15 Women in forensic institutions

Women are always in a minority in forensic populations, whether in prisons or secure hospitals. As a consequence neither the institutions nor policymakers seem to know how to respond to them. The earliest response was to ignore them; women were an afterthought whose needs were overshadowed by the male offender population that dominated the landscape. At other times there was an overemphasis on psychopathology, reaching a peak when the United Kingdom (UK) government in the 1970s proposed rebuilding its main women’s prison, Holloway, as a secure hospital. There was a tendency to see all the deviant behaviour of women as evidence of psychopathology, leading to Sim’s charge that doctors located them ‘at the centre of the professional gaze’ (Sim 1990) when they came to examine and to medicalize prisoners.

Over the last decade policy has become more rational. The main advance has been to recognize that the mental health of offenders is continuous with the mental health of all women. Most women spend only brief periods in custody and it is unrealistic to imagine that either their health or broader social needs can be addressed in isolation from their lives outside the walls of the institution.

The new approach is neatly summarized in the title of the UK government’s main policy document in this area: Women’s Mental Health: Into the Mainstream (Department of Health 2002). As part of the general policy of social inclusion and reducing inequality, women are to be brought in from the margins. Awareness of gender differences should go along with acceptance of the principle of equality of access to healthcare. In practice these goals are to be achieved through an emphasis on individual needs assessment with services designed to meet the needs that emerge from the assessment.

Too often in the past the service has come first. Women have been squashed and squeezed into forensic services designed primarily for men. We have known for many years that ‘women require different treatment and facilities to men if they are to have the opportunity to break the patterns of behaviour which have lead to their contact with the criminal justice system.’ (Department of Health 2006a, p. 5). Now there is an opportunity to do something about the problem and we are beginning to see the first examples of services designed to meet women’s needs. The rest of this chapter will consider what we know about women in forensic settings and how best we can develop responsive services.

Women are generally less antisocial than men (Moffit et al. 2001). They are convicted of fewer crimes and the difference is particularly marked for violent or sexual crimes (Ministry of Justice 2007a). Women are more likely to be the victims rather than the perpetrators of violent crime, and 38% of violent crime is committed against women (Home Office 2007, p. 83).

In 2006, men were convicted or cautioned for 407 100 indictable offences compared to 99 900 for women, a ratio of 4:1. Theft and handling offences are the most common, see Table 15.1.

There are qualitative differences between the offending of men and women. Women are more likely to be involved in domestic violence, less likely to commit violence against strangers or acquaintances (Home Office 2007, p. 83).

Prospective longitudinal studies show that women’s antisocial behaviour usually emerges in adolescence but fluctuates more than men’s antisocial behaviour, mainly according to circumstances. Typical antisocial behaviour in young women is greatly influenced by social factors and male peers are a particularly important—and usually negative—influence (Moffit et al. 2001). The Cambridge study of delinquent development (Farrington 1994) showed that one of the main factors helping young men to stop offending was the stabilizing influence of a partner, and this recent work shows the influence sometimes operates in the opposite direction.

As in men, a history of conduct problems in adolescence is associated in women with several negative outcomes in adult life. Apart from adult antisocial behaviour they include relationship problems; depression; a tendency to self-harm and suicide; and poor physical health (Moffit et al. 2001). So whilst antisocial behaviour may be less common in young women there is nothing to suggest that it is any more benign. In fact there is evidence to suggest that the adult outcome may be worse as antisocial young women are such a small minority and may experience additional rejection because of their failure to conform to gender stereotypes.

In offenders with major mental disorder the balance between men and women is much less skewed. The additional offending risk conferred by having a major mental disorder is greater for women than for men and this narrows the gap in offending between the sexes (Hodgins et al. 1996). Even so, women account for only a minority of those in secure institutions for mentally disordered offenders. They have generally accounted for less than 20% of the population in high secure hospitals, and a greater proportion in this setting were judged to be misplaced in that they did not need to be in such high security (Lart et al. 1999).

The prison population in England and Wales is at a record high (Ministry of Justice 2008), having increased by 41% between 1996 and 2006 (Ministry of Justice 2007b). During the same period the female prison population increased by 94% (Ministry of Justice 2007b, p.92), admittedly from a very low baseline. The reasons are complex and relate partly to women receiving lengthy sentences for their involvement in drug smuggling.

These changes mean that while women remain less likely than men to be sent to prison they make up an increasing proportion of those in prison (currently 5.4%) (Ministry of Justice 2007c). The number of women in prison has almost doubled since 1996 but remains small at 4430. The percentage of women within the prison population is similar in other English-speaking countries (Bartlett 2007a).

The most common convictions for sentenced women in prison are acquisitive, followed by drug and then violent offences. For men, violent and acquisitive offences are equally common and outnumber drug offences. Sentenced women in prison are almost twice as likely as sentenced men to have been convicted of a drug offence and are less likely to have been convicted of a violent offence (although this gap has been narrowing). Table 15.2 shows the offences for which women in prison had been convicted, in 2007, and the comparison figures for men.

Women in prison are a disadvantaged group and often have complex and interdependent problems. There is a high prevalence of adverse childhood experiences (O’Brien et al. 2003; Department of Health 2006a), poor educational attainment (O’Brien et al. 2003), and work histories (Social Exclusion Unit 2002), and high rates of mental disorder and substance misuse (Maden et al. 1994; O’Brien et al. 2003). The experience of being in prison can compound these problems; women are separated from their children and social networks and they are within a hierarchal and potentially authoritarian institution where they are often victimized (O’Brien et al. 2003).

Forty-three per cent of women are living with their children immediately prior to imprisonment, about half of these as a lone parent, and relationships with partners frequently break down while women are in custody (O’Brien et al. 2003). Each year up to 17 000 children are separated from their mothers by imprisonment, and only 5% of them remain in their own family home after their mother’s imprisonment (Department of Health 2006a). A study at Holloway prison found that in women with mental disorder transferred to hospital, almost half of the women who had borne children were still their primary carers (Rutherford et al. 2004).

The Prison Service provides a limited number of places in mother and baby units which take babies up to the age of 18 months. However, the admission criteria tend to select out mothers with mental disorder. Within the units, mental disorder remains under-diagnosed and treated, potentially putting mothers and their children at risk (Birmingham et al. 2006a).

Women in prison receive more adjudications than men for all types of punishable offences, including violence (Ministry of Justice 2007b, p. 110). In part this may be the result of less serious breaches of the rules: comparison with male prisoners is difficult due to the differing nature of the regimes; women, for example, are afforded more physical freedom than men with a similar security classification (Maden 1996). Women who have had three or more punishments are more likely to have attempted suicide than those with only one or two, suggesting a possible link between women causing breaches of discipline and mental illness (Department of Health 2006a).

Self-harm and suicide in prison are of considerable public and institutional concern. Twenty-six per cent of sentenced female prisoners report having committed an act of self-harm or having made a suicide attempt in the last year (O’Brien et al. 2003). The rate of completed suicide in custody was higher for women than men between 1999 and 2004 but is now similar, at about 1 per 1000 population per year (Ministry of Justice 2007b, p. 114).

Mental disorder is common in prisons (Fazel and Danesh 2002) and female prisoners are particularly at risk (Maden et al. 1994; O’Brien et al. 2003) (see Table 15.3).

Women in prison suffer higher rates of mental disorder than women in the community and their male counterparts in prison. They are a particularly vulnerable group who suffer additional stresses within prison.

The prevalence figures tell only part of the story and interpretation is complicated. The debate has sometimes been conducted in simplistic terms, according to which all mentally disordered women should be removed from prison to hospital. This extreme position in the debate over medicalization of female offending is unsustainable in principle (drug addiction is not usually a reason for hospitalization so why should it become so in prison?) and for practical reasons (there are not enough beds).

The real challenge is to assess and meet the full range of healthcare need in prisoners, whether it be for transfer to hospital or some form of help in prison, but the task is not straightforward. It is relatively easy to count heads in prevalence surveys but there is no standardized measure of the need for treatment. As a result surveys rarely address this dimension of the problem, and when they do they use idiosyncratic estimates or self-report of previous treatment.

In addition to presenting with high rates of mental disorder and comorbidity (Bartlett 2007a), women are more likely than men to have had psychiatric treatment in the past and to be receiving treatment in prison (Maden 1996). This presumably reflects, in part, the increased needs of this population but may also reveal a willingness by professionals to respond to evidence of psychological distress in women and for the women themselves to have a more positive attitude towards treatment. In their survey of women prisoners, Maden et al. (1996) measured attitudes to treatment and found significantly more women wanted treatment. For every inmate given a diagnosis, they also made a recommendation for treatment: ranging from no treatment to transfer to hospital. No treatment was more likely to be recommended for men (odds ratio (OR) 0.4, 95% confidence interval (CI) 0.3–0.5), there was not a significant difference between the sexes in the number recommended for transfer to hospital (OR 1.6, 95% CI 0.8–3.0) but women were more likely to be recommended for outpatient treatment (OR 2.4, 95% CI 1.8–3.4).

Treatment within the prison system can take a number of forms, for example, addressing mental disorder, substance misuse, or offending behaviour. These areas of need are not independent of one another, although the services provided may be almost entirely independent despite being located within the same institution. Mental disorder is treated by healthcare, substance misuse by the CARAT service (Counselling, Assessment, Referral, Advice and Throughcare), and offending behaviour by the cognitive behavioural ‘treatments’ which are part of the Offending Behaviour Programmes Unit.

On 1 April 2003, the NHS formally took over the provision of healthcare within the prison service in England and Wales and government policy is based on the idea of providing equivalence of care between the community and prison (Department of Health 1999a). However, in many ways prison is not equivalent to the community: prisoners are a highly selected population with a number of characteristics complicating their treatment; they are living in a custodial environment and there is no equivalent in the community to a prison healthcare wing (Wilson 2004; Birmingham et al. 2006b).

Despite the high rates of mental disorder, few women are transferred out of prison to a hospital bed: less than 2% of new receptions in a study at HMP Holloway (Rutherford et al. 2004). The women who are accepted may face long waits before transfer, and those with personality disorder and those requiring higher levels of security wait even longer. Factors associated with difficulty obtaining an inpatient bed include: having a personality disorder; reporting a history of sexual and or physical abuse; being chronically self-harming; and having a history of substance misuse (Gorsuch 1999).

The CARAT service was established as the universal drug treatment service in every prison in England and Wales in 1999. They assess prisoners, give advice about drug misuse, and refer to appropriate drug services: clinical services, CARAT services, rehabilitation programmes, and therapeutic communities. These services are provided chiefly by external drug agencies, prison officers, and healthcare staff working in partnership. CARAT figures show that women are more likely to refer themselves, spend more on drug use, and are more likely to have had treatment before (May 2005).

Over the last few decades there has been an increase in the evidence base for cognitive-based interventions for offending behaviour (McGuire 1995, 2002). These programmes are generally delivered by forensic psychologists and prison staff. However, the majority of the evidence relates to men and while the programmes are often considered suitable for female prisoners, a recent evaluation of female prisoners showed no significant reduction in one- or two-year reconviction rates for those who had had treatment compared to a matched sample who had not (Home Office 2006). Male prisoners with personality disorder also have available to them a prison run on the lines of a therapeutic community; no such service exists for women.

The Dangerous Severe Personality Disorder Service, however, does have a facility for women: the Primrose Project with 10 beds. This new service intends to draw on the best practice from both the Prison Service and the National Health Service (NHS) to develop a hybrid model of health intervention for women who have care needs that extend beyond that which current mental health services can address, and for whom transfer to the NHS is not deemed appropriate (Department of Health 2006b).

The challenge is to bring all these services together in order to help women who have multiple and complex difficulties, which are not neatly subdivided to fit in with service provision; moreover many women are in prison only for a short period. A history of trauma is a good example of a complicating factor, one which is common, can impact on all these areas, and is not necessarily specifically addressed by any of these services. Perhaps inevitably it is the women with the most complex difficulties for whom the provision is often the poorest. Furthermore, continuity of service between the prison and community rarely exists.

The Corston Report recommends that custodial sentences should be reserved for women who commit serious and violent offences and pose a threat to the public, and that existing women’s prisons should be replaced with suitable, geographically dispersed, small, multi-functional custodial centres (Corston 2007). Were this to be implemented, it would mean a complete transformation of current provision.

The number of women in prison may be small, but the number of women in secure psychiatric services is even smaller and they are outnumbered by men at all levels of security. However, proportionally women are more likely to receive a psychiatric disposal for criminal behaviour than men (Maden 1996; Lart et al. 1999). As with the case of women in prison, the evidence base primarily describes the population and does not extend to effective treatments (Lart et al. 1999; Bartlett 2000, 2007b).

Compared to their male counterparts in secure forensic services, the women appear more ‘psychiatric’ and less ‘criminal’ (Bartlett 1993): they are more likely to be admitted as transfers from other hospitals, following non-criminalized behavioural disorder (for example, assaults which have not lead to prosecutions or self-harm), they have fewer previous criminal convictions, and more previous psychiatric admissions (Coid et al. 2000). They are also more likely to be admitted under the legal category of ‘Psychopathic Disorder’ and to have a history of arson (Coid et al. 2000). Like women in prison there is a high prevalence rate of adverse childhood experiences, including physical and sexual abuse, poor educational attainment, and work histories (Bland et al. 1999). Compared with women in prison, there are fewer mothers, and those that are mothers are unlikely to have an expectation of caring for their children on discharge (Bland et al. 1999). Self harm is also common (Lart et al. 1999): in a case note study at Broadmoor (Bland et al. 1999) only 16% of the women did not have a history of self-harm. Comorbid substance abuse/dependence is also common (Lart et al. 1999).

The higher prevalence of the category of ‘Psychopathic Disorder’ among the female detained population has raised the question of whether it constitutes a medicalization of antisocial behaviour which would be criminalized in men (Bartlett et al. 2001). Some have debated whether the psychiatric response to women is excessive, or that to men inadequate. In fact, if we return to our starting principles, the question is unlikely to lead to a useful answer. The crucial issue is whether individual patients get the needs assessment and the tailored treatment they require. Even a cursory glance at the women’s wards of high security hospitals in the 1980s showed that many of their patients were far from having their needs met but fortunately the world has changed since then.

Whilst women are generally considered less of a threat to the outside world there is general agreement that they present more of a challenge within secure hospitals. The small number of women in high and medium security account for a disproportionate number of assaults on staff and other patients, deliberate self-harm, and episodes of seclusion. Women are said to have less need for perimeter security and more need for internal or relational security.

As a consequence, women were central to the discussions in the 1990s about the future of high security hospitals. Many patients in high security were judged not to need all aspects of high security. They could be looked after safely in medium security if adaptations were made to meet their particular needs, for example, tolerance of a longer stay than the usual two-year ceiling.

A high proportion of these misplaced patients were women and their main specific needs were for high staffing ratios and skilled nursing care in medium security. Their treatment needs were not being adequately met at the appropriate level of physical security (Milne et al. 1995; Lart et al. 1999) and as a result many women suffered from the unnecessary constraints of being detained in high security.

All was not well in medium security either. Women, as a minority in high secure and medium secure services, lacked privacy and were vulnerable to victimization by male patients (some of whom had histories of sexual and physical violence towards women) (Bartlett et al. 2001).

After years of debate the late 1990s saw drastic action in these areas. Two of the three high security hospitals no longer admit women. The remaining high secure beds are now concentrated in a single hospital, Rampton, which takes advantage of the concentration of patients in one place to provide a more specialized services. It is no longer considered acceptable to try and fit women into a service primarily designed for men (Department of Health 1999b, 2002, 2003).

As the number of beds in high security has fallen, a new service development aims to provide a more appropriate service. Women’s Enhanced Medium Secure Services (WEMSS) provide a more accessible, locally-based service where the emphasis is on skilled nursing care and psychological therapies rather than perimeter security. There are currently three pilot WEMSS services, the largest of which is the Orchard in West London with 45 beds, it opened in 2007 (The Orchard – Enhanced Medium Secure Mental Health Unit at St Bernard’s Hospital, Ealing). These services have been designed to cater specifically to the needs of women and aim to provide a therapeutic environment that is able to safely contain and manage distress and disturbance, while providing appropriate therapeutic interventions (Green 2007).

Being for women only, WEMSS units are in line with a general trend in the NHS to separate inpatient facilities for men and women. Debate about this issue continues. The common ground is that separate services are needed but there is fierce argument about their extent. On the one hand it is argued that women who have been victimized throughout their lives need protection when at their most vulnerable. On the other it is accepted that such protection is necessary when mental health problems are at their most acute, but the task of recovery and rehabilitation entails coming to terms with the world through controlled exposure.

There is, of course, no simple answer and the debate is an example of the general tension in mental health between protection from stress and allowing recovery by attempting more ambitious tasks, always with the possibility of failure. There is little research in the area. One exception is a study that shows (unsurprisingly) that while there may be other benefits, single-sex units do not eliminate the problem of victimization (Mezey et al. 2005).

Perhaps unsurprisingly, medication is the most common form of explicit treatment over and above a secure hospital environment. In their study at Broadmoor, Bland et al. (1999) found while 72% of women had a diagnosis of mental illness, almost all women were receiving antipsychotics (97%), and most were also receiving antidepressants (91%). A third (32%) were receiving or had received formal psychotherapy. The literature review by Lart et al. (1999) concluded that there was unmet need for responses to personality disorder, substance dependence, and trauma therapy relating to histories of abuse. They made a number of recommendations including a shift towards relational security which would enable more individualized levels of security to develop in response to women’s needs. These recommendations have been incorporated into WEMSS, which provide ‘high levels of therapy in a non-oppressive environment, coupled with effective observation of patients’ (The Orchard – Enhanced Medium Secure Mental Health Unit at St Bernard’s Hospital, Ealing, p. 1).

Research regarding the outcomes after discharge from secure forensic units has tended to focus on reconvictions. Women are less likely to re-offend than men (Buchanan 1998; Coid et al. 2007), however, some or all of the gender differences in offending may be explained by the fact that women are more likely to harm themselves, less likely to have a history of alcohol and drug problems, and less likely to have a previous criminal history (Milne et al. 1995).

Women and men are different: this includes differences in the nature and prevalence of mental disorder and antisocial behaviour, but also differences in their lives, expectations, and responsibilities. This is captured by recent policy documents relating to gender and mental health and prison services (Department of Health 1999b, 2002, 2003, 2006a; Corston 2007). They conclude that gender needs to be integral to our services, not an afterthought; our services need to be holistic, and we need to listen to the women themselves in order to facilitate this. However putting this into practice is more difficult. The evidence base for effective treatment in this population is lacking (Lart et al. 1999; Bartlett 2000, 2007b), and we need to think about how we give disempowered women a meaningful voice within hierarchical and coercive systems.