contribute to NOMS' review of the compliance and delivery of the ACCT process. ..... between healthcare and discipline s
Submission to National Offender Management Service (NOMS) Review of compliance and delivery of the Assessment, Care in Custody and Teamwork (ACCT) process (August 2015) Introduction In his last annual report, and in various other publications, the Prisons and Probation Ombudsman (PPO) called for a review of ACCT arrangements. While the review is not as comprehensive as we might have liked, we welcome the opportunity to contribute to NOMS’ review of the compliance and delivery of the ACCT process. Our investigations have frequently identified weaknesses in the implementation of ACCT and this has been emphasised by contact with prison staff. These staff have highlighted the need for ACCT to be updated and streamlined to ensure it can be effectively delivered in a Prison Service with many more prisoners and fewer staff than when ACCT arrangements were introduced. While we have made clear that, in comparison to most other jurisdictions, we consider that the ACCT process as a whole is an essentially sound set of procedures, there is scope for some changes to the design of the document itself and, more particularly, much work is needed to improve implementation. Further, a review of ACCT will only be successful if there is a cultural and resource commitment to the implementation of the review’s recommendations. Finally, it is important that the review involves revising PSI 64/2011 Management of prisoners at risk of harm to self, to others and from others (Safer Custody). Our investigations have found the PSI is an unwieldy, inconsistent and confusing document which should be re-written in plain English. Position We recognise that the PPO has a very particular experience of the ACCT process: in all of the cases we investigate where the prisoner has been on an ACCT, the ACCT process has failed to prevent a self-inflicted death. However, there is little reason to consider that the ACCTs we see are not representative and investigators’ examination of other ACCT documents in the course of investigations often identifies
1
similar issues. Sometimes the inadequate knowledge and training of staff leads to the poor supervision of the ACCT process. There is often a lack of awareness about risk, an absence of multi-disciplinary working, or inappropriate use of segregation, which heightens the vulnerability of already vulnerable prisoners. The PPO sees too many cases where the ACCT procedure is not followed as thoroughly as it should have been, or where case reviews are not carried out within specified timescales or key information is not recorded. The ACCT document should be a live plan which is reviewed and updated when a prisoner’s circumstances or risk change, but often this does not happen. A 2014 Learning Lesson’s thematic report, Self-inflicted deaths of prisoners on ACCT, showed that the ACCT process was not correctly implemented or monitored in half the deaths considered. A well-written ACCT plan alone will not prevent a person from self-harming or killing themselves. The plan has to be implemented in a rigorous and timely way to have the best chance of helping the prisoner address the underlying root causes of their distress. This is best achieved by taking a holistic approach to the prisoner’s care, and focusing on their individual needs and circumstances. Continuity of support and care for prisoners at risk of suicide and self-harm is key to ensuring a successful approach to managing their immediate and future risk. At a time of significant change and austerity in the Prison Service, this may become increasingly challenging. Yet prisons are under a duty of care to ensure that they do everything they can to keep a prisoner safe, in particular complying with the ACCT procedures, and providing continuity of care for those considered at risk. Context There were 76 self-inflicted deaths across prisons, approved premises and immigration removal centres in 2014/15. Although this was, thankfully, lower than the previous year (2013/14) when there had been 90 self-inflicted deaths, it was still 36% higher than the year before that (2012/13) when there were 55. In the first quarter of 2015/16 there have been three more self-inflicted deaths than the equivalent period last year. Our thematic review of self-inflicted deaths of prisoners in 2013/14, found 29% of the prisoners were being managed under ACCT procedures when they died.
2
Our response broadly follows the seven areas set out by NOMS for consideration. We held discussions with investigators into self-inflicted deaths who had some additional comments and suggestions about the ACCT process. Our response should be read in conjunction with the Learning Lessons publications we have previously shared with the review team and the key points from which can be found at the end of this document. 1.
The structure of the ACCT document
As part of our research to inform this paper, we held a discussion workshop with senior investigators to draw out their experiences of how prisons use the ACCT process. Staff highlighted, among other issues, some of the limitations of the ACCT document itself. These limitations, in language, structure and action, inhibited the ability of staff to implement the ACCT process properly. Investigators thought that it contributed to a culture of “action without critical thinking”, a tick box exercise where staff simply followed instructions rather than considering why a specific action was required and what the consequence of the action might be. Our investigators proposed redesigning aspects of the ACCT document to encourage staff to have ownership of their actions, and increased personal responsibility in their management of the ACCT process. This would lead to the better understanding of the ACCT procedures and better protection of vulnerable prisoners. A further concern is that each part of the ACCT process appears to operate in isolation from each other and there is often little evidence of a joined up, case management approach to allow more effective care – in the way that the policy underpinning ACCT envisages. We find that assessment interviews are often well conducted but there is little evidence that they are referred to subsequently – even at first case reviews, which the assessor rarely attends. Case reviews appear to operate in isolation from each other: it is rare to find a single consistent case manager and new staff chairing case reviews do not appear to review what happened before, including checking and updating CAREMAP actions, checking what the original risk factors were and reviewing the ongoing ACCT record or information from the prisoner’s medical record to identify anything significant that has happened since the last review. Most reviews rely too much on what the prisoner says – and prisoners intent on suicide rarely announce this at ACCT reviews. There needs to be 3
wider consideration of all the information available about the prisoner’s risks. We are particularly concerned at the number of reviews we see, including when ACCTs are closed, when only one member of staff is present. Guidance on ACCT needs to make it clear that this should never happen. More detailed comments are set out below. A number of aspects of the ACCT document would benefit from revision: moving the list of risk factors; better understanding of the triggers section; use of the Suicide/Self-harm risk guidance; use of the assessment interview; the need to include healthcare staff in the process; effectiveness of case reviews; rethinking CAREMAPs; and the use of active language. (i)
We often find that the section on the front cover of the ACCT document, which should specify the level of observations, is difficult to read and not completed in a readily understandable way. There is no space for the date, when the observations are changed, to be included. Staff often say that they find this section difficult to follow and there is too much scope for staff to misinterpret what is required.
(ii)
It would be helpful to move the list of triggers to the front of the ACCT document so that every member of staff administering the ACCT is forced to look at them and consider them. We rarely find any reference to the assessor’s summary of findings.
(iii)
The triggers section is often, in our experience, misunderstood, rarely fully completed, and often forgotten after the first entry. Entries need to be more specific and detailed. For example, where a court appearance is cited as a trigger, the date of the court appearance, once it is known, should be added to the document. The triggers should also be reviewed regularly. All staff engaging in the process need to be aware of what a prisoner’s triggers are and how to mitigate accordingly. Investigators suggested integrating the list of triggers throughout ACCT, and including triggers as part of the review CAREMAP, especially when they are ongoing.
(iv)
The Suicide/Self-harm Risk Guidance was considered to be a very useful resource which focused the thinking of staff. Referral back to the Guidance should be emphasised throughout the ACCT document. It might be helpful
4
to give some examples of what level of observations might be expected for each level of risk. (v)
The Immediate Action Plan requires too much information, is overly detailed and badly designed with boxes too small to write anything meaningful and the four tasks too squashed at the end of the document. We would expect a proper assessment conversation to take place between a senior officer and the prisoner, and for the results to be recorded.
(vi)
The Assessment Interview was considered to be well-structured and useful. However, as noted above, it is often ignored and not referred to again, even in the first case review. This should be better used, and perhaps a prompt to refer to it during each case review should be added to the document.
(vii)
The formal inclusion of healthcare staff into the ACCT process should be added with a compulsory section which requires information from, and attendance by, a member of healthcare staff. For example, this could require healthcare to complete a summary of the prisoner’s mental health history or risk information. This would serve to increase the engagement with healthcare staff with the ACCT process. The member of healthcare staff attending should be required to read the prisoner’s medical record. If they have not done so and do not know the prisoner there is often no value to their attendance.
(viii)
The First Case Review should include a list of questions that require an action and a commitment from staff to undertake the action. For example: Has information from healthcare staff been included and considered in the review? Has the PER been considered? Enabling this information to be gathered and presented here should allow more effective engagement with the prisoner about relevant matters.
The First Case Review is often not
properly attended or takes place without proper preparation or a check on the person’s history. We suggest that the first page of the case review could be made into a checklist and include background on the individual e.g. risk/triggers/PER/healthcare concerns. Investigators thought that the expectations on the person chairing ACCT case reviews are not sufficiently clear (and this is rarely a consistent case
5
manager.) This could be remedied by providing a series of prompts for the Chair, at the start of the Review form. For example “Have you contacted everyone who needs to attend?” Too often, investigators have found ACCT First Case Reviews were held with no other staff in addition to the chair attending and that healthcare staff were not informed. Prisons need to consider how case reviews best fit into the working day to enable ACCT assessors and others to attend reviews. This should be supported by increased and tailored training for assessors. ACCT guidance needs to be written more clearly to identify exactly who is required to attend. Investigators considered that the very small space on the case review form to complete each section limited an effective record and believed this should be increased in size, with the opportunity to continue over the page. (A move to an electronic record might help. It is noticeable that we often get more detailed account of ACCT case reviews from nurses entries in SystmOne.) Subsequent case reviews should require each member of staff to agree and sign decisions. This should encourage greater engagement, teamwork and ownership, of the process. Risk should be more prominently displayed and considered, including the question: “Has the risk has changed?” and “What are we going to do about it (i.e. update the CAREMAP)?” Each case review should also include a box for the level of observations which have been agreed and set at the review. (ix)
In our view, CAREMAPs require a radical rethink. Staff often fail to understand how the form should be completed or to realise its value and to often it is just given cursory attention. Where a risk of self-harm is identified, the CAREMAP should not have as an action “Prisoner to stop self-harming”. Similarly, a “refer to mental health in-reach” should be far more robust, for example “ensure that the Mr X had a mental health assessment within 24 hours, sufficient support is provided and a care plan put in place”. Too many CAREMAP actions are about process – such as making a referral or getting an appointment. While these are necessary first steps, this leads to ACCTs being closed because actions such as a mental health referral have been completed even though no one from the mental health team has seen the prisoner. We put forward using a flow map to indentify how staff intend to reduce risk which would require a more proactive response; for example “How are we 6
going to reduce risk?” The document could be renamed “Practical and achievable next steps”. Staff need better training as to the principle behind the CAREMAP and what it needs to achieve. Where verbal or written input from staff about the mental health of a prisoner is required or cited in the CAREMAP, a signature should be required. We argue that a member of healthcare should have to attend each case review and sign off on the CAREMAP to indicate a consensus has been reached as to the course of action. (x)
The ongoing record sheet is often characterised as a long list of information such as “asleep on left side”. Records of observations are important, but often, buried within the detail or routine observations, are significant entries which could play a key role in keeping someone safe. For example, an entry of “spoke to girlfriend” is important where the contact with the girlfriend is considered to be a trigger. There needs to be a clear mechanism to highlight key information. It could be questions adjacent to the entry “Is this information significant? Do I need to act?” or it could be a colour coding system. Case reviews often miss important information in the on-going record and it is apparent in many cases that the chair and other staff attending do not read it.
(xi)
Some investigators highlighted the vagueness and passivity of the language used in the ACCT document as a barrier to its effective use. They suggested that more active language, which pushed staff to critically engage with the person, and in consequence, the process, would be of benefit. For example, the Concern and Keep Safe form could be changed to Why is an ACCT being opened? The Immediate Action Plan could be renamed “How will we keep this person safe until the first review?“
1. (a) Tailoring ACCT The ACCT process is used for children and young adults as well as adult prisoners over 21. In 2013, the PPO published a Learning Lessons Bulletin on Child Deaths which found that two of the three young people who died were on an ACCT at the time of their death. A number of failings were identified in the application of their ACCT plans. These mostly concerned a lack of child-centeredness and conflicts between care or safeguarding and discipline or control mechanisms, such as 7
adjudications and the rewards system. This prevented ACCT operating as a consistent, holistic and integrated system appropriate for children. The report recommended that ACCT processes should be more child-centred and involve senior managers, families and outside agencies as part of an effective care planning approach to managing young people at risk of self-harm and suicide. Our report on Young Adult Prisoners (2014) highlighted that a common theme among the 80 cases of self-inflicted deaths in young people was challenging behaviour. Many had spent time in the segregation unit or on the basic regime of the incentives and earned privileges scheme; these privations reduced the protection factors available to young prisoners vulnerable to suicide and self-harm. It is thus important that managing risk, treating mental health and managing behaviour need to be better integrated to ensure a balanced, holistic and consistent approach to prisoners in emotional turmoil. We found that there is a need for a more tailored to approach to supporting vulnerable young people and that staff working with children and young adults need to recognise the particular importance of attachment to families and partners; their lack of life experience can mean they are more emotionally affected by the break up of relationships and family bereavements. Prisons need to take this into account when assessing their risk. The report also identified a need for staff to be aware of the risk factors for young foreign national prisoners who are under threat of deportation. The report noted that even those proficient in English can find it difficult to understand and come to terms with their situation, but there was a need to ensure that those who did not had access to interpretation and translation services. All foreign national prisoners and detainees needed immigration advice or legal representation. The 2014 report also noted that staff often did not sufficiently engage with the correlation between bullying and vulnerability to suicide and self-harm, and advised that indications and allegations of bullying should be recorded, investigated and, acted upon to protect the apparent victim and address the behaviour of the alleged bully or bullies. Similarly, the ACCT process cannot currently be tailored to the gendered experiences of suicide and self-harm. For example, self-harm is relatively rare in the male estate and might therefore be considered a key indicator of increased risk; in the female 8
estate, self-harm is more prevalent and has less power as a trigger for suicide. We believe that greater consideration should be given to gender, particularly in the tailoring of risk assessments which then feed into the ACCT process. 2. Compliance Many PPO reports contain recommendations relating to the poor compliance by prison staff with the requirements of the ACCT process. These include, among others: a lack of staff understanding of the ACCT process; staff failure to understand risk; staff failure to open an ACCT; a lack of guidance to assist staff in identifying vulnerable prisoners; the need for ACCT to be a multi-disciplinary process; the need for CAREMAP’s to be achievable. We encourage the review to refer to our Joint Recommendations database which lists all recommendations made in cases relating to ACCT and which we recently shared with the Equality, Rights and Diversity Group at NOMS. Particularly worrying are those prisons where poor practice relating to the management of the ACCT process appears to be ingrained and endemic. For example, several investigations found that, at one large London local prison, staff made inappropriate alterations to the monitoring records of the prisoner who died. It would be helpful for the review to identify what is going wrong at those prisons where there have been persistent problems with ACCT procedures. The review should look at places where the PPO has identified compliance with ACCT to be particular poor, though we note that there may not be a direct correlation between the number of deaths at a prison and ACCT but rather poor risk assessments may be significant.
2. (a) A holistic approach: mental health provision and the involvement of Healthcare staff As noted above, the implementation of a successful ACCT does not occur in a vacuum but requires engagement with, and operation by a wide range of staff. Consequently, while this review is being run by NOMS, close liaison is required with the NHS and those private organisations who deliver healthcare in prisons. Due consideration must be given to the quality, type and availability of mental healthcare provision. It is important that there are good working relationships between 9
healthcare staff and prison operational staff. Where there is a reliance on locum staff, the transitionary nature of these staffing arrangements should not be allowed to impede the delivery of good quality healthcare. Equally, prison staff need to be clear that opening an ACCT is not restricted to a certain category of staff. Our experience is that there is a lack of integration between healthcare and discipline staff meaning that when an individual under the care of the mental health team, discipline staff sometimes disengage from their responsibilities under ACCT. It is important that ownership of ACCT is shared between these two sides of the prison and team working is a reality. 3. Delivery of ACCT 3. (a) Sources of information Investigators’ experience is that, too often, the ACCT process, and those administering it, rely too heavily on the prisoner as a source of information and not enough on the diverse, and important, alternative sources of information available. For example, contact with healthcare staff was often limited so operational staff were unaware that a prisoner had a history of mental illness or self-harm. Prison staff tell investigators, at interview, that they “spoke to the prisoner and he said he was fine” or “he had good eye contact”. While we acknowledge that direct engagement with the prisoner is important, and an assessment by staff of the individual should be central to ascertaining the risk, it must be balanced by a consideration of objective sources of information. We would like to see a system which enables risks and triggers from previous ACCTs to be carried over between ACCTs so that key information is not lost. For example, an individual may be on an ACCT and a trigger may be a court appearance which will not occur for several months. If the ACCT is then closed before the court date, knowledge of the court date as a trigger is then lost. A system, which enables a flag to pop up, on P-NOMIS or SystmOne, which would alert staff to the potential increase in a prisoner’s risk would be very useful. 3. (b) Understanding risk PPO investigations have shown that rather than being risk averse, some prisons and staff fail to fully understand the concept of risk and ignore or miss evident risk
10
factors. Suicide and self-harm warning forms are discounted and information from PERs is often not considered. We consider that where a prisoner arrives with identified risk factors it is better to open an ACCT so that a proper assessment can be made, even if the ACCT is closed shortly afterwards. In contrast, case managers are often not risk averse enough. For example, they need to realise that constant self-harmers are never going to be low risk. Prisoners on an ACCT require monitoring, at a frequency which directly correlates to their assessed risk. These observations must be undertaken, at irregular intervals within the specified frequency, and recorded clearly. PPO investigations have highlighted problems around this monitoring process. Staff may not adhere to the timetable of observations required, either due to competing pressures on their time, forgetfulness or neglect, and, where observations have not been carried out, entries are either not made or may be falsified. Understanding the risk posed by a prisoner, and the need to comply with the ACCT process, is integral to effective monitoring and protection. 3. (c) Enhanced case reviews For challenging prisoners and those with complex needs, enhanced case reviews may provide an appropriate mechanism for safely managing these individuals. Again, staff understanding of when a prisoner’s risk is sufficient to warrant this next step is often lacking and our experience has shown they are not used often enough. While we acknowledge that there are resource implications for this level of ACCT management, in some cases, enhanced case reviews, complemented by constant observations, may be the most effective method of managing challenging prisoners. We have investigated a number of self-inflicted deaths where the prisoner displayed behaviour which we considered should have at least led to some consideration of whether an enhanced case review process was needed. In some cases, we found that the failure to do so was to the detriment of the prisoner’s safety.
4.
Roles
Though we are not particularly focused on role, we highlight that when interviewed, prison staff often raise concern with investigators about the negative effect of benchmarking on the ACCT process and prisoner safety. For example, a number of prison staff commented on how reduced staff numbers had left a segregation unit at 11
one prison with insufficient staff to undertake ACCT checks, leaving prisoners vulnerable and staff under increased pressure. Reduced staffing numbers are often cited as reasons for poor attendance at ACCT reviews and a lack of consistency of case managers. Recent IMB Annual Reports have consistently highlighted the impact of benchmarking on the provision of services within prison and the delivery of protective measures for vulnerable prisoners, not only those on ACCT. For example, at one prison, the IMB found that reduced staffing levels had a negative affect on self harm, bullying, bad behaviour, drugs and violence within the prison. NOMS is clearly aware that the cuts to staffing levels have had an impact on the ability of prisons to provide a safe custodial environment. Some prisons have been able to manage the impact better than others, although this is also connected to the kinds of prisoners that establishments receive and the subsequent pressure exerted by this population on resources. It is clear however, that some prisons are struggling to manage prisoners at risk of self-harm and suicide. 5. Changes to the ACCT process In the PPO thematic report on ACCT, we found that over one in four (16 out of 60) prisoners did not have an adequate CAREMAP drawn up. The inadequacies found in CAREMAPs included insufficient support being given to help prisoners achieve their specified goals. Too often it was stated on the CAREMAP that it was the prisoner’s sole responsibility to reach their set goals, with no other support mentioned as part of the action. It is advisable to specify a named individual as opposed to a department because if the action is not assigned ownership, it is likely that it will not be completed. In one case, a prisoner was given the responsibility on their CAREMAP to achieve the goal of completing a transfer request to another prison. While it is important for the person on the ACCT to be given some responsibility for improving their situation, it is inappropriate to put the onus of achieving a prison transfer solely on the prisoner. In this case, the goal would have been more achievable had a member of staff also been named, as the prisoner had no power to secure a transfer. There is too much focus on process rather than outcomes. If a transfer was the outcome to be achieved to reduce the prisoner’s risk then that should have been the CAREMAP action rather than a request for a transfer.
12
CAREMAPS should be assessed to ensure they are still relevant to the prisoner’s current needs and risks. This process should be carried out in subsequent reviews and should be the subject of rigorous management checks. This will also help to ensure the effective continuity of care which is key to the successful implementation of any ACCT. In the PPO research, over one in four prisoners did not have known trigger points correctly recorded on their ACCT plan. The implications of not recording trigger points in detail can result in a missed opportunity to intervene when a prisoner presents at their highest risk. While triggers can sometimes be difficult to identify, there is a better chance of uncovering them if there are staff from all areas of the prison present at the case review. Information which might not have been shared before may come to light. If there are no triggers identified, this should be recorded, instead of leaving the form blank. The form can then be updated as and when case reviews happen in order to review whether the individual’s circumstances have changed. 6. Culture It is not always obvious that the Prison Service is clear on the purpose of the ACCT process. The impact of this on the culture within prisons about ACCT has been the over-emphasis on the need to close the process. For those at risk of suicide, it is vital that the fluctuations in risk are properly assessed and responded to. Too often the first few signs of stability or reduced vulnerability leads to ACCTs being closed prematurely. The personal officer scheme appears, at least initially, to offer a productive opportunity for staff to engage with vulnerable prisoners. We found, in our examination of self-inflicted deaths since January 2012, for which we hold data, that half of those who died had personal officers and that 47% of those were quite well known to their personal officer. This data raises the question of the efficacy of the personal officer scheme and how it relates to providing support to vulnerable prisoners and those under ACCT. 7. Training In our experience, most prison staff genuinely want to do a decent job and keep prisoners safe. However, the ACCT process can easily become a hollow 13
administrative exercise, lost in the challenge of delivering the normal daily regime. In order for all sections of the ACCT process to be conducted and managed in the most effective way, it is imperative that staff receive relevant training on ACCT procedures. Training should be kept up to date and refresher courses should be provided for staff. We have found that staff often lack the suitable level of training or indeed may have had little to no training at all. Where training has been received, staff lack confidence or may not have fully understood all the relevant elements of the training. For example, staff have told us that they “don’t want to know the prisoner’s offence, so they won’t be biased”. What this means in practice is that a key risk factor will be ignored where it is relevant, for example, at an ACCT review. Statements about not wanting to judge a prisoner by being aware of their offence, suggests that staff do not have all of the information to keep a prisoner safe. Prison staff have also told us that the training was not sufficiently practical and so they rely on their intuition so assess risk. Investigators have found that ACCT case managers often over-estimated their capabilities in assessing risk and their knowledge of the prisoner. Assumptions relating to how to keep the prisoner safe were built on this foundation but were found to be weak. We acknowledge that risk assessments are difficult but they require the effective gathering and use of all available information, an aspect of the process often missing. We are also concerned that the training focuses heavily on the completion of the ACCT documentation and not on understanding suicide and self-harm, risks and triggers. Finally, training needs to tackle the myths and stereotypes which abound about suicide and to rid the Prison Service of attitudes such as “We can’t open an ACCT on everyone”. 7. (a) Learning lessons Our research has shown that some prisons have warranted similar recommendations in a number of cases where problems with the ACCT process was a factor in the death. The response to many of these recommendations has been for the Governor to issue a notice to staff about the issue. This is clearly insufficient. However, we hope that part of the ACCT review will identify prisons that need to go beyond the issuing of notices and engage with training and more robust methods of
14
improving the implementation of the ACCT process. It is clear that some prisons need to work harder at changing the culture and attitude within their establishments towards the ACCT process and how it should be used effectively. In prisons where there have been a high number of self-inflicted deaths, it could be argued that this need is obviously urgent. Certain prisons have been flagged by investigators as ones where lessons have consistently not been learnt. It is not enough that the PPO merely reiterates previous recommendations and more people continue to die. NOMS needs to consider more effective mechanisms to share lessons within prisons and across the prison estate, highlighting good practice and robustly focusing on areas where things go wrong. 8. Specification and Policy considerations Please see above our comments above regarding amending the PSI. We have set out below, the key lessons from our recent publications, as they relate to ACCT. Learning Lessons Bulletin on Segregation (2015) noted that twenty-eight prisoners took their own lives while being held in segregation units between January 2007 and March 2014; nine of them were subject to ACCT procedures at the time of death. In several of these cases, no exceptional reasons to justify holding the prisoner in segregation were recorded. Prisoners who are managed under ACCT procedures are particularly vulnerable, and locating them in segregation units should be avoided wherever possible. Some vulnerable prisoners may also be very challenging, particularly if they have complex mental health needs. This can leave prison staff with some very difficult decisions about where prisoners managed under ACCT procedures should be held, in order to minimise the risk of harm to themselves – and others. Our investigations have found that, too often, prisoners identified as at risk of suicide and self-harm and being managed under ACCT procedures were held in segregation units without sufficient evidence that staff had considered other options or identified exceptional circumstances to justify their segregation. Where there are exceptional reasons to justify holding prisoners who are managed under ACCT procedures in segregation units, there are some additional requirements that need to be met by prisons. As well as the initial health screen for all prisoners moved to a segregation unit, PSO 1700 makes it clear that prisoners subject to ACCT procedures should have a mental health assessment within the first 24 hours of their segregation. An ACCT review 15
should also be held within 24 hours. Mental health safeguards, for example observations and dialogue, should be put in place, and consideration given to the possibility of moving the prisoners to a safer cell or monitoring by CCTV. The PSO makes it clear that prisoners on ACCTs should remain in segregation only as long as the exceptional circumstances continue to apply. Segregating a prisoner when they are already identified as at risk of suicide and selfharm often heightens their vulnerability. If there is no alternative, it is essential that staff follow all the specific procedures designed to ensure the prisoner’s safety while in segregation, in addition to those that apply to all prisoners being managed under ACCT procedures. Ensuring ACCT procedures are correctly followed is essential for all at risk prisoners, but is especially important for prisoners whose vulnerability may be increased by segregation. Learning Lessons Bulletin New Psychoactive Substances (2015) highlighted that for some people, it appears that NPS can be a trigger for self-harm. Whether taking NPS was a direct causal factor in self-inflicted deaths is very difficult to establish. In most such deaths, we can only ever hope to make informed judgements about the individual motivation. To add to the difficulty, in some of the cases we investigated, we only had hearsay evidence that prisoners were using NPS type drugs. However, we have seen a small number of self-inflicted deaths among prisoners believed to be using synthetic cannabis, where their actions were unexpected and seemingly unplanned. Learning from PPO investigations: Self-inflicted deaths of prisoners on ACCT (2014)
The goals set on CAREMAPs should be realistic, achievable and relevant.
A named member of staff should be specified next to each CAREMAP goal and a target date set for completing the goal.
The trigger section must be completed on all ACCT plans, even if it is to say that there are no known trigger points.
Triggers should be reviewed and updated as and when is necessary for the individual.
Staff must conduct an immediate case review if a trigger behaviour is observed, as set out in PSI 64/2011.
ACCT reviews must be timely and multi-disciplinary, with consistent staff attendance, if possible. 16
Staff from across the prison and agencies working within it should be invited and encouraged to attend ACCT reviews and offer input into the individual’s care.
ACCT plans should be treated as a holistic approach to managing an individual’s care and support.
All staff who come into contact with the individual should be responsible for updating the ACCT plan if they feel that their risk of self-harm or suicide is heightened.
All staff should be up to date on their ACCT training and those that are not should be provided refresher training at the earliest opportunity.
Learning from PPO Investigations: Risk factors in self-inflicted deaths in prisons (2014)
There should be clear local procedures which require prison and healthcare staff in reception to actively identify risk factors together based on checks of relevant documents such as the Person Escort Record, pre-sentence reports, NOMIS, and clinical records. Reception screening needs to take fully into account concerns from others about an individual’s risk to themselves, such as the police, escorts, the courts and families.
Evidence of risk should be fully considered and balanced against how the prisoner presents themselves. Reception staff should record what factors they have considered and the reasons for decisions.
An individual’s level of risk is not fixed. Distressing and stressful events can have a sudden and critical impact. Where such information is known, staff working closely with the prisoner should be made aware.
A third of the prisoners had seen healthcare staff in the 72 hours before their death. This represents a key opportunity to intervene. Healthcare staff need to be confident about initiating and using ACCT monitoring and be clear when to share concerns about prisoners more widely. Similarly, prison staff need to ensure, particularly in reception, that healthcare staff are given all relevant information about risk and that this is discussed with them.
Many prisoners will attempt to withhold the extent of their distress from staff and other prisoners. In this context it is important to act promptly on any concerns family and friends convey to the prison.
NOMS should amend PSI 64/2011 to set out a clear, standardised list of risk factors for suicide and self-harm which includes being held on remand as a risk factor for suicide.
Learning from PPO investigations into three recent deaths of children in custody (2013) 17
Busy YOIs struggle to give the individual attention necessary to care for the most vulnerable. Accordingly, allocation to an STC or specialist unit within YOIs, such as the Keppel Unit at Wetherby, needs to be considered and pursued.
There are problems sharing information and a lack of shared understanding of vulnerability which can hinder co-ordinated care of the children across agencies.
Assessments of vulnerability and risk of self-harm did not adequately weigh static risk factors against presentation or fully take into account the complex ways children can show emotional distress.
ACCT processes were insufficiently child-centred, and the involvement of senior managers, families and outside agencies in care planning was too limited.
Managing risk, treating mental health, and the wider operation of YOI processes, particularly disciplinary procedures, need to be better integrated to ensure children are treated holistically and consistently.
YOIs need to ensure a more robust response to bullying and that reports of bullying are acted upon.
Personal officers offer an important point of contact and support. They should be assigned quickly upon reception and regular contact with the child or young person fully documented.
Sources of external support, including but not limited to families, can be extremely important. Enhanced access to this support at times of crisis and for those at particular risk of self-harm should be facilitated wherever possible.
August 2015
18
