Do women need special secure services?

Clinicians working with women patients in secure units will already know that they are the centre of a debate not of their own making, which is about the appropriate specification of services for women patients with security needs. This paper attempts to outline the relevant issues and proposed solutions.

Women and psychiatry

Gender differences in psychopathology and the resultant treatment of patients have been discussed for the past 20 years from medical, historical and anthropological perspectives (Showalter, 1987). Some of this discussion has been critical of psychiatric practice. Most recently, Seeman (1995) concluded that there are real and essential gender differences in patients' expression of psychopathology. She summarises:

“psychiatric diseases are inherently multidetermined but […] biological sex (genetics, anatomy, hormones) and psychosocial gender (assigned and adopted roles within family, political and economic structures) taken together are two powerful determinants. The facets of psychiatric disease that appear to differ between the two sexes are:

1. • the prevalence of certain syndromes (some found more commonly in men, others in women)

2. • the age at onset of certain syndromes (some begin earlier in one sex than in the other)

3. • the character and diversity of symptoms (sometimes identical in both sexes, sometimes not)

4. • the course and severity of illness (sometimes more progressive and more lethal in one sex than in the other)

5. • the response to existing interventions (sometimes particular to one sex or the other)

6. • the known risk factors (often distinct in women and men).” (p. 379)

The debate has led to a discussion of the services that women receive in general practitioner (GP) surgeries, psychiatric hospitals, secure psychiatric hospitals, learning disability units and prisons. Despite this, mainstream psychiatry has ignored the critiques of both contemporary and historical practice (Berrios & Porter, 1995; Gelder et al, 1996). The validity of psychiatric categories applied differently to men and women goes substantially unchallenged.

Treatment approaches

Women patients are seen in high numbers in general psychiatric hospitals, where there is a 3:1 female to male ratio (Gomberg, 1995). Women report that they are prescribed and use more psychoactive drugs than men, which seems to relate to the fact that women are more likely to report subjective distress while men exhibit behavioural manifestations of disorder. It is unsurprising then that women are generally diagnosed more frequently with neuroses and depression, and men with antisocial personality disorder and alcohol misuse. The current practice of prescribing psychoactive drugs for neuroses and depression would therefore lead to women being prescribed more medication than men (Gomberg, 1995). This raises an interesting question about the social and biological bases of different patterns of presentation, as well as the validity of such psychiatric categorisation in theory and practice.

If women generally present with depression and neuroses, they are more likely to be seen in GP surgeries and general psychiatric hospitals. Men presenting with antisocial personality disorder and alcohol misuse are more likely to be seen in forensic psychiatric hospitals, substance misuse facilities and prisons. But what happens to those women who do not fall within the parameters of the general psychiatric population? These are the women who are seen in secure facilities.

Women in secure facilities

Classification and placement

The total number of women in any type of secure setting (i.e. criminal justice or mental health) is small in absolute terms. Special Hospitals Service Authority (1995) figures for all types of secure psychiatric facilities indicated that in total there were only 1085 women requiring secure psychiatric care, 20% of the total population requiring such care. These were distributed in the following way: 539 in low secure psychiatric hospitals, 89 in medium secure psychiatric hospitals, 255 in high secure psychiatric hospitals and 202 in prison. The ratio of men to women in different types of secure facilities varies, but women always constitute a minority. Although the ratio of men to women admitted to hospital under Part II, Sections 2, 3 and 4 of the Mental Health Act (MHA) 1983 is roughly equal, women constitute less than a tenth of those currently put on restriction orders (Section 37/41) or transferred to hospital as sentenced prisoners (Section 47/49) (Department of Health, 1998). In both instances this reflects the small number of women convicted of serious offences. Similarly, among individuals sent from the courts to hospital for treatment without restriction (Section 37), women still form only 13% of the total. The ratio of women to men in special hospitals has been 1:5 for many years, and has been maintained in admission ratios (Jamieson et al, 2000). The female to male ratio for medium secure units varies from 1:4 to 1:7 (Higgo & Shetty, 1991; Milne et al, 1995; Murray, 1996). Figures for low secure hospital services indicate a 1:2 female to male low secure provision ratio (Special Hospitals Service Authority, 1995).

Concern about the current pattern of distribution of women in different secure placements has been supported by several major pieces of research (Dell et al, 1993a , b ; Maden et al, 1993; Maden, 1996). There are two main issues. First, both men and women are unnecessarily detained in high security, and this is true for a higher proportion of women in special hospitals than men. Second, categorisation of men and women under Section 1 (Types of mental disorder) of the MHA 1983 differs. Women are more likely than men to be detained in high secure hospitals under the legal category of ‘psychopathic disorder’, and men are more likely than women to be detained under ‘mental illness’. The high secure figures have led to a challenge of psychiatric orthodoxy in the forensic arena. As elsewhere in psychiatry, many clinicians would suggest that a ‘gender-blind’ diagnostic approach is both useful and used. However, commentators have argued that the significance of the psychopathic disorder label applied to women is that it constitutes the medicalisation of antisocial behaviour that in men would be criminalised (Special Hospitals Service Authority, 1995; Hemingway, 1996; Bland et al, 1999; Lart et al, 1999; Women in Secure Hospitals, 1999). The large number of men with personality disorders in prison (Fryers et al, 1998) lends support to this argument. The relative absence of serious convictions among women in high secure care also points to women being considered differently (Bland et al, 1999).

From both the crude markers of the MHA 1983 and patterns of placement, it appears that women and men are construed differently by the forensic world. It is less clear from this information that when hospitalised they have intrinsically different therapeutic needs and if so what these are.

Clinical characteristics of forensic women

Studies suggest that women are described as ‘mad’ rather than ‘bad’ (Department of Health & Home Office, 1994; Special Hospitals Service Authority, 1995; Hemingway, 1996; Bland et al, 1999; Lart et al, 1999; Women in Secure Hospitals, 1999). Despite this, forensic psychiatry has been justly criticised for the absence of research on treatment interventions (Bartlett, 1993). In the case of forensic female patients, there had until recently been almost no clinical information on which to base treatment strategy and in particular to identify specific unmet treatment needs. Phenomenological psychiatry has dominated practice for many years and there has been a corresponding lack of enthusiasm for treating the category of personality disorder by comparison with that of mental illness. For the disproportionate number of women detained in ostensibly therapeutic environments it could be argued that this has had a discriminatory effect.

Women are often admitted to secure services because of damage to property, self-harm or aggression towards hospital staff (Women in Secure Hospitals, 1999). Bland et al (1999) reported that the female patients in Broadmoor Hospital had a forensic history of assault (79.3%), arson (47.1%), theft (37%) and murder (21%). The same group of women exhibited self-harming behaviours, including self-injury (84%), alcohol misuse (38%), drug misuse (37%), eating disorders (17%) and sexual disinhibition (17%). Ten women in Holloway Prison reported a high incidence of arson, self-harm and substance misuse (Gorsuch, 1999). These findings were replicated to a lesser extent in a group of nine women with a dual diagnosis of mental illness and learning disability. These women, who were residents on a mixed-gender unit at the time of the survey, reported a history of violent and abusive relationships and some self-harming behaviour (Namdarkhan, 1995). Maden's (1996) larger comparative study of male and female sentenced prisoners found higher rates of self-harm, substance misuse, personality disorder and neuroses in women than men (Box 1).

Box 1. Characteristics of female and male patients in high secure psychiatric hospitals

Female patients are more likely than male to:

1. • be detained under Part II of the Mental Health Act 1983 as civil patients

2. • be classified as having a personality disorder and meet the diagnostic criteria for borderline personality disorder

3. • have an index offence of arson

4. • be admitted because of suicidal or self-harming behaviour, aggression towards hospital staff or damage to property

Male patients are more likely than female to:

1. • be detained under a Mental Health Act 1983 Restriction Order

2. • be classified as having a mental illness

3. • have committed homicide

4. • have a prior offending record

5. • be admitted because of their sexual behaviour or the symptoms of mental illness

Service planning and treatment needs: evidence-based medicine?

In the absence of work comparable to the large epidemiological studies of prisoner populations, service planning within hospital units at the level of treatment modalities is dependent on small studies and clinical impression. Bland et al (1999), Women in Secure Hospitals (1999) and Lart et al (1999) have emphasised the clinical significance of women's histories of sexual and physical abuse. Heads et al (1997) found that even in a population of patients with schizophrenia and mental illness, the incidence of histories of sexual and physical abuse is higher for women than men. A small study of 10 highly disturbed women on the psychiatric wing of Holloway Prison found that 90% of them had a primary diagnosis of personality disorder and a history of childhood sexual and physical abuse (Gorsuch, 1999). Similarly, 85% of the female patients at Ashworth Hospital classified as having borderline personality disorder reported some history of childhood sexual abuse (Potier, 1993). Evidence suggests a relationship between the diagnosis of personality disorder and a history of childhood sexual and physical abuse that warrants further investigation.

These elements of patients' life histories, which plausibly contribute to clinical presentation, should be considered more frequently in the development of treatment protocols. Male and female patients with similar early experiences may exhibit different behavioural problems as adults. Clinicians ignore the implication of events in patients' life history at their peril. A medical model, with its reliance on pharmacological intervention, may be unhelpful in dealing with behavioural problems of complex aetiology. In this regard it is interesting that in a survey of female patients (n=87) at Broadmoor Hospital, over 90% were prescribed both antipsychotics and antidepressants, while only 32% had received any formal psychotherapy (Bland et al, 1999). At the same time it is anecdotally established that some women detained under the legal category of psychopathic disorder are subject to episodic descent into psychosis. In practice the potential interaction of behavioural problems, conflicts from early life and episodic psychosis complicates clinical management.

Vulnerability and risk

The differential assessment of self-harming behaviour and suicidal behaviour can be difficult, as the presenting symptomatology is often similar – cutting, overdosing, risk-taking. Current practice is focused on the pharmacological management of behaviour, but the concept of self-harm is changing. New treatment strategies are evolving based on psychological frameworks that propose that to treat self-harming behaviour, we must first understand the life history of patients and the resulting motivation for their behaviour (Gorsuch, 1999; Smith Benjamin, 1999). Childhood trauma contributes to self-destructive behaviour, and lack of secure attachments helps to maintain it. At times of distress and when feeling a lack of control, women patients may engage in self-harm in an attempt to “appease [their] internalised attacker” a result of old abusive familial interpersonal rules (Smith Benjamin, 1999: p. 123).

For mental health professionals it can be frustrating and tiring to work with patients who exhibit self-harming behaviour. The patients are often labelled as having a personality disorder or being difficult to manage, and are relegated to the ‘untreatable’ group. Focusing primarily on managing behaviours contributes to the sense of hopelessness in treating people who practise self-harm (Smith Benjamin, 1999). Instead, staff need to focus on patients' interpersonal interactions and the themes of powerlessness, rejection and feeling misunderstood that are prevalent in women patients' life histories. Management of patients may be problematic, even where there is a therapeutic focus on resolving conflicts from early life. Even enthusiastic therapists can get it wrong. Pushing therapy too fast can exacerbate feelings of powerlessness, which resonate with early life events. Women may find themselves unable to cope with therapy and unable to tell their therapist. Cutting, overdosing or slipping into psychosis can be the only avenues available. It is only as women begin to feel understood that they can progress, and start to find productive ways of expressing their distress instead of self-harm. The changing concept of self-harming behaviour means that there are no definitive answers in its assessment and treatment. However, it is clear that the management of risk behaviour must be different from management of suicidal behaviour. Staff must be trained to be aware of and respond to these differences.

Whereas the focus for women patients is primarily on harm to self, for male patients in secure settings it is primarily on assessing and managing potential harm to others. It is important to recognise that women and men patients present different risks. Clinical experience shows that female patients on discharge often present no risk to others, or occasionally some risk to children, whereas male patients present a high risk for attacking women. The potential behaviour of patients is important in determining the appropriate location, level or type of security, and treatment. It is therefore important for both general and forensic psychiatrists to be aware of the differences between women and men patients and to be cautious of homogenising patients either across or within groups.

The secure units in which women patients live are not all equipped to provide a safe and private environment, and the safety of women diminishes in units where there are few female patients, as these often have no separate male and female facilities (Mental Health Act Commission, 1999). The women often have low self-esteem, poor assertiveness skills and few interpersonal resources to help them cope with a large male population. Their experiences of childhood abuse make them particularly vulnerable to a group of men, many of whom are known to have exhibited violent and dangerous behaviour towards women (Bland et al, 1999). The culture of normalisation may encourage emotional and sexual attachments between patients that are ultimately damaging to the women concerned. Women from ethnic minority groups often find themselves doubly isolated. Cultural differences are at times compounded by linguistic isolation and unwanted close proximity with men, which violates religious beliefs.

In addition to needs that stem from ‘abnormality’ (personality disorder and mental illness for the most part), it is important not to forget ordinary aspects of women's experiences. In secure hospitals, these elements of identity (being someone's daughter, mother, sister, work colleague or friend) may not be recognised, and with the passage of time may wither and die. These parts of the self can be caught up in pathological attachments and sometimes in violence, no doubt just as they can be for men similarly detained. But nurturing healthy parts of our patients is as important as dealing with the unhealthy part. Some of these aspects of self are highly gender specific and should be considered as such in our assessments (Box 2).

Box 2. The ‘special’ experience

“A couple of months later I was called into the office by Dr. L who said he'd heard that I'd not been going over the male side. I told the doctor I wasn't interested and of course it was men that put me in there so I wasn't interested. But going over the male side is supposed to mean you're normal. He said I'd have to try and be interested because if I didn't go over the male side, the next time that he did his ward rounds he'd have to write to the Home Office and tell them I wasn't co-operating… I had no choice. So the next time there was a disco I went over to the male side. I was crying my eyes out but one of the nurses I got on with said she'd come with me and she did and I sat there and all these men are coming at me, asking me to dance and I kept saying “no”. I thought “this is good enough isn't, I've come haven't I?” Then the nurse came up to me and said she had to do a report when she got back, so could I at least have one dance then she could put it in her report. So I danced and I said “you touch me where you're not supposed to and I'll…” but then I thought even if he did do something I couldn't do anything. They could whip me away into the punishment block if I objected, tell me this was just the way they were. I was in a no-win situation.” (Brown, 1996: p. 43)

Current services

Separate wards for male and female patients were the historic norm, but in the 1970s there was a trend towards mixed wards in all psychiatric hospitals (Namdarkhan, 1995; Warner & Ford, 1998). Within forensic hospital facilities, special hospitals have moved only very slowly toward integration. The more recent and increasingly numerous medium secure units have been mixed gender. In addition to the debate about treatment interventions targeted at women's needs, there is the issue of suitable environments in which to deliver care to women. This can be considered in terms of physical facilities and staff–patient composition.

Research into the physical environment of wards has mainly been done in non-forensic settings. A national audit of 263 adult acute admission wards and 33 adult intensive care wards in 118 trusts across England and Wales found that 94% of the wards were mixed, 65% of the female patients did not have access to women-only sleeping areas, 33% had access only to bathrooms used by both male and female patients and 3% of female patients used sleeping areas also used by male patients, in which their beds were only enclosed by a curtain (Warner & Ford, 1998).

In 1992, an investigation of Ashworth Hospital (Department of Health, 1992) noted that “the environment and culture at the hospital has been especially insensitive to [the women's] needs” (p. 229). The committee found that the high incidence of female patients' self-harming behaviour was met with derogatory comments from staff, and the gender-blind approach of the staff led to acceptance of close and intimate expressions of physical affection between female patients and male staff. The committee concluded that the current regime for female patients at Ashworth is “infantilising, demeaning and anti-therapeutic” (p. 232).

Kaye (1998) and the Special Hospitals Service Authority (1995) echo this dissatisfaction with psychiatric services provided for female patients. They add that, paradoxically, despite the service deficiencies, the cost of caring for female patients in high secure psychiatric hospitals is higher than for male patients. During the financial period 1995–1996, the average annual cost for a female patient was £102 000, compared with £80 000 for a male patient (Kaye, 1998).

Medium secure units have not been subject to the same level of scrutiny as special hospitals. It is hard to know the extent to which findings in other parts of the health service would generalise to them. It is possible that in medium secure units the situation for women is at its worst. In small units, women may have no way of avoiding sexualised social contact with men who have a history of sexual and physical violence towards women. Women who have previously experienced such violence can be ill prepared to deal with the threat of similar behaviour in the unit. Even on acute psychiatric wards sexual harassment and sexual assault is not uncommon and may not be recognised by staff groups unused to identifying such phenomena (Thomas et al, 1995; Warner & Ford, 1998). The reluctance of some medium secure units to deal with women patients because of unsafe environments and the historically well-established reluctance of medium secure units to deal with patients with personality disorder who stay too long may create a further difficulty. Women patients may simply bypass medium security and go straight to the special hospitals.

The response

The past 10 years have seen an increased awareness in the need to address some of the problems with secure psychiatric services provided for female patients (Box 3; Mental Health Act Commission, 1999). The Special Hospitals Service Authority spent a significant proportion of £15 to 20 million per annum over a 6-year period on improving facilities for female patients (Kaye, 1998), and developed a strategy for women requiring secure psychiatric services (Special Hospitals Service Authority, 1995). This strategy outlined a range of service requirements (environmental, treatment, staffing and the establishment of local links for individual women).

Box 3. Four main areas of dissatisfaction with current secure psychiatric services for female patients

Female patients do not have adequate privacy in the units

Vulnerable female patients are placed in potentially dangerous situations

The regimes of care do not reflect society's current values regarding women's roles

Current services are expensive and their efficacy under-researched

A national charity, WISH (Women in Secure Hospitals), works with and on behalf of women patients who are detained in secure settings. WISH has been active in campaigning for the rights of such women and has been instrumental in the heightened awareness of their concerns in the Department of Health. Mind provides a more general advocacy service to the women in hospitals. The Department of Health has funded a project to develop a national educational programme of gender awareness training. In the interim, some National Health Service (NHS) and private secure psychiatric hospitals have begun to develop and reassess their equal opportunities and sexual harassment policies. Staff have formed groups with special interests in women's issues and together are working to improve the services that they provide for female patients in secure psychiatric hospitals. In 1997, the Government issued a policy stating that by the year 2002, 95% of health authorities should no longer have mixed-gender psychiatric units. Several new units have opened within the past year that provide services only for women patients in medium secure units, and within mixed medium secure units there has been a shift towards opening at least one ward that will accommodate only women patients. The advantages and disadvantages of such restructuring are currently being studied.

Despite the increased awareness of the problems of women in secure psychiatric services, the funds and resources allocated to improvements are still inadequate. It was hoped that the Department of Health's recent investigation into NHS policy guidelines would lead to the development of an NHS policy on gender issues. However, the results of the study do not recommend a separate policy on gender issues. Instead, it has been suggested that there is adequate gender-based information in the general NHS framework to provide a set of standards for women patients in secure psychiatric services. For those who supported the idea of a specific NHS policy on gender issues, this has been a great blow. The concern now is whether or not regional NHS trusts will allocate sufficient resources to the provision of improved services for women in secure psychiatric services when there is no clear mandate for them to do so.

Conclusion

In summary, despite gaps in evidence, there is a good case to be made that women in secure facilities may present different clinical challenges than men. Although there is value in the debate questioning whether such women should be in these facilities in the first place, the immediate problem is how to deal with the existing population. Unlike many psychiatric populations, the enduring characteristic of forensic populations is that they do not go anywhere fast. Treatment based on individual need and a holistic approach to care and informed by contemporary discourse on gender is a good place to start – provided that it can be made meaningful to the multi-disciplinary and multi-agency group of staff involved in women's care. Rather than being marginalised by the term ‘special’, treatment should simply be both ‘specific’ and ‘sensitive’. Although the research terms are useful rhetorically, research evidence is thin on the ground. Diagnostic or behavioural labels are insufficient material on which to base comprehensive interventions. Many staff grasp the contrasting profiles of male and female patients (see Box 1) but find it difficult to tailor ‘specific’ treatments to problems. Part of the difficulty is in analysing how the environment in which a female patient finds herself can resonate with earlier experiences. Hierarchical environments of a kind normally found in secure settings feed into established feelings of powerlessness and exacerbate current illness behaviour. Sensitivity to the details of women's lives and their relationship to the present and future seems an essential prerequisite to adequate service delivery at whatever tier of security. This will be easier if senior staff support such an approach. And if women are not adequately represented among senior staff it seems unlikely, on the evidence from recent hospital history, that anything will change.

Box 4. Key conclusions

Sensible service delivery to and for women both in prisons and in mental health services requires a more sophisticated grasp of gender issues than is current

An understanding of gender issues can complement and enhance what are otherwise inadequate psychiatric and psychological models

This will enable better identification of specific needs in a population of women for whom power, violence and gender have been historically connected

Both the users of the services and the front-line staff are integral to the creation and monitoring of such services

Two environmental options intended to avoid or reduce the problems cited above are being explored. Single-gender medium secure units have been and are being developed, and in mixed-gender secure units consideration is being given as to how facilities, therapeutic regimes and timetables can be altered to provide safe and empowering environments. The extent to which such units will materialise, and how they will run, is unclear. Their emergence now both reflects and requires that staff attitudes should come under scrutiny. Professionals bring to work ideas about gender, as they do about many other things. Part of an empowering therapeutic regime for women in secure care is the staff's ability to reflect on their own attitudes. Only in this way will the wholesale imposition of staff views on vulnerable women be avoided. It seems unavoidable, however, that the institutional hospital regimes will, by their very nature, to some extent echo the coercion of women's early lives. This is best acknowledged.

But the real test of change is not theoretical argument. Evaluation of therapeutic goals and outcomes is necessary. Research into staff training, therapy focused on what appear to be specific women's needs and adjustments to the therapeutic environments designed to make them comfortable for women are essential. Randomised control trials of interventions are seldom appropriate for potentially dangerous individuals. None the less, valuable research can and must be done to see if any improvement would result from the implementation of current thinking on the needs for specific and sensitive, although not special, services for women.

Multiple choice questions

1. 1. In special hospitals:

   1. a the ratio of men to women is 15:1

   2. b the ratio of men to women is 2:1

   3. c the ratio of men to women is 1:5

   4. d men and women are equally likely to be admitted under the MHA 1983 category of mental illness.

2. 2. Men and women detained in hospital under the MHA 1983 are equally likely to be detained under:

   1. a Section 2

   2. b Section 37

   3. c Section 37/41

   4. d Section 47/49

3. 3. With regard to psychiatric wards:

   1. a most high secure hospital wards are mixed

   2. b most medium secure hospital wards are mixed

   3. c most female patients have access to bathroom/ toilet areas used only by women

   4. d most female in-patients have separate women- only sleeping areas.

4. 4. In high secure hospitals:

   1. a the cost of keeping male patients is more than that for female patients

   2. b the annual cost of keeping a female in-patient is between £50 000 and £75 000

   3. c the annual cost of keeping a female in-patient is between £75 000 and £105 000

   4. d the total number of women currently detained there and thought to require high secure psychiatric care is more than 300.

MCQ answers

1		2		3		4
a	F	a	T	a	F	a	F
b	F	b	F	b	T	b	F
c	F	c	F	c	F	c	T
d	F	d	F	d	F	d	F