for a story like that rests on the estimate that you use for number of people in prisons with a psychiatric condition”

“It doesn’t make any distinction between conditions, like depression for example … is markedly different from a psychotic condition like bipolar disorder or schizophrenia. And the relative risk of a condition like that may be very different from somebody with depression.”

Equally, people detained under the Mental Health Act will have been detained for a variety of reasons and varying mental health conditions.

In summary, it was a rough estimate. A very rough estimate. The tactical inclusion of “up to” within the ‘five times more likely’ claim acts as a disclaimer, when trying to falsify Ramesh’s statistical statement.

Prisons should adopt the EHRC recommendations accordingly, and properly measure the mental health of our prisoners. Only then will we be able to make a fairer and more accurate comparison, to determine the future of mental health care policy in our prisons.



TWITTER: @prisonwatchuk 

FACEBOOK: facebook.com/PrisonWatchUK

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s