Development of a specialised forensic service for women with learning disability: the first three years

Development of a specialised forensic service for women with learning disability: the first three years

Berber, Elizabeth


In recent years there has been growing interest in the fate of those women with mental disorder who come into contact with the criminal justice system. This interest has stemmed from growing recognition that traditional forensic services could not offer the appropriate care required by this group in a conventional mixed-gender environment. Women-only services have begun to be developed in generic psychiatric settings, spurred on by the national service framework (NSF) which set a time limit for the development of segregated in-patient facilities. Forensic services for those with learning disability have been slower to take up the challenge of how best to place women with learning disability who offend and require an in-patient secure environment. This article describes how one such service attempted to rise to this challenge and build a service for this often neglected group.


In January 2000 a new service for learning-disabled women opened in the West Midlands. This service was conceived out of a recognition that women with learning disability who had attracted the attention of the criminal justice system were often ill-placed in the setting in which they subsequently found themselves.

Specialised services for male offenders with learning disabilities had already been in the region for some time, initially as part of the services located in a large learning disability hospital, and later in the form of the Janet Shaw clinic, a 12-bedded purpose-built medium secure unit (MSU) which opened in 1990 (Sansom & Cumella, 1995). Although this unit catered initially for both men and women, there were never more than one or two women on the unit at a time, and the last woman was admitted in 1996. Between 1984 and 1991 about 14% of admissions to the Janet Shaw services were women. This figure fell to only eight per cent between 1991 and the opening of the new unit. It had become apparent that a mixed-gender environment of this type was highly inappropriate, particularly given that many of the female patients had a past history of sexual abuse and some of the male patients had histories of sexual offending. Before January 2000 there were no other dedicated forensic units for women with learning disability in the West Midlands, and very few in the UK. Other secure learning disability services had continued to admit women, as women continued to be referred. Alexander et al (2002) found that about 20% of referrals to the Eric Sheppard Unit, a learning disability medium secure unit in Hertfordshire, were for women. With the reduction in the number of beds in high security psychiatric hospitals, and the increasing reluctance of MSUs to admit women, it was clear that the needs of such women had to be addressed.

The need for specialised forensic services for women

Interest has been growing in the suitability of some of the disposal options for mentally disordered women offenders. A sizeable body of literature has been produced on this subject and, although very little of it is concerned specifically with those with learning disability, these women are often identified in the populations under study. Studies of women prisoners have generally found high rates of mental disorder. Parsons et al (2001) found rates of 60% among the inmates of HMPs Holloway and New Hall, while Maden et al (1994) found women prisoners to have higher rates of learning disability than male prisoners – six per cent compared with two per cent. Dell and colleagues (1993), in a study of remand prisoners at HMP Holloway, identified 19 women who were classified as ‘mentally handicapped’ (learning disabled) in a sample of 101 women who had been referred to an outside psychiatrist in some capacity.

Their findings illustrate very well the typical problems encountered when there is virtually no suitable disposal. Ten of the women had been charged with minor offences, and had been remanded in custody not because of the seriousness of their offences, but because of concerns over disturbed behaviour. Although this group showed a high degree of instability, along with serious social problems, only one of them received a hospital placement (she had a moderate degree of learning disability). The remaining nine were not offered placement, because the visiting psychiatrist did not believe that hospital would be ‘likely to alleviate or prevent deterioration’. Five women in the sample had been charged with arson; four of them met placement difficulties. One was eventually admitted to a high security psychiatric hospital, although the authors commented that few doctors thought that maximum security was necessary. For another, no local secure facilities for people with learning disability were available so, after she had waited for two months in custody, a placement in the private sector was eventually found. Two other women suffered months of delay during which consultants disagreed about the level of security required and, losing patience, the magistrates in one case discharged the woman and in the other imposed a two-year prison sentence. The study by Maden and colleagues (1994) identified similar gaps in the availability of adequate services, and described two women with learning disability and personality disorder who were receiving ‘nursing care’ in a segregation unit from untrained staff’in an environment reminiscent of the worst asylums’.

These studies are almost a decade old, but day-to-day experience suggests that nothing has changed. Although it is easy to criticise the services which appear so indifferent to a so obviously needy population, the closure of large psychiatric hospitals and the ever-dwindling long- and medium-stay inpatient resources have severely limited what local and regional services can realistically offer to an extremely disturbed group of individuals.

For those women who are offered placement in mental health services, the appropriateness of the provision has been called into question. Despite the Government’s plans, via the national service framework, to revert to single-sex wards, most of the NHS psychiatric provision remains mixed. Women patients on these wards report high levels of sexual harassment and intimidation from men (Thomas et al, 1995). The Reed Report (DoH & HO, 1992) identified women offenders and those with learning disabilities as special groups, and noted the fact that there appeared to be a number of women detained in high security psychiatric hospital who did not require that level of security. Women were outnumbered in these settings, and had often entered this level of security because of limited options elsewhere.

High security psychiatric hospitals at least have always operated strict segregation policies. Women make up about ten per cent of the population of these hospitals. Bland and colleagues (1999) reviewed female admissions to Broadmoor Hospital and found that 26% of those admitted were felt to have ‘borderline learning disability’. They state that 35-50% of admissions to high security psychiatric hospitals do not require high security. Thomson et al (2001) reviewed female admissions to Carstairs Hospital between 1992 and 1993, and found that, despite having lower levels of criminality, they appeared to be more controlled and restricted than their male counterparts. Although they were more controlled, the women were thought to require a lower level of security, lack of alternative facilities once more being the reason suggested for women’s presence in this environment. Woods & Mason (1998) identified admissions classified under ‘mental impairment’ to Ashworth Hospital over a 20-year period. This group made up 13% of the admissions (the majority of them being in the mild range). Twenty-seven percent of women in-patients had not been charged with any criminal offence (three times the rate of non-impaired admissions).

Medium secure units came into existence following a report by the Butler Committee which highlighted the gap in provision of service level between high security psychiatric hospitals and non-secure psychiatric services. The potential patient population who would inhabit these new units was described in the Glancy Report (DHSS, 1974) and included those:

who may be mentally ill or mentally handicapped, or suffer from psychopathic or severe personality disorder alone, or in conjunction with mental illness or mental handicap.

Those with severe handicap were not included. The original intention was to include those with learning disability, but the Royal College of Psychiatrists thought that separate secure facilities were needed (Royal College of Psychiatrists, 1980). Puri and colleagues (2000) compared two medium secure units, one for those of normal intelligence and one for learning-disabled people. They found significant differences in diagnoses, index offences and contact with the criminal justice system, and pointed out the differences in milieu between the two types of unit. Nursing staff in the unit for those with normal intelligence were trained to manage severe mental illness, whereas the learning disability units used a more behaviour-orientated model of nursing.

The Butler Report also suggested that length of admission should be in the region of 18-24 months which is unrealistic in learning disability units, where long-term behavioural interventions may be used. Hassell and Bartlett (2001) conducted a telephone survey of 39 medium secure units in England and Wales and found that about 19% of the patients on these units were women. Ninety-four percent of the NHS establishments were mixed gender. Where singlegender units existed in the NHS, 71% were for men; only 23 beds specifically for women existed at this time in the NHS. This contrasts with the private sector, where 50% of women were in single-sex units. Four of the NHS units surveyed said that they had never, or no longer, admitted women patients. The authors concluded that this was because units were aware of the inadequate and inappropriate service provision for women in a mixed environment and so had decided to cease admitting them. Very few units reported plans to open women-only facilities.

Development of a service model

The first steps in choosing a service model for the new secure unit for female learning-disabled offenders was to review the literature and determine the range of problems encountered among women offenders. Studies have been consistent in their findings of high levels of deliberate self-harm, past histories of sexual abuse and diagnoses of personality disorder, especially borderline personality disorder, in this group. Adshead (1994) looked at women referrals to the forensic service at St George’s Hospital. She found that 81% reported childhood sexual abuse, mainly by perpetrators known to the patient, and a further 56% had experienced sexual assault during adulthood. The most common diagnosis was personality disorder (66%), a further 44% of whom were specified as having borderline personality disorder. All the women had committed at least one act of deliberate self-harm, and 87% were multiple repeaters. Thirty-eight percent had received a diagnosis of schizophrenia, although the author was doubtful about the validity of the diagnosis, questioning whether the voices heard were in fact true auditory hallucinations. A third of the sample had been referred because of threats of, or actual, assault on health care workers (nearly all of whom were themselves female). It is often assumed that all-female units are best staffed by all-female staff, but from a safety angle this may not be the best policy. Adshead (1994) notes that there is sometimes a perception that only other females can help these patients, but that female workers could actually be more at risk.

Coid and colleagues (2000a; 2000b) reviewed admissions to all high security psychiatric hospitals in England and Wales and half of the MSUs, comparing male and female patients. They found higher levels of depression, anxiety, phobias and self-harm in the women. They were also more likely to have epilepsy, and had a lower mean IQ. Cluster analysis on the women, to attempt to define different categories of symptomatology, revealed three groups: personality disorder, major mental illness and organic brain syndromes. The largest group was the personality-disordered group, consisting mainly of antisocial and borderline personality disorder subtypes. Four per cent of the antisocial group and five per cent of the borderline personality disorder group had IQs below 70. Those with a diagnosis of mental illness or organic brain syndrome had rates of learning disability of nine per cent and seven per cent respectively. Bland and colleagues (1999), in a survey of women patients in Broadmoor, also found high rates of sexual abuse and deliberate self-harm. Dissociative episodes were common, and were often mistaken for psychotic symptoms, leading to discrepancies in diagnosis. More than a quarter of the sample was felt to be in the borderline learning disability range.

Although the problems inherent in the female offender population are diverse, the themes of past abuse and self-harm recur frequently. Features of borderline personality disorder include affective instability, lack of impulse control, disturbances of self-image and sexual preference, chronic feelings of emptiness and a tendency to form intense unstable relationships which, if threatened, lead to excessive efforts to avoid abandonment, often with threats of, or actual, self-harm. Any service which plans to admit such patients needs to be geared up to be able to contain and manage therapeutically the high degree of disturbance, both emotional and behavioural, so often present in those who live with the aftermath of a traumatic past life. These women have a knack of exhibiting the types of behaviour that leave staff feeling anxious, angry and frustrated. They often appear to be deliberately sabotaging their treatment plans, and can cause splitting in the clinical team and generally cause bad feelings. Such conduct is particularly difficult to manage in a mixed-gender environment which is set up for the severely mentally ill, and it is probably no coincidence that Coid et al (2000a) found that a third of women in high security psychiatric hospitals had been transferred from MSUs, even though their rates of absconding showed that they were less likely to need high security.

Once it had become clear that a dedicated service for women was indeed the most appropriate way to deal with these particular problems, it became necessary to look at what could realistically be done to address these issues. Most of the research into treatment options has been done with non-learning-disabled populations, and it is unclear whether direct extrapolation is justifiable. Sequeira and Hollins (2003) reviewed the available literature on the consequences of sexual abuse for people with learning disabilities and, although the methodology of many of the papers was not rigorous, it did appear that people with learning disabilities who have been sexually abused do suffer from a range of psychiatric and behavioural problems similar to those of the general population.

Bland and colleagues (1999), in the Broadmoor study, pointed out that, despite the high levels of psychological disturbance in their sample, only 32% were involved in any sort of psychotherapy. They suggest that, in order for women to make meaningful connections between their past experiences and their current predicaments, psychotherapeutic input is necessary. There are a growing number of models of therapies aimed at, or adapted for, the treatment of borderline personality disorder. Dialectical behaviour therapy, interpersonal therapy, Hobson’s conversational model of more traditional dynamic psychotherapy and cognitive analytic therapy have all been used with outpatients (Winston, 2000). Dialectical behaviour therapy has been adapted and evaluated as an in-patient treatment (Barley et al, 1993) and there is some evidence that a therapeutic community setting can convey benefit to those with a diagnosis of personality disorder (Dolan et al, 1996). Dolan et al (1996) also suggests that hospital admission can actually save money for this group, particularly for those with forensic histories. Other models for treating borderline personality disorder currently in favour include considering the symptoms consequent upon chronic abuse and emotional deprivation as part of the spectrum of post-traumatic stress disorder, and devising treatments along these lines. The principles of attachment theory and their relevance to the genesis and management of these problems are also gaining ground.

Adshead (1998) described the concept of maladaptive behaviours arising from insecure attachment in childhood (as the result of abusive parenting) being repeated in adulthood, especially at times of separation or threat to security. She describes how professional carers may be at risk from the anger and violence from the past abuse during these traumatic re-enactments, as care givers may take on the role of attachment figure. While providers may not be able to alter radically the patient’s psychopathology, their role is to contain anxiety and arousal and provide a safe base while the patient develops new skills in order to modulate their own emotions and reflect on their attachment patterns. The philosophy of care should be one of ‘being with’ rather than ‘doing to’.

As with any complex disorder, the therapies purporting to help are legion, and variable in outcome as far as the individual is concerned. Treatment models need to be flexible and tailored to the individual, particularly as not all women who find themselves in the forensic services fit into the category of borderline personality disorder. Any over-riding model, therefore, must hold this diversity in mind and be broad enough in its aims and objectives to encompass this. Charles Kaye, formerly Chief Executive of the Special Hospitals Service Authority, reviewed the provision for women and made proposals for an alternative pattern of care that would apply nationally (Kaye, 1998). As well as improving the physical environment, the proposals aimed to provide a strategy and standards of care. Following consultation with outside agencies, a number of ‘hallmarks’ were identified by which they felt a service for women could be judged. They were to be applied equally to both high and medium security in both NHS and private settings. The proposals were broadly as follows.

Providing an appropriate environment

This should be a haven for women, but not be isolated – the women’s unit could be linked to a similar men’s unit. The atmosphere should be sympathetic and gender-sensitive. There should be a demonstrable philosophy of care, agreed by all professionals and written down.

Care and treatment

Patients should have active involvement in their care plans. The significance of post-traumatic events should be recognised. There should be clinical strategies to respond to and reduce deliberate selfharm. Members of staff (both men and women) should be trained to work with women. There should be effective networking with outside contacts.

Demonstrating validity

There should be criteria for demonstrating clinical progress both on the unit and afterwards, and research with feedback to enable practice to be improved and modified. Treatment should be costed for each individual.

The first steps to building the service

The physical environment

Brooklands (formerly known as Chelmsley Hospital) is situated on the edge of Birmingham and is part of North Warwickshire Primary Care Trust. The site originally consisted of villa-style wards, typical of the Victoria asylum era, which had been erected in 1878 as a children’s home. Today’s landscape would be unrecognisable to its old inhabitants. The original buildings have been largely replaced by a series of mainly low-risk ‘homes’ and admission units. The previously spacious grounds have been swallowed up by one of the now ubiquitous housing developments which encroach to within a few inches of the perimeter fence.

Not only has the physical landscape changed, but also the patient population has altered. Pressure to discharge the traditional long-stay patients into community settings has caused the emphasis of medium- and longer-stay admissions to shift towards those with challenging behaviours and forensic problems. The efflux of the old residents left several vacant buildings across the site, providing the potential to use these premises for the development of new services. It is in this context that a former admissions unit, One the Oval, became the first home of the women’s learning disability forensic service.

One the Oval is a five-bedded, single-storey unit which is semi-attached to a similar unit. They share a common entrance area where storage and laundry facilities are located. The adjacent unit is an open male forensic ward. The door leading into the unit is kept locked, but there is no airlock. The door leads directly into a large lounge/dinning area. A small staff office and a kitchen open directly from this area, making observation reasonably easy. Access to the kitchen for patients depends on current mental state.

The dining area leads, via patio doors, to a garden enclosed by a sturdy six-foot fence. Leading away from the main area is a corridor on which the bedroom and bathroom facilities and a de-escalation area are situated. All the bedrooms are single and have a small observation window. If a patient has been unsettled and wishes to be in her room with the door closed, then there is an agreement that the curtain covering this window is left open. There are no seclusion facilities.

The unit was not purpose-built for a population prone to disturbed behaviour and self-harm, so a number of adaptations were required to the fabric of the unit before opening – window restraints and a lock on the kitchen door, for example. Other safety measures came to light once the unit was running. Radiators thought to be securely attached to the walls were found not to be, light fittings which appeared tamper-proof at first glance could be used for self-harm, and so on. Measures are in place to minimise the risk of absconding, cutlery is not available at times other than mealtimes and is counted in and out, lighters and matches are kept by staff and non-breakable crockery is used on occasions. However, the environment was never intended to be 100% risk-free. It was felt important that ‘relational’ and ‘internal’ security, provided by a high staff-to-patient ratio, was more important than a high perimeter fence or a materially impoverished environment. We have sought a fine balance between acceptable safety and a reasonably ‘normal’ environment, the aim being for the locus of control for self-harming behaviours to come from within rather than from artificially restrictive surroundings.

In May 2001 a second five-bedded unit on the site, Three Sycamore Crescent, became the next building to be vacated. This was one of the ‘homes’ used for a more disabled group, and was even less suited to those with any degree of challenge than the first unit. The living, dining and kitchen areas were more discrete, making observation less easy. Some adaptations were made but, as it was envisaged that this unit would cater for a less disturbed population, a more homely environment was possible. It also had the advantage of being joined to a second bungalow, soon to be vacated.


The concept of ‘relational’ security has been key to maintaining a safe environment in the absence of those environmental measures normally associated with medium security. From the outset, a high staff-to-patient ratio was felt to be the most appropriate therapeutic method of ensuring that risk was kept to a minimum. All the qualified staff who currently work in the service have a background in learning disability nursing. In addition to the qualified members of staff, including a home manager and deputies, there are 15 unqualified care assistant positions. All but one member of staff are female. It was initially planned that all staff would be female, but it became apparent that this was neither necessary nor desirable. During the daytime the aim is to have five or six members of staff present each shift, with three on each night shift. One member of unqualified staff has taken on the role of providing day services specifically for this unit, three days a week. Unfortunately, this person is not supernumerary and the staff member is counted within the day’s staffing numbers. The second unit has a lower staff-to-patient ratio.

Both units are currently run by a single home manager. Many of the nursing staff were already employed by the Trust and were working on site. No-one had worked specifically with women offenders, and staff were drawn from diverse backgrounds, some having worked with those with a high level of mental and physical disability and others within the forensic service for male offenders. Newly qualified members of staff spend a period of time rotating through the various services offered on the site. Staff usually decide quickly whether this sort of work is for them, and it has been very apparent that to make such a service succeed, those at the coalface must want to be there.

The service also has one full-time psychologist and 0.5 full-time equivalent of a speech and language therapist, although, sadly, this position has not yet been filled. Medical staff time for the original unit was calculated at 0.4 of a consultant and 0.4 of a staff grade. Currently no occupational therapy services operate on the unit. There are wellestablished services outside the women’s service, which are available for all the units on the site. They include crafts, snoezelen, exercise facilities, swimming pool, horticulture, a light industrial therapy unit and music therapy. For patients who are able to benefit, community activities have been made available, including horse riding, college courses and voluntary jobs.

Staff training has been mainly in-house and has been provided by our own nursing, medical and psychology staff. The format of the training has included occasional lectures, covering different aspects of borderline personality disorder, for example, and full-day presentations. However, for all staff to be absent from a specialised unit together is difficult logistically and depends on the goodwill of management and other site staff. Staff members have also been able to attend and speak at national conferences and at events specifically targeted at those interested in the provision of women’s services both in learning disability and more generally. These events are very useful for ‘networking’ with other similar services, which are also in the early stages of development, to allow sharing of experiences and ideas. Personal, formal study of relevant psychotherapies has shown the extent of the commitment to the service of a few of the members.

The patients

Once the structure and staffing were in place, the first patients were admitted. The service was needsled from the outset. Referrals came from a number of sources, including high security psychiatric hospital, regional secure units, local community services, private sector facilities and prisons. It would have been possible to fill the first five beds several times over. Not surprisingly, the majority of referrals were for those with mild learning disability and borderline personality disorder, especially those who were difficult to manage in community settings with high levels of deliberate self-harm and offending behaviours, particularly violence. Many referrals came for those in other secure facilities who, because of their gender or learning disability, were misplaced in that setting. Although the unit had been set up for five admissions in the first few months, we found ourselves with six patients, which lost us the use of the de-escalation room. Of the first six patients two were on Section 37/41 of the Mental Health Act, two were on Section 3 and one on Section 37, and one was admitted informally.

Offending histories were typical of this population – violence to others, including care staff and their own children, fire-setting, damage to property and theft. Psychiatric histories were also typical: high levels of deliberate self-harm (usually self-injury rather than self-poisoning), some of which was extreme, substance abuse and a past history of sexual and physical abuse. Axis I diagnoses included those of depression and psychosis. The latter diagnosis, as discussed above, has been difficult to confirm, as the events often witnessed on the ward which could be perceived as psychotic in nature could also be explained as dissociative phenomena. Most patients were receiving antidepressants and antipsychotics on admission. The profiles of those admitted to the new unit were similar to those of the original patients in terms of past psychiatric history and offending behaviour, but these patients were not felt to require the level of security offered by One the Oval. Two of these patients, however, were rather atypical in that there was no past history of abuse and offending behaviour had taken place due to circumstances rather than personality problems per se.

In the early days of the service there were high rates of deliberate self-harm, some of which were severe, for example the insertion of a hypodermic needle into the neck causing a pneumothorax, ingestion of inedible objects including batteries, a coat hanger and a pair of glasses, and the insertion of objects into the forearm and vagina. Some of this activity resulted in surgery. All the selfharming behaviour was longstanding and had been a problem even in areas of higher security. Violence and aggression towards staff have also been prevalent, ranging from verbal aggression to premeditated physical attacks. Absconding has occurred, once on each unit. Yet despite all this, the service has so far survived without having to give up on any of the admissions.

Progress of the service

The first major step forward for the service was the opening of the second unit. Many of the referrals had been for women who, though inappropriately placed in their current situation and in need of a specialised service, did not present with a degree of disturbance necessitating admission to the acute unit and for whom such a placement might actually have proved detrimental. The often-noted phenomenon of women copying each other’s behaviour has proved very real. The question had also arisen of where women on the acute unit would move on to, once they no longer required the security of One the Oval. None of the units on site was suitable to offer a continuation of the rehabilitation started in One the Oval, as most of them cared for a more disabled population with more traditional types of challenging behaviour. The work needed to continue in an environment with a similar structure and philosophy, but with a lesser degree of security. It had been accepted from the outset that short admissions or a quick turnaround back into the community were not going to be an apt pattern of care for these patients, particularly those subject to Home Office restrictions. A clear pathway of progression through the service would be needed for some of the patients. The second unit made both these aims possible. Shortly after opening, two patients were transferred from the acute unit, which gave back the de-escalation room and allowed the seventh acute admission.

Progress of the patients

During the first year of the service there were high levels of disturbance in both units. However, as the months passed the initial – at times almost constant – tumult gave way to a substantially more peaceful truce. Relationships have been built between patients and between staff and patient. Although we have never aimed for a therapeutic community approach, the natural evolution of ward dynamics has contributed greatly to the improvements seen. Incidents of violence and self-harm have decreased significantly. We have discussed on many occasions the most appropriate way of monitoring progress, but have not yet found any suitable instrument which could chart the outcomes we feel are relevant. All patients on site have been monitored using HONOS-LD (Roy et al, 2002) and the aberrant behaviour checklist. These do not, however, capture the type of behaviour we are concerned with, as they were developed for a very different population. One crude way of attempting to document clinical improvements has been simply to monitor incident forms. Such a form is filled in each time a patient harms herself, harms another or damages property, etc. They are filled in consistently and filed chronologically, and allow quick comparisons to be made over time.

The less tangible aspects such as trust and the quality and depth of relationships are almost impossible to measure. For those patients who had been admitted from other long-stay establishments there was a concern about their attachment to that institution. These establishments are often the only secure base a person from a highly disorganised background has ever known, and leaving the buildings and the staff who have cared for them for many years can be a huge emotional wrench. These patients were introduced gradually to the new unit on increasingly long periods of leave prior to transfer. Staff from the old placements have kept in contact with their old charges by phone and visits, which has smoothed the transition process considerably. At times of high disturbance, such patients initially would demand to go back to where they had come from or behave in such a way that they believed they would be sent back. Eventually this ceased to be a problem, and the service has developed so that there is a real possibility of moving on from it; the attachment has clearly shifted from the old institution to ourselves.

Nursing and therapeutics

The mere act of moving into an appropriate environment has, for most of the patients, proved to be one of the most significant therapeutic contributions we have made to their ‘treatment’. Being in an environment which has been provided specifically for their type of problem, with staff who are there because they want to be, is a sharp contrast to the situations from which some were admitted. The service has aimed to provide a setting in which relationships can be explored in a safe, containing way. Each patient has a ‘named nurse’ and individual care plans for each aspect of her treatment and areas of risk. The latter is assessed using a standard trust assessment ‘tool’ and also by the Mental State Checklist, which is a home-grown instrument felt to be more practical for everyday use where risk levels for an individual patient can vary markedly during the course of even a day.

Pharmacological treatments continue to be used. Most patients on admission were on high levels of psychotropic medication, particularly antidepressants, antipsychotics and mood stabilisers. The continuing need for medication was assessed individually for each new admission, and we have been able to make substantial reductions for some. This has been possible because they are in a specialised setting. For others, many years had been spent getting the medication right, and for them we left well alone and resisted the temptation to interfere. We have recently become interested in the use of Clozapine for severe borderline personality disorder (Chengappa et al, 1999; Frankenberg & Zanarinin, 1993), following its successful use with a community patient.

Psychotherapeutic approaches are the mainstay of treatment for our patients. As with most things in the world of learning disability, we have taken mainstream approaches and adapted them accordingly for the developmental level of the group of patients we work with. No particular model is used to the exclusion of any others, but the themes from attachment theory and transactional analysis are prevalent in the units. Patients have individualised therapies, and a psychotherapy group was developed and established by the ward manager and the unit psychologist. This group began life as a deliberate self-harm group and was made up of all of the original admissions except one who chose not to take part. Although the initial focus of the group was on deliberate self-harm, it soon became apparent that the women all had a good grasp of why they harmed themselves and the scope of the group widened. The group runs for one and a half hours a week, and both ward manager and unit psychologist are present to facilitate. Strict boundaries govern the group, and subject matter discussed there is not taken out into the domestic setting. The going can be tough, but this group approach has provided the right sort of opportunity for the women to explore the impact of their past experiences.

The problems

As with any new service there were various teething problems’, side by side with all the usual problems that working in an NHS setting brings. The units themselves were not purpose-built for this group of patients, and unanticipated modifications were required in the early days to ensure an acceptable level of safety. Although patients are thoroughly riskassessed before admission, they could still come up with new and imaginative ways of using the environment to self-harm. As the concept of specialised services for learning-disabled women offenders is a relatively new one, most staff had little if any experience of this sort of challenging behaviour, and the learning curve was steep. Knowing theoretically that splitting occurs or that these patients somatise a great deal is one thing; dealing with it practically day in, day out is another matter. To have the ability to manage these issues successfully and consistently can only come with experience and from knowing each patient well. In order to have any hope of achieving a measure of success it is necessary to have consistent staffing.

We have been fortunate in having a hard core of dedicated staff who have been with the unit since it opened and have remained in the service. However, as in all NHS settings, having enough staff to fill all the available posts has been a problem. The service is not atypical in this respect, and has had to compete with the services on the rest of the site to obtain its staff complement. Because of the policy that new staff are rotated through various services on site, some staff who would not have chosen to be involved with such patients are asked to work on the wards, and staff who initially thought they would like to try this work have found that it is not for them. In any service, both in the NHS and in the private sector, rapid turnover of staff is a problem and countertherapeutic, particularly in settings such as ours which have attracted a label of being difficult environments in which to work. We have attempted as far as possible to avoid the use of bank or agency staff, although this has not always been possible. The lack of speech therapy input has been a major concern for the treatment of some of the patients, and general on-site services have had to be used in the absence of a dedicated person familiar with the communication problems which our service encounters.

Day services which are available were set up in the days when the site fulfilled the role of long-stay accommodation rather than the more acute/mediumterm establishment. A wider range of more appropriate activities would be appreciated. A newer thorn in the flesh of the forensic service as a whole has materialised, in the form of the local (and occasionally national) press. There has long been a baseline interest in the forensic services which originally existed but, since the building of many new homes in what was the hospital grounds, local opposition to the services has intensified. Public relations meetings have been held, but any incident is eagerly seized by the press and further adds to the negative image already present for this stigmatised group of patients. There have been incidents of abuse towards patients going about their business in the grounds, and nuisance from youngsters getting up to mischief on the site has already begun to cause more problems.

The future

We are currently in a situation where the two units are up and running, but there still remain patients on waiting lists and referrals continue to come in. There is a continuing need for these services, and rather than seeing the opening of newer units as a threat, we have seen it as an opportunity for mutual support and idea-sharing. Despite this slowing of the referral rate, we feel that there is still room for expansion in our service for accommodating those on the waiting list, having the capacity to take on new acute referrals and for those already in the service to move on into less secure areas to continue their rehabilitation. We have just secured the use of the unit adjacent to the second unit, and are interviewing for staff. It is envisaged that, once the unit is opened, staff from the first unit will move across temporarily, as we did with the second unit opened, to ensure initial continuity of staff for those patients who are transferred and to help establish a similar ethos in the unit. The third unit will act as a ‘step down’ from the previous two in terms of security and the level of responsibility we expect the patients to take.

We have not yet discharged any of the patients back into the community, although some are getting very close to being ready for discharge, and we hope that this third unit will be a final step on the way. This last hurdle of discharge will be difficult. The resources available in the community have already proved unsuitable for some of the patients, and providers are reluctant to take on people with forensic histories, especially those known for setting fires and violent behaviour. Until the right kind of supportive accommodation can be provided outside the hospital setting, where the patients can maintain the improvements they have achieved in hospital, the hope of resettlement will remain remote for at least some of them. For the few who will require much longer-term hospital-based provision we will need to ensure a decent quality of life.


As services for women offenders develop and expand and experiences are shared, we hope that we will gain a clearer understanding of the needs of this previously neglected group of patients and be in a better position to use effective therapeutic interventions and develop further appropriate services. This group of patients are certainly challenging, but the rewards have been great.


Adshead G (1994) Damage, trauma and violence in a sample of women referred to a forensic service. Behavioral Science and Law 12 235-49.

Adshead G (1998) Psychiatric staff as attachment figures: understanding management problems in psychiatric services in the light of attachment theory. British Journal of Psychiatry 172 64-69.

Alexander RT, Piachaud J, Odebiyi L & Gangadharan SK (2002) Referrals to a forensic service in the psychiatry of learning disability. British Journal of Forensic Practiced (2) 24-33.

Barley WD, Buie SE, Peterson EW et al (1993) Development of an mpatient cognitive-behavioural treatment programme for borderline personality disorder. ]ournal of Personality Disorders 7 232-41.

Bland J, Mezey G & Dolan B (1999) Special women, special needs: a descriptive study of female special hospital prisoners. Journal of Forensic Psychiatry 10 (1) 34-5.

Chengappa KNR, Ebling T, Kang JS, Levine J & Parepally H (1999) Clozapine reduces severe self-mutilation and aggression in psychotic patients with borderline personality disorder. Journal of Clinical Psychiatry 60 (7) 477-84.

Coid J, Kahtan N, Gaul S & Jarman B (2000a) Women admitted to secure psychiatric services: I Comparison of women and men. Journal of Forensic Psychiatry 11 (2).

Coid J, Kahtan N, Gault S & Jarman B (200Ob) Women admitted to secure psychiatric services: II Identification of catagories using factor analysis. Journal of Forensic Psychiatry 11 (2) 296-315.

Dell S, Robertson G, James K & Grounds A (1993) Remands and psychiatric assessments in Hollo way prison II: the non-psychotic population. British Journal of Psychiatry 163 640-44.

Department of Health & Home Office (1992) Review of Health and Social Services for Mentally Disordered Offenders and Others Requiring Similar Services (The Reed Report). London: HMSO.

Department of Health and Social Services (1974) Revised Report of the Working Party on Security in NHS Psychiatric Hospitals (The Glancy Report). London: HMSO.

Dolan BM, Warren FM, Menzies D & Norton K (1996) Cost-offset following specialist treatment of severe personality disorders. Psychiatric Bulletin 20 413-7.

Frankenberg FR & Zanarinin MC (1993) Clozapine treatment of borderline patients: a preliminary study. Comprehensive Psychiatry 34 (6) 402-3.

Hassell Y & Bartlett A (2001) The changing climate for women patients in medium secure psychiatric units. Psychiatric Bulletin15 340-2.

Kaye C (1998) Hallmarks of a secure psychiatric service. Psychiatric Bulletin 22 137-9.

Maden A, Swinton M & Gunn J (1994) Psychiatric disorder in women serving a prison sentence. British Journal of Psychiatry 164 44-55.

Parsons S, Walker L & Grubin D (2001) Prevalence of mental disorder in female remand prisons. Journal of Forensic Psychiatry 12 (1) 194-202.

Puti PK, Lehk SK & Treasden IH (2000) A comparison of patients admitted to two medium secure units, one for those of normal intelligence and one for those with learning disability. International Journal of Clinical Practice 54 (5) 300-5.

Roy A, Matthews H, Clifford P, Fowler V & Martin D (2002) Health of the Nation Outcome Scales for people with learning disabilities. British Journal of Psychiatry 180 61-6.

Royal College of Psychiatrists (1980) Secure Facilities for Psychiatric Patients: A Comprehensive Policy. London: Royal College of Psychiatrists.

Sansom D & Cumella S (1995) One hundred admissions to a regional secure unit for people with learning disability. Journal of Forensic Psychiatry 6 (2) 267-76.

Sequeira H & Hollins S (2003) Clinical effects of sexual abuse on people with learning disability: critical literature review. British Journal of Psychiatry182 (1) 13-19.

Thomas C, Bartlett AEA & Mezey GC (1995) The extent and effect of violence amongst psychiatric mpatients. Psychiatric Bulletin 19 600-4.

Thomson LDG, Bogue JP Humphreys MS & Johnstone EC (2001) A survey of female patients in high security psychiatric care in Scotland. Criminal Behaviour & Mental Health 11(2) 86-93.

Winston AP (2000) Recent developments in borderline personality disorder. Advances in Psychiatric Treatment 6 211-8.

Woods P & Mason T (1998) Mental impairment and admission to a special hospital. British Journal of Developmental Disabilities 87 119-31.

Elizabeth Berber


Harm Boer


Address for correspondence

Harm Boer, Consultant Forensic Psychiatrist, Janet Shaw Clinic, Brooklands, Brooklands Way, Birmingham, B37 7HL.

Copyright Pavilion Publishing (Brighton) Ltd. Dec 2004

Provided by ProQuest Information and Learning Company. All rights Reserved