6.1 The circumstances surrounding any death which which has taken place in prison are subject to investigation by an independent Fatal Accident Inquiry (FAI), which also makes recommendations where appropriate. However, as all our inspections include an assessment of the safety of individuals in custody, our observations below are produced with the aim of improving public awareness and of assisting prison management in its onerous task of trying to achieve a balance between conflicting priorities in this area.

6.2 There were seventeen suicides* in Scottish prisons between 1 April 1996 and 31 March 1997, eleven of which involved prisoners who were on remand. Included in this total were three young women, which sadly is a record.

6.3 Comparative statistics for self-inflicted deaths over the last five years are as follows:

Click to view table (25KB)

6.5 These figures appear to reflect the increase in young male suicides which has been experienced in the wider community in recent years. However, the prison rate, which is rising at about 6% per annum, appears to be growing at a greater pace. At a time when the prison population is expanding, this could be due to prisoners tending to come from poorer socio-economic backgrounds or from areas where drug abuse is rife. More detailed observations are shown below, though it should be noted that these are based entirely on self inflicted deaths; the issue of attempted suicides is even more complex and does not always present a similar picture.

Open Establishments and Category ‘C’ Prisons

6.5.1 It would appear that the risk of fatalities in these establishments is much lower.

6.5.2 There have been no fatalities at HMYOIs Dumfries, Glenochil and Polmont for four years, despite the fact that young men in the community are at the greatest risk. Whilst we are unable to provide compelling reasons for this, our previous reports have referred to the coordinated approach which is adopted towards vulnerable individuals at Polmont, which includes the opportunity for much closer integration in its hospital ward.

6.5.3 Until April 1997, there had been no fatal suicides amongst long term prisoners at Peterhead, Shotts or Glenochil since 1989.

6.5.4 More recent figures for deaths among remand prisoners - 12 out of 17 in 1996-97 - indicate a high rate for this group, who account for approximately one fifth of the daily prisoner population and more than half the annual throughput. Many tragedies have taken place in the early stages of custody, apparently with most having given no indication to prison staff of their intentions, though some may have had a history of previous attempts.

6.5.5 It is perhaps not unexpected that so many remands are involved, given that their particular situation includes the added anxiety of not always knowing how long the period of custody will last or the outcome of any subsequent trial. Additionally, the conditions experienced by remand prisoners are some of the worst within the prison system, partly due to the pressures created by overcrowding. They are also locked up in their cells for long periods at most establishments as under Prison Rules, they are not required to work.

6.5.6 It is equally possible that there could be a connection between the high throughput rates at remand prisons and the number of fatalities, as staff do not always get the chance to establish proper supervision or close relationships. Drug withdrawal symptoms can lead to both physical and psychological problems whilst depression may also be behind some deaths; for example, there is good evidence that withdrawal from methadone is severe and unpleasant in the first few days after any prescription is withdrawn and that this condition can last for several weeks. It is clear, therefore, that it is important to ensure that all those withdrawing from drugs receive appropriate treatment.

6.5.7 However, the attribution of any one factor - either amongst remand or convicted prisoners - is difficult to make, though in our recent report on HMP Edinburgh it was noted that there had not been a suicide for three years despite the very high throughput of remands. The following possibilities were highlighted - the comprehensive anti-drugs strategy (which included the prescription of methadone and drug counselling for remands), the integrated hospital ward, excellent psychiatric care (which was also linked to the community) and the Listeners Scheme which was being operated by prisoners. Elsewhere, we found that where one or more of these aspects was missing or incomplete, a higher incidence of suicide was more likely - see inspection reports for HMPs Greenock and Perth. At HMP Barlinnie, the exceptionally high throughput of remand prisoners is probably a key factor in its death rate. (A formal inspection of that establishment is scheduled for May 1997.)

6.5.8 The cluster of deaths amongst the very small population of women at Cornton Vale is at odds with any comparable statistics and a joint FAI for three of the deaths was ongoing at the time this report was being prepared. Without prejudice to its outcome, we offer the following observations; many of its prisoners come from the Greater Glasgow area and a number of local factors may be involved, including the high proportion of poly drug abusers amongst admissions, together with individuals with psychiatric problems, or elements of both. Some incidents may have been ‘cries for help’ which then uninentionally became fatal, whilst others could have been some form of copy-cat behaviour caused in part by the publicity surrounding these tragedies.

6.7 The majority of suicides have involved hanging and there have been expensive attempts to remove bars from the inside by the installation of anti-ligature windows at many prisons, including recently at Cornton Vale. However, it is unrealistic completely to remove the many means by which prisoners can kill themselves, in the same way as it would not be possible totally to eliminate suicides in prison, especially at a time when the population is reaching such record levels; nevertheless we support any reasonable strategy which aims to reduce the risk of death and self injury amonst prisoners. The SPS is, therefore, to be commended for its current re-emphasis on integration, coordination and care of prisoners, an approach which we believe is eminently pragmatic.

6.8 Concern about the growing number of suicides has led to the commissioning of two major studies by the SPS. Conducted by Professor Gunn and Dr Power, these have involved the examination of cases occurring over the last 20 years and their general conclusions concentrated on the following areas:-

• Methods to improve the identification of those at risk
• Communication and staff training
• Better care arrangements.

6.10 The SPS constantly tries to balance conflicting priorities, for example in choosing between improved security, availability of quality regimes, improving anti-drugs strategies or improving suicide prevention measures. Following receipt of the Gunn and Power reports, an SPS Steering Group is now directing on policy adjustments which will compliment its suicide prevention strategy, which has been in place since 1992. Drafts have been seen at Governor level and where appropriate, some changes have already been introduced. Implementation at all prison establishments is likely to follow, once resource implications have been completely resolved.

6.11 We have also made a number of separate recommendations specifically relating to remand prisoners at Cornton Vale (see Chapter 11), with the undermentioned possibly having some wider application at other prisons:

• More comprehensive and special induction arrangements should be made for all remand prisoners
• A properly trained Personal Officer should be allocated to every remand prisoner.
• Drug addictions work should apply to remand, as well as convicted prisoners.
• More therapeutic regimes should be provided for remands, especially around the week-end (when the majority of deaths occur).
• There should be greater access for remands to telephones - and therefore contact with their families and the community.
• Further consideration should be given to providing in-cell TV for remand prisoners. (This might combat boredom and therefore act as an anti-suicide measure.)

In addition, our 1996 Thematic Study report recommended that visit entitlements for remand prisoners should be doubled.

6.12 In general, we strongly suggest that priority is given to preventive measures which concern remand prisoners, as they would appear to be most at risk of death on a daily basis. The need for greater focus on their induction arrangements - which should be treated as a process and not just as a one-off event - needs highlighting. Special care should also be taken where there are any drug withdrawal complications, particularly during the very early periods of custody or where self harm has previously been indicated; this should of course also apply to convicted prisoners.

6.13 However, if it is accepted that imprisonment can add to the suicidal tendencies already suffered by some individuals, then the greatest priority should be directed towards imprisoning fewer people in the first place. This policy should apply in the first instance to remands, with many more bail and community placement hostels being made available - see also paragraph 4.20. It has been encouraging to note the recent lead that The Scottish Office has taken in establishing more bail beds for women in Greater Glasgow; there has also been a considerable reduction in the number of women being sent on remand to Cornton Vale which will complement the anti-suicide measures which are being vigorously implemented at this Institution by the SPS. There are also some prisoners who are suffering from mental disorder, who ought not to be in prison custody.

6.14 Finally, the effect on prison staff must be considered. Those involved in the discovery of suicides or in resuscitation attempts, can be severely traumatized. Proper, professional counselling arrangements are therefore fundamental and should not be overlooked - either at the time or later - following any violent death in custody.