Individual at Gartree

Investigation into the circumstances surrounding the
death of a male prisoner on 12 October 2006 in
hospital, while in the custody of HMP Gartree
Report by the Prisons and Probation Ombudsman
for England and Wales
November 2007
This is the report of an investigation into the circumstances surrounding the
death of a male prisoner. It raises some significant issues.
The man died on 12 October 2006 in hospital, while a prisoner in the custody
of HMP Gartree. He had been admitted to hospital on 10 October after
complaining of abdominal pain. The hospital discovered that the man was
suffering from peritonitis as a result of a perforated bowel, but he was too ill to
undergo surgery and died two days later.
I would like to offer my condolences to the relatives and friends of the man
who died. He was an elderly man who had been in poor health for some
years. Nevertheless, his relatively sudden death came as a shock to his wife
and family. Losing someone while they are in custody only adds to the pain of
any bereavement.
The investigation was led by one of my investigators. One of my family liaison
officers contacted the man’s family to ask them if they had any questions or
concerns about his death. In addition, an independent review into the man’s
medical care and treatment was undertaken by a local GP. I am most grateful
to this GP for his assistance. I am also grateful to the Governor, staff and
prisoners of Gartree prison for their co-operation with this investigation.
The clinical review makes seven recommendations and highlights a particular
concern about the triage policy in operation at Gartree. Although the delay in
taking the man to hospital cannot conclusively be said to have hastened his
death, there is a real concern that healthcare staff are making diagnoses and
prescribing treatment they are not qualified to make. I am also very critical of
the fact that the man remained chained to one of his escorting officers until
some 35 minutes before he died. He was gravely ill throughout his brief stay
in hospital, and I consider that use of the escort chain in these circumstances
was unnecessary and distressing for the man who died, his family and the
attending officers.
I make six further recommendations endorsing those arising from the clinical
review and addressing the issue of the use of restraints on prisoners during
hospital escorts and bedwatches.
Stephen Shaw CBE
Prisons and Probation Ombudsman
November 2007
CONTENTS
Summary 4
The investigation process 5
HMP Gartree 7
The events leading up to the man’s death 8
What other prisoners said 11
Issues considered during the investigation 13
Recommendations 21
SUMMARY
The man who is the subject of this report was 80 years of age when he died.
He had been sentenced to life imprisonment in 1999 with a minimum time to
serve of ten years. He suffered from numerous chronic medical conditions,
including heart disease, lung disease and diabetes, and used a wheelchair.
He was transferred to HMP Gartree in March 2001 and was initially located on
the healthcare centre before being moved to a specially adapted cell on G
wing in June 2005.
On 6 October 2006, the man complained of abdominal pain and vomiting.
Several prisoners and staff were concerned about his condition. He was seen
by a member of healthcare staff on the same morning and diagnosed as
suffering from food poisoning. He was seen again on 7 and 8 October by the
same member of staff, and on 9 and 10 October by another healthcare officer.
The initial diagnosis of food poisoning was not changed and the man was not
referred to a more senior member of the healthcare team. Despite the fact
that the man was complaining of severe abdominal pain, neither of these
members of staff performed an abdominal examination. Both are trained
mental health nurses.
On the morning of 10 October, the man was found to be in a “dire” state. At
8.40am, he was referred to hospital and was taken to the healthcare centre
while the prison waited for a hospital bed to become available.
The man was taken to a local hospital at 12.30pm where he was diagnosed
as suffering from peritonitis as a result of a perforated bowel. He was
considered by the hospital doctors as too ill to survive an operation. His wife
visited him on Wednesday 11 October and he died at 9.37am on 12 October.
The post mortem gave the causes of death as peritonitis, perforated small
bowel diverticulum, ischaemic heart disease and chronic obstructive
pulmonary disease (lung disease).
The clinical review makes seven recommendations which I endorse. In
particular, I draw attention to the comments made about Gartree’s triage
policy. The review concluded that this system had failed the man who died in
that the severity of his illness was not discovered for some five days. The
review also concluded, however, that an earlier admission to hospital would
not have prevented the perforation of the man’s bowel and, given his
underlying chronic medical conditions, would have been extremely unlikely to
have prevented his death.
Apart from three brief periods on 11 October, the man was chained to an
escorting officer with an escort chain until some 35 minutes before he died.
Circumstances such as these are distressing for the deceased, their family
and the officer who remains chained to a dying person. I make a
recommendation to the Governor of Gartree that she review the prison’s
guidance to staff to bring it in line with the requirements of the National
Security Framework (NSF). I also make a national recommendation to the
Prison Service to review the NSF to give guidance on the removal of
restraints from dying prisoners on compassionate grounds.
THE INVESTIGATION PROCESS
1. I was notified of the man’s death on 12 October 2006. Notices were
issued to staff and prisoners at Gartree telling them that an investigation
would be taking place and inviting those who wished to see the
investigator to make themselves known. The Coroner was contacted
and a copy of the Post Mortem report was requested and received.
2. My investigator visited Gartree on 17 October 2006. She met the
Governor, and a representative from the Independent Monitoring Board.
She was provided with the prison record of the man who died, and
copies of the notices, reports and other records associated with his
death. She visited the wing and saw the man’s cell and spoke informally
to some prisoners and staff. My investigator had a long telephone
conversation with the Head of Healthcare at Gartree on 19 October.
3. On 6 November, my investigator received a letter from a prisoner at
Gartree telling her that several prisoners would like to speak to her about
the man’s death. She spoke to this prisoner on the telephone on two
occasions. She then visited Gartree on 28 November and interviewed
eight prisoners.
4. A clinical review of the medical care of the man who died while he was a
prisoner in Gartree was commissioned from Leicestershire County and
Rutland Primary Care Trust. A local GP conducted the clinical review.
5. One of my family liaison officers contacted the man’s family. My family
liaison officer and my investigator subsequently visited the man’s wife,
eldest son and daughter in law.
6. The man’s family was concerned that the hospital had not made it clear
to them how ill he was when he was admitted on 10 October. Only his
wife had visited him in hospital before he died, and the family was upset
as they said they would have all visited had they known how ill he was.
The man’s daughter in law was disappointed that, when she had called
the hospital to find out how ill he was, a nurse had passed her on to one
of the prison officers escorting her father in law.
7. The man’s family was concerned that he had been ill for some time and
should have been admitted to hospital earlier. They thought he had been
feeling ill since Friday 6 October. His wife said she had been worried
because he always rang her every evening, but she had not heard from
him during the week before he was admitted to hospital.
8. The man’s wife was also upset that her husband had been “chained up”
while in hospital. The family expressed surprise that this had been
necessary as the man had been unable to walk, had been attached to
several tubes and had needed an oxygen mask to breathe. The man’s
wife said that the escorting officers had obviously been uncomfortable
with the “chains”, and one of them in particular had been very keen to
take them off.
9. The man’s family said that prison staff had attended his funeral and had
brought a wreath. The prison’s family liaison officer had been helpful and
had offered practical assistance. They had found it especially helpful to
visit the prison and see the man’s cell. His family said they had been
touched by the concern of the other prisoners and a wreath that one of
them had placed on his cell door.
HMP GARTREE
10. Gartree is a category B prison whose principal function since 1997 has
been to accommodate and rehabilitate adult male mandatory life
sentence prisoners. The average tariff (minimum time to serve) for
these prisoners is 15 years. Around 18 per cent of the population now
consists of prisoners sentenced to indeterminate sentences for public
protection.
11. Gartree is part way through a major refurbishment which will continue for
the next two years. When complete, it will give Gartree a certified
normal accommodation of some 680 prisoners and make it the biggest
lifer centre in Europe. The rebuilding project and the associated
disruption has caused some uncertainty among staff and prisoners, and
staffing levels have been lower than expected for some years. Staffing
levels have recently begun to increase and the Governor told my
investigator that the psychology department alone now stands at 43
members. Gartree runs more courses for prisoners than any other
prison in the country, although the waiting lists for these remain very
long.
12. The death of the man who is the subject of this report was the second of
five deaths to have occurred at Gartree since I was given the
responsibility for investigating all deaths in prison custody in April 2004.
In my investigation into the first death, the clinical reviewer made similar
comments to those of the GP in the current case about the difficulties of
understanding prisoners’ medical records at Gartree. At the time of
writing this report, the clinical reviews into the other deaths were not
available.
KEY FINDINGS
From 6 October until the man was taken to a local hospital on 10
October
13. The man who died first complained of feeling ill on 6 October 2006. His
medical record shows he was seen by a qualified mental health nurse
(RMN). The man complained of vomiting and stomach pain and was
diagnosed with food poisoning. In the GP’s clinical review, he says that
the RMN recalled in interview that the man said he had eaten some
pilchards which had been kept in his cell unrefrigerated for four days.
There is no formal record of this. The man’s pulse, blood pressure,
respiratory rate and temperature (his ‘baseline observations’) were taken
and found to be normal.
14. On 7 October, wing staff contacted the healthcare centre after the man
complained that he was still being sick and still had abdominal pain.
They called the healthcare centre an hour later because the man’s
condition was worsening. The man was seen on the wing by the RMN
again and was reassessed. The man said that he had drunk plenty of
water and was given sachets of oral rehydration powder. His baseline
observations were taken and found to be normal. He was advised to
stay in bed and alert staff if he felt worse.
15. The man’s medical record shows that he was seen again on 8 October
by the same RMN. The only detail of the assessment is that his
baseline observations were again normal.
16. On 9 October, a second RMN was on G wing assessing another patient.
Wing staff asked her if she would see the man. The second RMN wrote
in the man’s medical file, “still complaining of D + V [diarrhoea and
vomiting], given more dioralyte [oral rehydration powder], to see doctor
in morning.” In interview, she told the GP who conducted the clinical
review that the man’s baseline observations were taken and were
normal, but there is no record of this on his medical file. The second
RMN told the clinical reviewer that she remembered contacting the first
RMN and accepting his diagnosis of food poisoning. The second RMN
said she was happy with the man’s clinical condition and did not feel the
need to involve more senior staff.
17. On 10 October, the man was seen again by the second RMN at 8.40am.
His condition was described as “dire” with pallor, sweating and difficulty
in breathing. He was referred to hospital and taken to the healthcare
centre where he received oxygen therapy. He was taken by ambulance
to a local hospital at 12.30pm when a bed became available.
The man’s time in hospital between 10 and 12 October
18. Once the man had been referred to hospital, a hospital risk assessment
form was completed at Gartree. This is a standard form which must be
completed in all cases when a prisoner is escorted to outside hospital.
Its purpose is to ascertain the appropriate level of escort and type of
restraint to be used. In the section to be completed by prison medical
staff, it is recorded that there were no medical objections to the use of
restraints in the case of the man who died. There is a note to the effect
that a “closeting chain” should be used if possible. (A closeting chain is
otherwise known as an escort chain. It is a length of chain attached to a
single handcuff at each end. The prisoner wears one cuff and the
escorting officer wears the other.) The security assessment was that the
man represented a medium risk to the public, no risk to staff, a medium
risk of taking a hostage and a medium risk of escape. The escort
conditions were that the man should be accompanied by two members
of staff and that restraints should be used at all times. The reason for
this was recorded as “cat B prisoner serving life sentence.” This level of
escort and restraint was confirmed by the Head of Operations who wrote
on the form:
“Use of escort chain due to his mobility and illness. Staff to
consult D/Gov [Duty Governor] if removal required.”
19. The Prisoner Escort Record (PER) completed on 10 October shows that
the man arrived at the hospital’s Clinical Decision Unit at 1.10pm. The
section on risk categories on the front of the form shows indicators for
“medical condition” and “violence”. The medical condition was explained
as “heart”. The indicator for violence was added because the man had
been convicted of a violent crime (murder). The record of events section
on the PER form shows that the man was seen by a doctor at 4.00pm,
and that the doctor told the escorting staff that the man would be in
hospital for “at least 24 hours” whilst tests were completed. The PER
form shows that at 5.15pm the man had a catheter inserted. At 6.00pm,
the escorting officers started a hospital bedwatch form. At 6.40pm, the
man had x-rays taken and at 8.25pm a nasal tube was inserted.
20. A second hospital risk assessment form was completed by the Duty
Governor, on 11 October. It is not clear at what time this was
completed. The Duty Governor wrote that the escort chain was to be
used at all times and should only be removed in a medical emergency or
for routine treatment and with the permission of the Duty Governor.
Family visits were to be allowed. The man was again assessed as
presenting a medium risk to the public and a medium risk of escape.
The reason for the use of restraints was given as “cat B prisoner serving
life for murder”, and the reason for the use of the escort chain as “due to
illness and infirmity”.
21. At 8.45am on 11 October, the man was seen by the surgeon and told he
would be having a scan. The escort staff obtained permission from the
Orderly Officer at Gartree to remove the escort chain during the scan,
which took place at 10.25am. At 11.35am, the chain was again removed
briefly with the permission of the Duty Governor to allow the man to have
blood tests. At 1.35pm, the man received permission to call his wife.
The record shows he told her which ward he was on. At 2.30pm, he was
visited by a worker from Age Concern and also saw the surgeon. The
worker from Age Concern called the hospital at 3.30pm to confirm that
the man’s wife would visit at about 5.00pm.
22. At 4.45pm on 11 October, the bedwatch log shows that the surgeon told
the man that he had suffered a “twisted bowel” and that there was a
“very high” risk that he would die if he had surgery to correct it. The
record shows that the man agreed to talk to his wife about the situation
and would then make a decision about whether to have an operation or
not.
23. The man’s wife arrived at 5.15pm. The man is recorded as becoming
agitated after being told the implications of surgery and at 6.20pm the
escort chain was removed when he pulled out the tubes he was
attached to. He then appears to have had a heart scan and the chain
was replaced some 15 minutes later.
24. The man’s wife told my investigator and my family liaison officer (FLO)
that her husband was struggling to breathe when she visited and was
wearing an oxygen mask. She said he found it difficult to talk and had
“lots of tubes in him”. She said she thought hospital staff thought that he
would not survive an operation. She said her husband had wanted to go
ahead with surgery and she had been asked about consent.
25. At 7.00pm, the surgeon told the man that he intended to insert a wire “to
control heart rate before attempting any further operation”. At 8.00pm,
the man was again seen by the surgeon who explained to him all the
risks of undergoing an operation. The man signed the necessary
consent form. Throughout the rest of the evening he underwent further
tests and was given morphine to ease the pain he was in.
26. At 12.45am on 12 October, the surgeon told the man who died that he
was too unwell to undergo surgery. The GP who conducted the clinical
review told my investigator that the hospital records showed that,
although the surgeon was willing to operate on the man, the anaesthetist
refused on the grounds that the man’s heart condition meant he would
not survive the operation.
27. At 4.00am, an entry shows that the man’s wife was informed that her
husband was very unwell. It is not clear whether it was prison or
hospital staff who made this call. The man’s wife told my investigator
that she was called in “the middle of the night” but that she was very
drowsy as she had taken a sleeping tablet. The man’s wife speaks
relatively little English and she said that she did not understand that her
husband was dying. She said she thought that the hospital told her to
come in the next morning.
28. The hospital bedwatch log shows that the man who died was seen by a
doctor at 8.24am who decided he should be moved to a side room. At
9.00am, another doctor asked that the escort chain be removed and
permission to do so was given by the Duty Governor. The log shows
that the prison was trying to contact the man’s wife. The hospital
chaplain came to the man’s bedside at 9.31am. At 9.37am, the man
was pronounced dead.
WHAT OTHER PRISONERS SAID
29. My investigator spoke to a number of the man’s fellow prisoners. In
interview, one prisoner said he had been on G wing “near enough since
it opened” and had known the man who died since then. This prisoner is
a wheelchair user and occupies the other specially adapted cell on G
wing next door to the cell in which the man who died was located. The
prisoner said that he spoke to the man who died on Thursday 5 October
and he “was not quite himself”. On 6 October, he remembered the man
complaining of stomach pains. During the night of 6 or 7 October, he
remembered hearing the man groaning loudly and coughing. He said he
pressed his cell bell and asked the night officer to check on him. The
night officer reassured him that healthcare staff knew the man was ill
and he was under the impression that a doctor might have visited the
man who died during the night. The prisoner who I spoke with said he
saw the man who died on Sunday morning and noticed his stomach was
very swollen. He said the man was in so much pain he could not speak.
30. The prisoner who my investigator spoke with said that he knew that
healthcare staff thought that the man who died had food poisoning. By
Monday 9 October, the man’s stomach was “like a whale” and several
prisoners were asking staff to do something to help him. He said the
man looked very unwell before he went to hospital on 10 October.
Prisoners were angry when they heard that he had died because they
thought something should have been done for him sooner.
31. A second prisoner who my investigator spoke with said he had known
the man who died for 16 months since G wing opened. He said the man
had been very ill for five days before he died and had not come out for
his meals which was very unusual. He said the man had complained
about pains in his stomach and he had seen him holding his lower
stomach.
32. A third prisoner who my investigator spoke with said the man had
complained of feeling unwell about four days before he went to hospital
and had looked “really unwell”. This prisoner said that the healthcare
department were good at sending staff out to the wings to see prisoners
who were ill, but these staff were usually triage nurses and he thought
that it would be more appropriate for more senior medical staff to be
sent. He described his own experience of the triage system which
involved seeing a nurse four times before being sent to see the doctor.
He said the doctor diagnosed irritable bowel syndrome (IBS) but his
symptoms persisted. He saw a triage nurse a further two times before
he was referred to the doctor again. The doctor then referred him to
outside hospital where he was diagnosed with diverticulitis. This
prisoner added that, while he was in outside hospital, he had undergone
an endoscopy while chained to a member of staff by an escort chain.
He said that the doctor had asked the escort staff to remove the chain
because he was to be sedated but staff had refused.
33. A fourth prisoner who my investigator spoke with said he had known the
man who died since G wing opened in July 2005. He said that “about a
week” before the man was taken to hospital he had taken some library
books to him. The man had complained that he was not feeling very
well. Over the following three or four days, the man did not come out of
his cell and staff told the fourth prisoner that he was “resting”. This
prisoner said he heard the man moaning frequently because he had the
cell directly below him.
34. A fifth prisoner who my investigator spoke with said he had known the
man who died for about a year and had helped him clean his cell and
change his bed. This prisoner said the man had been ill for five days
before he was taken to hospital. He said he saw a nurse visit him once
or twice during the five days, but thought that once it was “by chance”
because the nurse had come to see another patient. This prisoner said
that the man looked very ill and his face had changed completely. He
said he was eating little and his stomach had become very swollen.
Staff told him the man had food poisoning.
35. Three other prisoners who were interviewed told my investigator that the
man who died had been ill for about five days before he was taken to
hospital.
ISSUES CONSIDERED DURING THE INVESTIGATION
Medical record keeping at Gartree
36. In his clinical review, the author said the quality of the man’s medical
records was poor. He said there was no logical order to the filing
system, and he found it difficult to find any relevant information regarding
a particular condition without reviewing all the entries. Although the
majority of clinical entries were appropriately dated, they were not always
legible. This problem was compounded by the fact that staff signed the
entries but did not also print their names. The author also commented
on the large amount of non-clinical information filed in the man’s medical
record. He found that this made it even more difficult to follow the
threads of the man’s medical treatment. The problem of illegible
signatures, and the difficulty of following the chronology of events in the
medical record, are also raised by the clinical review in my investigation
into Gartree’s previous death from natural causes. As that review
pointed out, “accurate, informative, contemporaneous and legible records
are essential to support communication between staff and improve
patient care.”
37. The author also commented that the man who died did not appear to
have undergone a healthscreen on arrival at Gartree. The man did arrive
with a detailed summary from the medical officer at his previous prison
(Whitemoor), but this does not seem to have been collated or referred to
when the man experienced further ill health at Gartree. Although an
elderly man with a long history of poor health, his GP records were not
requested by Gartree, or by either of his previous prisons. The clinical
reviewer concluded that staff at Gartree failed to appreciate that the man
suffered from a chronic heart condition. For example, on four separate
occasions between July 2002 and January 2004, episodes of shortness
of breath were treated as asthma despite the summary provided by
Whitemoor clearly stating a history of chronic obstructive airways disease
and chronic heart failure. The clinical reviewer has indicated these were
a more likely cause of the man’s symptoms.
38. The clinical reviewer concluded that the poor record keeping at Gartree,
and a lack of patient summaries from previous prisons and the prisoner’s
GP, undermines the ability of healthcare staff to adopt a co-ordinated
approach to the treatment of chronic medical conditions. In this case, the
man arrived at Gartree with chronic medical conditions of some years
standing. As he was at that stage relatively new to the prison system, it
would have been wise to have requested a summary from his GP.
The treatment of the man’s chronic medical conditions
39. As noted above, the clinical reviewer concluded that healthcare staff at
Gartree did not appear to be aware that the man suffered from chronic
heart disease. The author also judged that, during the man’s time in
Gartree, the management of his diabetes and chronic obstructive airways
disease fell short of the National Institute of Clinical Evidence (NICE)
guidelines of 2002 and 2004 respectively. The man who died does not
appear to have had the required blood tests to check his diabetes control
and cholesterol levels, or had the required retinal screening to monitor
eye damage caused by diabetes. It is possible that some of the
treatment and interventions for which the man should have been
considered were in fact considered, but it is not possible to establish this
from his medical record.
40. However, the clinical reviewer noted that, until 2005, GP provision at
Gartree was provided on a sessional basis and the visiting doctors did
not hold responsibility for chronic disease management. The author
believes that nursing staff at Gartree have shown great dedication in the
last few years to introducing a number of chronic disease registers and
clinics, but they have been hampered by insufficient clinical input and
leadership. The author concluded that the situation has improved since
the tendering process for medical services in 2005. He has suggested
that Gartree develop an annual audit system to demonstrate equity of
service with a GP practice.
41. The clinical reviewer concluded that, for a variety of reasons not all
directly under the control of the nursing team at Gartree, the treatment of
the man’s chronic medical conditions in prison fell below that he could
reasonably have expected to have received in the community.
The treatment of the man’s recent medical conditions
42. The clinical reviewer concluded that the details of the assessment and
treatment of the man’s recent medical conditions – those which appeared
after he was transferred to Gartree – were appropriate. The man’s
referrals to clinics were timely and apposite and, for the new complaints,
his medical treatment was comparable to that which he would have
received in the community.
The treatment of the man’s final illness
43. The clinical reviewer raised several concerns about the triage policy in
operation at Gartree. This policy was introduced in 2005 with the aims of
making sure that prisoners who are ill are sent to the appropriate
practitioner, and ensuring a more effective use of scarce healthcare
resources. I agree with the author that the founding principles behind
this policy are laudable, and I commend what he found to be “the
enthusiasm of the staff to react to the unique pressures of the Prison
Service”. I note also that one of the prisoners interviewed by my
investigator confirmed that staff are quick to attend sick prisoners on the
houseblocks. However, the author concluded that:
“The system failed in the case of the man who died as the
presumed diagnosis of food poisoning, which was likely from the
history of unsafe food storage, was not balanced against the
many possible causes of abdominal pain and vomiting. The staff
making assessments did not have sufficient clinical knowledge of
alternative diagnoses, nor understand the implications of the
man’s chronic medical conditions, nor hold the clinical skill to
make a full examination to support the diagnosis.”
The clinical reviewer judged that there was some doubt that the man who
died had had episodes of diarrhoea which made a diagnosis of food
poisoning less likely. Neither member of staff who examined the man
between 6 and 9 October performed an abdominal examination which
was required to exclude surgical causes of abdominal pain and vomiting.
The man did not have a chest examination. Moreover, given his history
of chronic heart failure, poor oral intake would have affected the
absorption of his daily medicines.
44. I am very concerned by the conclusion of the clinical reviewer, namely
that neither of the assessing healthcare staff had the medical knowledge
or the clinical skills to make a proper diagnosis in the case of the man
who died. Although the Gartree triage policy itself correctly identifies the
training and experience necessary for staff to undertake triage, these
requirements have not been met. In this case, the healthcare officers
who assessed the man were qualified mental health nurses and have not
had the experience or training in minor ailments and prescribing required
by Gartree’s own policy document.
45. The admitting consultant at the local hospital told the clinical reviewer
that, once the perforation of the man’s bowel had occurred, the outcome
for him would have been grave due to his underlying medical conditions.
It is therefore impossible to tell whether the delay in taking the man to
hospital contributed to his death. It is also the case that the man’s final
diagnosis was obscure and not readily apparent to hospital staff.
Although this in part excuses the delay in taking him to hospital, the
underlying concerns about the inadequacy of the triage system remain.
Unless these are speedily addressed, I fear the outcome in other cases
may indeed prove fatal. I judged this issue to be of sufficient concern
that I raised it with the Governor of Gartree prior to the publication of this
report in draft.
46. The clinical reviewer made seven recommendations (listed in the
recommendations section below) which address the issues of patient
summaries, the management of chronic diseases, staff training, the
triage policy, the out of hours service and the unacceptable drain on
healthcare resources caused by 12 out of the 14 in-patient beds in the
HCC being used as a solution to overcrowding on the houseblocks.
I endorse the recommendations of the clinical reviewer and
recommend that the Governor of Gartree ensures that a
comprehensive review of his report takes place within three
months of the publication of this PPO report.
I recommend that, within this review, consideration of the
triage policy should take place as a matter of urgency and
measures put in place to ensure that staff who are required
to operate it have up to date and appropriate training in
identifying minor ailments and prescribing.
I further recommend that a priority should be to review
medical record keeping and put measures in place to ensure
that records are ordered logically, contain only clinical
information and that all entries are signed and names
printed alongside. Particular attention should be given to
identifying prisoners with chronic medical conditions
diagnosed prior to transfer to Gartree.
The use of the escort chain while the man who died was in Glenfield
Hospital
47. The Prison Service National Security Framework (NSF) replaced the
Prison Security Manual in 2004. It is available on the Prison Service
intranet and gives guidance on the procedures for escorting prisoners on
all occasions they leave prison and on the use of restraints. There is
particular advice on hospital escorts and bedwatches, and when
restraints should be applied or removed during medical treatment.
48. The section of the NSF relating to hospital escorts says that the prison
must first undertake a risk assessment to decide the level of escort and
restraint required for the safe custody of each prisoner. Factors which
must be taken into account during this risk assessment include:
(cid:127) the prisoner’s medical condition
(cid:127) the prisoner’s security category
(cid:127) the nature of their offence
(cid:127) their risk to the public and hospital staff
(cid:127) their motivation to escape
49. According to the NSF, the normal arrangements for prisoners being
escorted from closed establishments are that they will be accompanied
by two officers and “restraints must be used unless there are medical
objections”. The NSF goes on to say:
“Restraints are applied when out of the prison up to the point of
medical consultation or treatment. The restraints will be taken
off at this point unless the risk assessment shows the risk of
escape is too high.”
50. Although this is the normal arrangement, the section on ‘Escort options’
which follows it says that other options are available to prison managers.
These are:
“an escort with two officers or more with no restraints
“an escort with one officer and no restraints (appropriate where
the prisoner’s medical condition or lack of mobility is such that he
or she cannot escape unaided and there is no evidence that an
escape attempt is likely)”
51. The section on ‘Reviewing the escort arrangements’ says:
“The level of security necessary in all cases must be kept under
review to take into account the prisoner’s developing medical
condition, the physical surroundings in which the prisoner is
located and any emerging intelligence.”
52. The NSF section on ‘Restraints’ lists the circumstances in which
handcuffs are not usually necessary. One of these circumstances is:
“On prisoners attending for medical treatment outside the prison,
if the prisoner’s medical condition renders restraints
inappropriate or a risk assessment demonstrates they are
unnecessary in all the circumstances. Restraints will not
normally be necessary when the prisoner’s mobility is severely
limited, e.g. when he or she is on crutches … “
53. The NSF section on ‘Removal of restraints’ says that restraints may be
removed in certain circumstances during hospital treatment. It does not
list these circumstances but refers back to the section on hospital
escorts. It goes on to say that restraints may be removed, “when a
medical professional requests their removal on health grounds.” If
necessary, escorting officers should first obtain the permission of the
Duty Governor before agreeing to such requests.
54. Gartree’s local policy on hospital escorts and bedwatches is
comparatively stark and unhelpful. There is no mention that restraints
may not be appropriate in certain circumstances, particularly when a
prisoner’s medical condition and lack of mobility is such that they are
unable to escape unaided. There is no suggestion that reviews of the
level of restraint should take into account the prisoner’s developing
medical condition. The sole reference to circumstances in which
restraints may be removed is:
“Preservation of life [bold in original]. Remember that if a
prisoner is restrained by either handcuffs or an escort chain and
the medical staff need to have them removed to administer
emergency/life saving treatment, they must be removed.
This is in compliance with the security manual.”
55. When the man who died left Gartree on 10 October, it was not known
how ill he was or whether he would be admitted to hospital. The level of
escort was set at two members of staff and the use of the escort chain
was specified instead of handcuffs in view of his poor health and mobility
problems. I am satisfied that this was a reasonable decision in the
circumstances. This level of escort and restraint remained in place until
the man was seen by the doctor at 4.00pm and staff were told that he
would be admitted for “at least 24 hours”. Oscar 1 (the code for the
Orderly Officer, the officer in charge of the prison at that time) was
informed at 4.25pm. There is no documentary evidence that a further
review of the level of escort and restraint was done at this time, but I
presume that a decision was taken not to change it.
56. The situation was reviewed by the Governor on the morning of 11
October. By this time, the man who died had a catheter and a nasal tube
inserted. The Governor decided that the level of escort should remain at
two members of staff and that the man should continue to be chained to
one of them using the escort chain. The reason given on the risk
assessment form was that he was a category B prisoner serving life for
murder. At 4.45pm, the man was informed by the surgeon that there was
a high probability that he would die during the operation he needed. He
was asked to discuss with his wife whether he was prepared to take the
risk of surgery. When the man’s wife visited some 30 minutes later, she
said her husband was using an oxygen mask and could hardly speak
because his breathing was so laboured. At 12.45am, the man was given
the news that his potentially life saving operation would not be taking
place because he was too ill to survive it. He remained cuffed to staff by
the escort chain until a doctor asked staff to remove it some 35 minutes
before he died later the same morning.
57. It seems clear from the bedwatch log and from what the man’s wife said
that the man was gravely ill throughout his brief stay in the local hospital.
I understand the importance of security on bedwatches and realise that,
for often valid reasons, the current climate in the Prison Service is risk
averse. However, the man who died was both chronically ill and 80
years old. He was at best unsteady on his feet and at worst unable to
move without aid of a wheelchair. Within hours of his admittance to
hospital, he was fitted with a catheter and a nasal tube and needed an
oxygen mask to breathe. I do not believe that the man required to be
chained to an escorting officer, and quite probably did not need more
than one member of staff in attendance. The man’s wife told my staff
that the escorting officers were obviously uncomfortable that her
husband remained chained when he was so ill. It is a situation which is
undignified for the prisoner and distressing for his family and the
escorting officers.
58. I do not consider that Gartree’s policy reflects the National Security
Framework as required. It is outdated and appears to be based on the
contents of the Prison Security Manual which became obsolete in 2004.
The NSF (and indeed its predecessor, the Prison Security Manual)
makes it clear that prisoners should not normally be restrained during
medical treatment. I make no criticism of the individual staff involved in
the case of the man who died. All staff complied with Gartree’s local
policies. However, the local policy needs urgent review to more fully
reflect the requirements of the NSF.
I recommend that, within three months of the publication of
this report, the Governor of Gartree ensures that a review of
local escorting procedures and the local instructions for
staff on a bedwatch takes place to more fully reflect the
standards set by the National Security Framework.
Specifically, the review should seek to expand the policy to
encompass the circumstances in which restraints may not
be appropriate and give the Duty Governor discretion to
remove restraints from gravely ill or dying prisoners without
waiting for a request from medical staff.
I further recommend that the Governor of Gartree reminds
escorting staff that restraints should normally be removed
from all prisoners during medical consultation and
treatment.
59. In a recent report of an investigation into a death of a prisoner from HMP
Birmingham, I was critical of the lack of flexibility in the local policy on
bedwatches. I made a recommendation that the Governor amend the
policy to allow the duty governor discretion to authorise the removal of
handcuffs in non-life threatening situations. A lack of flexibility in the
local policy is again a feature of this case but I think the problem is
deeper. Prison managers must base their local instructions on the
standards set by the National Security Framework. The NSF does not
deal specifically with procedures for dealing with gravely ill or dying
prisoners in outside hospital or hospices. There is no allowance for
prison staff to make a decision to remove the restraints from gravely ill or
dying prisoners on compassionate grounds when the risk of escape is
clearly much reduced. I know of cases where Governors have
nevertheless given permission for restraints to be removed on
compassionate grounds and I have always commended them for doing
so. However, I now think consideration should be given to building this
into the NSF so the practice may become more widespread.
I recommend that, within three months of the publication of
this report, the Director General of the Prison Service
ensures that a review of the escorts, restraints and
bedwatch sections of the National Security Framework
takes place, with a view to providing advice to staff on the
removal of restraints from gravely ill or dying prisoners on
compassionate grounds when their risk of escape is much
reduced.
60. Finally, I would like to add that I am pleased to see that the family of the
man who died was appreciative of the help they received from the HMP
Gartree after his death. From what I have learned, this was an example
of good practice and should be recognised as such. Although I make no
formal recommendation in this regard, the Governor and her Area
Manager will wish to ensure that my comments are brought to the
attention of those concerned.
RECOMMENDATIONS
Recommendations from the clinical review
1. A medical record summary should be obtained from the general
practitioner of all new inmates in all instances. Information from
previous prisons should be collated and summarised. There
should be a formal review of all diagnoses, investigations and
therapy, particularly as the inmates are likely to be drawn from a
group of the population that has not fully accessed the health
service. There should be a designated role for filing and upkeep
of medical records, including a protocol for summarising
important information regarding diagnosis and results.
2. The prison should identify funding for note summarising of all
current inmates’ records. The first phase should be for inmates
on medications suggestive of significant chronic diseases and
should be completed as a paper exercise prior to the installation
of the prison Health IT system. These prisoners should have a
summary sheet of all significant past medical conditions
recorded in a prominent place in the medical record. This
should be done as a matter of urgency. The prison should
engage with the medical officers to determine the mechanism
and priority of the summarising process. It is my opinion that
the majority of prisoners without significant chronic disease
issues should have their records summarised at the time of the
introduction of the prison Health IT system to prevent
unnecessary duplication of labour, providing there is no undue
delay.
3. The same audit standards sought in the Quality Outcomes
Framework (QOF) in general practice for chronic disease
management should be applied to the health provision for
prisoners at HMP Gartree. This would be a future marker of the
quality of service provision. It is assumed that the new Health IT
system will be able to deliver this recommendation; failing this
the prison should make provision for audit clerks to perform the
task manually. It is anticipated that there would be a delay
caused by moving from paper to computer records before this is
achieved.
4. A staff training needs analysis should be performed and
matched to the needs of the service. Staff should be released to
attend recognised training courses relating to chronic disease
management and minor illness, then to consolidate knowledge
in the general practice setting with sufficient backfill. Prison
staff should be discouraged from working beyond their ability
until they have received training and supervision comparable to
their colleagues working in similar circumstances in primary and
unscheduled care. The unique pressures of the prison should
not be an excuse for staff to work beyond their scope of
practice. Doctors at the local practice would be well placed to
judge the competency of the nursing staff in these roles after
training, although this role falls outside their medical contract.
5. The triage policy needs urgent review with greater clinical input.
Only staff demonstrating appropriate clinical experience and
training should perform unsupervised triage. Staff should reflect
on their roles to ensure they are not working beyond their scope
of practice and highlight their learning needs.
There are software programs to support nurse triage, training
sessions and audit tools to measure performance, in established
use for the PCT out of hours service. The prison healthcare and
PCT should explore the application of these systems in HMP
Gartree.
6. Closer liaison with the OOH service should be encouraged. Staff
should be aware that they can get telephone advice from the
triage centre. It may be useful to create a brief orientation pack
for visiting doctors explaining the prison setting, to be carried in
the visiting doctor’s car.
7. Hospital wing inpatient facility staffing should be reviewed.
Nurses should only be involved in the health needs of inmates in
the two “medical” beds. This will allow nursing staff roles to
expand into chronic disease management and allow for backfill
for training. The healthcare staff should as far as possible be
disengaged from the custodial function of the prison, in order to
strive to provide independent holistic care. It would be useful to
review the language used in the prison healthcare system such
as “special sick parade” in order to refocus on the role of
delivering modern efficient primary care to individuals.
Recommendations from the PPO
1. I endorse the recommendations of the author of the clinical
review and recommend that the Governor of Gartree ensures
that a comprehensive review of his report takes place within
three months of the publication of this PPO report.
2. I recommend that, within this review, consideration of the
triage policy should take place as a matter of urgency and
measures put in place to ensure that staff who are required to
operate it have up to date and appropriate training in
identifying minor ailments and prescribing.
3. I further recommend that a priority should be to review medical
record keeping and put measures in place to ensure that
records are ordered logically, contain only clinical information
and that all entries are signed and names printed alongside.
Particular attention should be given to identifying prisoners
with chronic medical conditions diagnosed prior to transfer to
Gartree.
4. I recommend that, within three months of the publication of this
report, the Governor of Gartree ensures that a review of local
escorting procedures and the local instructions for staff on a
bedwatch takes place to bring them in line with the standards
set by the National Security Framework. Specifically, the
review should seek to expand the policy to encompass the
circumstances in which restraints may not be appropriate and
give the Duty Governor discretion to remove restraints from
gravely ill or dying prisoners without waiting for a request from
medical staff.
5. I further recommend that the Governor of Gartree reminds
escorting staff that restraints should normally be removed from
prisoners during medical consultation and treatment.
6. I recommend that, within three months of the publication of this
report, the Director General of the Prison Service ensures that
a review of the escorts, restraints and bedwatch sections of the
National Security Framework takes place, with a view to
providing advice to staff on the removal of restraints from
gravely ill or dying prisoners on compassionate grounds when
their risk of escape is much reduced.