A criminological and psychiatric survey of women serving a prison sentence.
The experiment failed and Holloway became notorious for high rates of self-mutilation and violence and the plight of a small number of mentally disordered women (Moorehead 1985). Doctors were no longer carrying out their gatekeeping function and there was confusion about the appropriate response to disturbed behaviour. Mentally disordered women were punished rather than treated or transferred to hospital,. whilst other women resented being treated as if they were not responsible for their own actions.
Holloway now operates more effectively, having reverted to the conventional separation of disciplinary and medical services that is found in male prisons. Carlen (1983) describes a similar, failed attempt to |psychiatrize' Cornton Vale in Scotland. With hindsight, the failure of these experiments is less surprising than the fact that the schemes were adopted in the first place. It is therefore important to examine both the empirical evidence and the theoretical assumptions that prompted these medical responses to female criminality.
The main empirical evidence is Gibbens's (1971) survey of every fourth reception into Holloway during 1967 (Heidensohn 1985: 73-% Dobash et al. 1986: 127-9). He reported |mental illness' in 15 per cent of sentenced women, 39 per cent of medical remands and 25 per cent of other remands; the proportions reporting inpatient psychiatric treatment in the last three years were 17, 26, and 18 per cent respectively. Gibbens concluded that women in prison have high rates of psychiatric disorder but, as Morris (1987: 56) points out, psychiatric morbidity in male prisoners is also high. A 1971 survey of sentenced men in the South East prison region (Gunn et al. 1978: 217) found a history of inpatient treatment in 12 per cent of men but 15 per cent were described as having a drink problem, compared to 7 per cent of Gibbens's sample. Whilst mental health was |a major problem' in 15 per cent of sentenced women, 34 per cent of sentenced men were identified as current psychiatric cases (Gunn et al. 1978: 218-20). These comparisons are of dubious validity because of methodological differences but they make the point that gender differences are not clear cut. There is a need for a study in which identical methods are applied to comparable samples of women and men in prison but, in the 25 years since Gibbens carried out his survey, this has not been done (the authors review psychiatric studies of female prisoners in more detail elsewhere (Maden et al. 1993)).
Theoretical support for the proposition that female offenders will have higher rates of psychiatric disorder is more convincing. |Normal' women have such a low rate of offending that mentally abnormal offenders are likely to account for a greater proportion of total offending (although mentally abnormal female offenders will still be outnumbered by similar men). Whether this will be reflected in the sentenced prison population depends on the filters at various stages of the criminal justice system which divert some mentally disordered offenders away from prison.
Mental disorders do appear to be associated with a higher proportion of female offending. West's (1965) comparison of 148 cases of murder followed by suicide and a control group of 148 other murder cases found that women accounted for 60 (41 per cent) of the murder-suicide group (many of whom were mentally disordered), compared to only 11.5 per cent of the murder group. In the former German Federal Republic, Hafner and Boker (1982) described all mentally disordered offenders who committed a crime of serious violence against the person in the period from January 1955 to December 1964. Women accounted for 23 per cent of the total of 533 offenders, in marked contrast to male:female ratios for all recorded violent crime in England and Wales (8:1 (Home Office 1990a: 113)) and for prisoners sentenced for a violent offence (39:1 (Home Office 1990b: 24-5)). Note that, in all these studies, although women account for a greater proportion of mentally disordered offending than other types of offending, they are always outnumbered by male mentally disordered offenders.
Although there are few studies, it also appears that a greater proportion of female offenders receive a psychiatric disposal. Allen (1987) looked at decision-making processes surrounding the trials of female and male defendants and found evidence for a gender bias operating mainly at the sentencing stage and favouring a psychiatric disposal for the female offender. Rather than a systematic tendency to |psychiatrize' female offenders (the explanation favoured by Sim (1990: 157-64)), Allen believed there ;vas EL greater reluctance to use a psychiatric disposal in the case of male offenders, even when they were more disturbed than women who did receive such a disposal. Suggested reasons for this attitude included a greater emphasis on moral and retributive factors in male cases, favouring punishment rather than treatment, and the shortage of facilities for the chronically and dangerously disordered offender. Existing figures cannot separate these two explanations but they confirm a higher rate of psychiatric disposals among female offenders. Overall, women appearing before the courts are roughly twice as likely as men to be dealt with by psychiatric means (Allen 1987: 123-6). This appears to be a stable finding over time; in 1961, psychiatric disposals (including probation with a condition of treatment) accounted for 0.8 per cent of all women convicted, compared to 0.5 per cent of men (Walker 1965). It must be stressed that, irrespective of gender, psychiatric disposals account for a tiny proportion of all cases appearing before the courts. Also, as men appear before the court in much greater numbers, the actual number of men receiving a psychiatric disposal is greater than the corresponding number of women.
Edwards's (1984) study of female defendants and the attitudes of criminal justice professionals also suggests that the criminality of women is more likely to be seen as an indicator of mental disorder deserving of treatment. Grounds's (1991) survey of transfers of sentenced prisoners to a special hospital found that the ratio of men to women was 14:1, whereas the corresponding ratio in the adult sentenced prison population was 3 1: 1. Overall, it appears that rates of psychiatric disorder are higher in female offenders and a greater proportion of female defendants are diverted away from prison. It is therefore difficult to predict rates of psychiatric disorder in sentenced women. They may be close to zero, suggesting that diversion procedures are operating at optimum efficiency and may even be diverting too many women, with little evidence of disorder. On the other hand, there may be considerable unmet need for psychiatric treatment.
The following account describes a 25 per cent cross-sectional sample of the sentenced female prison population and compares it to a similar, 5 per cent sample of the male sentenced population. A semi-structured interview by a psychiatrist was supplemented by information from prison and hospital records in order to assign psychiatric diagnoses. In addition, behavioural, criminal, and social characteristics were recorded, using operational criteria. Apart from the prison statistics, little is known about the characteristics of women serving a prison sentence and we collected data on the extent and nature of violent offending, self-harm, and disciplinary offences within prison. In addition to the presence of psychiatric disorder, we describe gender differences in previous psychiatric contact and the authors' assessment of current need for psychiatric treatment.
Sentenced women in England and Wales are held in eight prisons and four were surveyed, comprising an open prison, a closed training prison, a closed allocation centre, and a maximum security prison. A 50 per cent sample was selected at random from each prison, yielding a total of 301 women (24.5 per cent of the total population of 1,229 sentenced women (Home Office 1989& 32)). All women were invited to be interviewed, being informed that this was a confidential survey conducted by doctors from outside the prison, that participation was voluntary and no individuals would be identified. The selection and characteristics of the 5 per cent sample of the male sentenced population (1769 inmates) are described elsewhere (Gunn et al. 1991 a). Only inmates sentenced to immediate imprisonment were included in the study i.e., remand prisoners and those imprisoned for defaulting on fines were excluded.
Many women in prison are drug couriers from overseas and their presence complicates any comparison of male and female prisoners (Maden et al. 1992). They live in countries where patterns of psychiatric disorder and treatment services are likely to differ from those in the United Kingdom and it was therefore decided that all inmates with a permanent address in another country (39 women and 18 men) would be excluded from the comparison of psychiatric disorder, leaving a total of 262 women and 1751 men. The description of women from other countries is presented separately.
Interview and data collection
The prison file for each woman provided demographic and criminological information, reports of behaviour within prison and social enquiry reports. All criminal convictions were recorded, CRO dam being obtained later if it was not included in the prison file.
Each woman was interviewed by a psychiatrist (AM or MS), in private within the prison. The semi-structured interview was designed for the project and piloted on 50 volunteer prisoners, discarding items where inter-rater reliability was less than 90 per cent. It included questions about past and present medical and psychiatric problems, substance abuse, and self-harm. The Clinical Interview Schedule (CIS) (Goldberg et al. 1970) provided a standardized assessment of mental state. The interview could be completed in 30 minutes for subjects without evidence of psychiatric problems, expanded as necessary for others. Further information was obtained from the inmate's prison medical record and reports were obtained from NHS hospitals if previous treatment was reported.
Subjects with significant pathology were assigned a diagnosis, based on their present state, using clinical criteria according to the definitions in the International Classification of Diseases, 9th revision (ICD9) (World Health Organisation 1978). All diagnoses referred to the inmate's present state, except for diagnoses of drug or alcohol abuse or dependence. As drugs and alcohol are, in principle, not available in prison, inmates were asked about their behaviour in the six months preceding the index offence. Each individual could receive up to three diagnoses. The neurological diagnosis of epilepsy is also included because of previous findings of a high rate of epilepsy among prisoners.
Every inmate given a diagnosis was allocated to one of five categories of recommended treatment options:
(i) None. This is automatically applied to subjects with no diagnosis and also given
to those with a diagnosis who do not want treatment, unless the nature a nd degree of mental disorder is such as to require involuntary treatment un der the Mental Health Act (MHA) 1983.
(ii) Treatment within prison is treatment which an average general practitioner or
psychiatrist could provide on a mainly outpatient basis e.g., supportive psychotherapy and/or medication.
(iii) Therapeutic community refers to the type of contract-based regime where people
suffering from personality disorders are treated within the prison syste m at HMP Grendon and outside it at the Henderson hospital and some drug and alcohol rehabilitation centres.
(iv) Further assessment was used when there was uncertainty over diagnosis, treatment
or motivation. It was assumed that at least the initial stages of assess ment would take place in prison.
(v) Hospital meant inpatient treatment outside the prison system within the NHS,
including the special hospitals. It includes all cases needing involunta ry treatment under the MHA 1983 and all inmates willing to accept treatment voluntarily but suffering from psychiatric disorder which cannot be mana ged adequately and safely in a prison setting.
Decisions on recommended treatment were made on clinical grounds. In many cases, the decision was made by the interviewers. In all cases of psychosis, in cases where a recommendation for hospital transfer was likely and in all cases where the interviewers were in doubt about appropriate management, the case was referred to a research panel consisting of a group of psychiatrists, psychologists, and a social worker, all working in clinical forensic psychiatry and brought together at monthly intervals. Cases were presented to the panel, as if to a clinical case conference, and they were asked to select the most appropriate of the five treatment, options.
A list of the prisons visited and copies of the data collection sheet and coding manual can be obtained from the authors and are included in a report published by the Home Office (Gunn et al. 1991h).
When comparing proportions, odds ratios are used as an alternative to the [Chi.sup.2] test. The odds ratio is an estimate of relative risk. It has the advantage of describing the magnitude of the difference between proportions, rather than its statistical significance (Altman and Gardner 1989: ch. 4: 28-33). In place of significance levels, confidence intervals are given for each odds ratio. If the 95 per cent confidence interval for the population odds ratio does not include the value 1.0, the null hypothesis (that the population proportions are in fact equal) can be rejected and it can be assumed that a [Chi.sup.2] test would have been significant at the p <.05 level or higher.
Matching of sample and population
Table 1 shows that the distribution of index offences in the sample is similar to that found in the prison population.
[TABULAR DATA 1 OMITTED] Four UK resident women (1.5 per cent) refused to take part. The remaining 258 subjects were interviewed between April 1988 and july 1989. Three of the 39 women from other countries were not interviewed because of language difficulties, leaving 36 who will be described below; except where specified, the following account concerns prisoners ordinarily resident in the UK.
Details of sample: population matching for the male sample are contained in Gunn et al. (1991b). Both female and male samples were representative of their respective populations in terms of offence type, sentence length, and ethnic origin.
[TABULAR DATA 1 OMITTED]
History of violent offending
Of the 258 women interviewed in the present study, 53 (21 per cent) were serving a sentence for violent offences, compared to 22 per cent of the male sample. A group of inmates with a more pronounced record of violence was identified, using a stricter criterion which requires a record of three or more convictions for minor violence or a single conviction for severe, life-threatening violence, taking account of both the index arid previous offences.
Forty-three women (17 per cent) and 439 men (25 per cent) met these criteria, emphasizing the greater prevalence of violent offending among male inmates; the odds ratio for the difference in proportion is 1.7, the 95 per cent confidence interval (c.i.) being 1.2 to 2.4. In addition to this difference in prevalence, Table 2 reveals marked gender differences in the pattern of violent offending.
Table 2 Sentenced Prisoners: Violent Offenders by Index Offence and Gender Index offence Women Men n % n % Murder or attempt 26 60 154 35 Assault/wounding 6 14 95 22 Theft/robbery 6 14 141 32 Sex offences 22 5 Arson 3 7 6 1 Other 2 3 21 5 Total 43 100 439 100
The 26 women sentenced for |murder or attempt' includes 22 serving a life sentence for murder, none of whom had served a previous sentence. By contrast, the male sample includes 120 men sentenced for murder, including 39 (33 per cent) who had served at least two previous prison sentences.
The 22 women sentenced for murder can be compared to 22 similar men, matched for age at time of sentence. The victims of the women were husband or cohabitee in 16 cases, other family (two), a dependent adult (two) and a stranger (two). The victims of the men were a stranger (11), a male acquaintance (nine) and wife or cohabitee (two).
As assessed from reports of the trial, the circumstances of the killing in the case of the women were a triangular sexual relationship (ten cases), domestic violence (eight), financial gain (three), and terrorism (one). The circumstances of the killings by men were rows or disputes (ten cases), financial gain (five), sexual (four), domestic violence (two), and a triangular sexual relationship (one).
Disciplinary record within prison
Table 3 shows that women are more likely than men to have at least one disciplinary offence on record during the three months preceding the interview. Eight women (3 per cent) had a record of five or more disciplinary offences during this period, compared to 19 men (1 per cent). However, women were less likely to have been transferred to another prison for disciplinary reasons or to have lost remission during the present sentence (this analysis was limited to those serving a determinate sentence and therefore eligible for loss of remission)..
[TABULAR DATA 3 OMITTED]
Deliberate self-harm (DSH)
Eighty-two women (32 per cent )reported deliberate self-harm on at least one occasion; 31 (12 per cent) had harmed themselves more than once. Two hundred and ninety-seven men (17 per cent) reported self-injury, 131 (8 per cent) on more than one occasion. Table 4 compares the prevalence of deliberate self-harm in male and female prisoners.
[TABULAR DATA 4 OMITTED]
Method of deliberate self-harm
Reported methods of DSH are shown in Table 5; an individual may report more than one method. |Other methods' consisted of jumping from high places, fire (reported by seven men and no women), and gassing. The only significant gender difference is the higher prevalence of overdoses in women. Of the 25 women reporting cutting, 17 reported DSH only outside prison. Thirteen had cut themselves on more than one occasion. Self-cutting was reported by 15 (15 per cent) of the 99 women with a history of violent offending (defined as at least one conviction for violence against the person) compared to 10 (6 per cent) of non-violent women.
[TABULAR DATA 5 OMITTED]
Psychiatric treatment before imprisonment
Psychiatric contact prior to the current period of imprisonment was reported by 45 per cent of women, compared to 36 per cent of men (Table 6), with women less likely to report no contact and more likely to report adult out-patient or in-patient treatment. Men are more likely to report treatment only at a child guidance clinic.
[TABULAR DATA 6 OMITTED]
Psychiatric treatment within prison
Sixty-eight women (26 per cent) reported treatment with psychotropic medication during the current period of imprisonment, as did 131 men (8 per cent) (odds ratio = 4.4, 95 per cent c.i. = 3.2-6.2). None of the women (or the men) reported being given medication against their will. A few stated that other inmates were |drugged up' without consenting, but attempts to follow up these allegations revealed no evidence to support them. The most frequent complaint, among women and men, was doctors' reluctance to prescribe sedatives on request.
The Clinical Interview Schedule (CIS) is a standardized measure of psychiatric symptoms that has been widely used in hospital and general practice settings. It rates symptoms (mainly of anxiety and depression) as a single numerical score which takes into account the number of symptoms present and their severity. Most people score positively on a few items and the cut-off point, at which a person becomes a |case', is somewhat arbitrary.
The mean score for women was 7.1, compared to 4.8 for men (t = -5.26, p <.00 1). A |high' score (above 15) was recorded in 35 women (14 per cent) and 103 men (6 per cent) (odds ratio = 2.5, 95 per cent c.i. = 1.7-3.8). A high score does not confirm the presence of psychiatric disorder but makes it very likely.
Present psychiatric diagnosis
One hundred and forty seven women (57 per cent) were given at least one diagnosis. Thirty-nine of these women also received a second diagnosis and four were given a third. The prevalence of all diagnoses is shown in Table 8 alongside rates for male prisoners. Multiple diagnoses mean that totals will exceed 100 per cent. The commonest second diagnosis was personality disorder and most of the cases with multiple diagnoses consist of permutations of substance abuse, personality disorder, and neurosis. Five women with mild mental handicap and one woman with epilepsy also received a second diagnosis of personality disorder. Five hundred and ten men (38 per cent) received at least one diagnosis, compared to 57 per cent of women. The female:male odds ratio is 3.2, the 95 per cent confidence interval being from 2.5 to 4.2.
The four women with a psychosis consisted of three with schizophrenia and one with paranoia. All were rated as requiring transfer to hospital.
Three of the six women with a primary diagnosis of mental handicap were severely disordered, with many previous psychiatric admissions. Review of their hospital records did not produce unequivocal evidence of psychosis and they were given a secondary diagnosis of severe personality disorder. All were contained within one prison and it was our impression that the attitudes of medical staff locally, including the visiting psychiatrist, were part of the reason for these women having remained in prison rather than being transferred to hospital. Later, one of the women was transferred to special hospital; the other two were rejected by their local services as too violent and had not been referred to special hospital as they were not believed to be sufficiently dangerous.
[TABULAR DATA 8 OMITTED]
Table 9 shows the estimated number of cases in the prison population as a whole. The figures are estimated point prevalences during the period of the survey, from April 1988 to July 1989. Many more women and men would have passed through the system during this period; the table represents a snapshot view. The researchers' recommendations for treatment are shown in Table 10.
Table 9 Estimated Total Number of |Cases' in the Sentenced Prison Population, by Diagnosis and Gender Diagnostic group Women Men Psychosis 16 680 Neurotic disorder 160 2,080 Personality disorder 184 3,540 Alcohol abuse/dependence 96 4,060 Drug abuse/dependence 268 4,060 Mental handicap 24 220 Other disorders 8 900 Epilepsy 4 80
[TABULAR DATA 10 OMITTED] The women recommended for hospital transfer include the four cases with a primary diagnosis of psychosis, four with a primary diagnosis of mental handicap (see discussion above), three with severe personality disorder, and one with severe neurotic depression that had not responded to treatment.
Women from other countries
Thirty-one per cent of overseas women received a diagnosis (Table 11), compared to 57 per cent of women resident in the UK (odds ratio = 3.0, with a 95 per cent c.i. of 1.4-6.4).
Table 11 Female Sentenced Prisoners Ordinarily Resident Oversease: Psychiatric Diagnoses Diagnosis n % Neurotic disorders 10 28 neurosis (6) (17) adjustment reaction (4) (11) Drug dependence/abuse 1 3 No diagnosis 25 69 Total 36 100
Although the ram of neurotic disorder is higher than in UK resident women, the odds ratio of 2.1, with a 95 per cent c.i. of 0.9-5.0, suggests that this may be a chance finding.
A cross-sectional sample of prisoners is inevitably biased towards those serving longer sentences, for more serious offences. It is not representative of all offenders and contains relatively few women who serve frequent, short sentences. The same limitations affect both the female and male samples, so valid comparisons can be made. The survey shows that, in addition to differences in the frequency of offending, there are marked qualitative differences between the offending histories of women and men in prison, illustrated here by different patterns of violent offending. Whilst these figures emphasize the very different nature of the female and male prison populations, they do not suggest that women were serving sentences for less |rational' or less |serious' offences. Rather, the differences in motive and nature of the offence make numerical comparisons difficult. For example, most of the women sentenced for homicide had no previous convictions and had therefore been sentenced to imprisonment for their first offence; this was not the case for men sentenced for homicide. Such findings give a different slant to Kennedy's (1992: 22) statement that |women are actually imprisoned with fewer convictions than men'. NACRO figures show that 53 per cent of women have less than three convictions when they first go to prison, compared to 23 per cent of men, but this finding has no meaning without knowledge of the offences for which they are imprisoned.
Disciplinary offences within prison
A significantly greater proportion of women had disciplinary offences recorded against them during the preceding three months. Women in prison have had a reputation for a high level of disturbed behaviour since 1862 (Mayhew and Binny: 180-2) and a report in 1910 (Quinton) comments on the need for higher staffing levels to manage the women's section of Millbank prison. Dobash et al. (1986: 62-88) review the historical development of prison regimes for women, drawing attention to the |conventional wisdom' that women, whilst less violent outside prison, behave more violently than men once locked up.
Some evidence from other institutions supports this view and a survey of violent incidents in a maximum security hospital (Larkin et al. 1988) found that women, who account for only 25 per cent of the patient population, were involved in 75 per cent of incidents. The authors note that a small number of highly disturbed patients accounted for most incidents with one woman being responsible for 12 per cent of violent incidents. This uneven distribution emphasizes the pitfalls inherent in generalizing about the nature of |women in custody' from figures of this type.
It remains the case that women within prison are punished for disciplinary offences more often than men (NACRO 1986). The rate of offences per 100 prison population for all female establishments in 1985 was 335, whereas for men it was 160, but the majority of these offences were disobedience or disrespect.
The findings presented in Table 3 suggest that, although a greater proportion of women are disciplined, they are much less likely to be punished by loss of remission, the response to more serious offences within prison. They are also less likely to have been transferred for disciplinary reasons, the most common sanction against inmates who absconded from prison (or home leave) or intimidated other inmates. These results do not suggest that women in prison are generally more disturbed or violent than men. It appears that the management of men within prison presents more serious disciplinary problems but women are more likely to be disciplined for trivial offences.
This point has been made by Carlen (1985a: 134), who cites it as evidence that prison regimes for women are more oppressive than those for men and require a higher standard of behaviour. Mandaraka-Sheppard argues that aggressive offences by female prisoners can best be explained as |. . . a function of their response to the particular negative aspects of institutions' (1986: 189). Her study of three open and three closed female prisons found institutional rather than individual characteristics to be more important in accounting for violent incidents. She criticized the trivial nature of many recorded offences, the vague nature of some rules and the consequent inconsistency in their application. It has also been suggested that the attitudes of female prison officers are partly responsible for the high number of trivial offences recorded in women's prisons (Kozuba-Kozubska and Turrel 1978).
All these factors may be important. The comparison with male prisoners is complicated by the difference in regimes. In the prisons visited for the present study, women usually had greater physical freedom within prison than men with a similar security classification. The long lock-up periods that characterize male prisons, even those for long-term inmates (King and McDermott 1989), are less common in female prisons. Styal provided considerable freedom of movement within a secure perimeter for the majority of its inmates. This allows more interaction with prison officers and increases the potential for transgressing minor prison rules. For the same reason, many male inmates stated that they found |open' prisons more oppressive; the increased freedom meant increased regulation of behaviour and some chose to return to closed prisons. Mandaraka-Sheppard's description of an open prison regime (1986: 52-3) brings out the paradox whereby (conditional) freedom of movement increases the opportunities to breach regulations.
Deliberate self-harm (DSH)
The women in our sample had a higher lifetime prevalence of deliberate self-harm but this difference was mainly due to overdosing outside prison. Prevalence rates for DSH during the current period in custody are the same for both sexes at around 5 per cent. The lifetime prevalence of cutting in female prisoners is 10 per cent (the 95 per cent c.i. is 7-13 per cent), similar to the rate in men and not significantly different from the 8 per cent prevalence rate found in Wilkins and Coid's (1991) survey of remanded women. This finding does not take account of possible gender differences in the frequency of self-harm but it is surprising, as self-cutting in male prisoners has received less attention in the literature. Looking at other self-destructive behaviour, Dooley (1990) found that only 1.7 per cent of all prison suicides were female in the period 1972 to 1987, whilst women accounted for 3 per cent of the prison population. The present study's finding of an association between self-cutting and violent offending reinforces We results of Wilkins and Coid, who found associations with behavioural disturbance, substance abuse, and personality disorder. An earlier study of women who injured themselves in Holloway (Cookson 1977) also found an association with violent offending.
The high prevalence of self-injury outside prison emphasizes the importance of individual vulnerability in any attempt to explain self-injury in prisoners. Only four women reported self-harm exclusively within custody.
History of psychiatric treatments
Women in prison report higher rates for all forms of adult psychiatric treatment before imprisonment. This is consistent with gender differences within the general population where the female: male ratio in 1989 was 2.4:1 for GP consultation due to mental illness and 1.1:1 for first hospital admissions (Department of Health 1990: 54-61).
More pronounced differences are found in treatment rates within prison, in line with official figures (Home Office 1986) which show a higher total number of doses of medication per 100 inmates in female prisons. The use of psychotropic medication in female prisons has been criticized (Benn 1983) and it has been suggested that A serves a control function (Glick and Neto 1977). An alternative explanation is that women prisoners have greater needs for psychiatric treatment (Morris 1987: 124) and our findings show that rates of treatment by psychological methods, usually counselling or psychotherapy, were also higher. Our impression was that treatment demand was higher in women's prisons and that prison medical staff were more willing to respond. All hospital officers in women's prisons are trained nurses and may be seen as more sympathetic by inmates than hospital officers in male prisons, where only a minority have a nursing qualification. In addition, the culture of male prisons often saw emotion as a sign of weakness, so that distress may result in an act of violence, rather than a request to see the doctor.
Treatment facilities in women's prisons may be better than those in many male prisons but this is not to say that they met the needs of all or most women. Many women did not regard nursing or medical staff as sympathetic, seeing them as primarily agents of the institution. Lack of confidentiality was often cited by both men and women as a reason for not revealing personal problems to officers.
The prevalence of psychiatric disorder
The higher level of anxiety or neurotic symptoms among women (as measured by the CIS) also suggests a greater potential demand for treatment and this is reinforced by differences in the prevalence of psychiatric disorder. Overall rates of psychiatric disorder are less useful than looking at rates of particular diagnoses and these show that significant gender differences are not found in rates of psychosis, whilst women have a higher prevalence of personality disorder, neurosis, drug abuse, and mental handicap/ learning difficulties (Table 8).
The validity of our findings is likely to be highest for psychosis (a severe disorder with some specific symptoms), followed by neurotic disorders and substance abuse (the assessment included standardized, operational measures for symptoms). There are greater reservations about mental handicap (in the absence of standard measures, cases could have been missed) and personality disorder (a diagnosis that presents many problems of both principle and practice). In the latter cases, our priority was to avoid over-diagnosis, so our assessments of demand for services should be regarded as minimum estimates.
Our study cannot explain the observed gender differences in psychiatric disorder. They must arise from an interaction of rates of mental disorder among offenders and the operation of procedures at various stages of the criminal justice system which are intended to divert mentally disordered offenders away from the prison system. Our results do not imply any tendency of the courts to imprison mentally disordered women more readily than men in a similar position. This point can be made by considering in more detail drug dependence and schizophrenia.
Table 9 suggests that there are 268 drug dependent women in prison, compared to 4,060 men, i.e., women make up about 6 per cent of the total number of drug dependent, sentenced prisoners. In the UK as a whole, women accounted for 29 per cent of new addicts notified to the Home Office in 1989 and the figure has been between 26 and 29 per cent during the preceding ten years (Home Office 1989h). A similar argument can be constructed for personality disorder, where men also have higher rates in community surveys (Regier et al. 1988).
It is estimated that about 150,000 people suffer from schizophrenia in England and Wales (Bebbington and Hill 1985: 38) and roughly half are women. Even if it is assumed that all the cases of psychosis in Table 9 represent schizophrenia, our figures suggest that less than 0.05 per cent of women with this disease are serving a prison sentence, the corresponding figure for men being 1 per cent. These findings contradict Penrose's (1939) |law', with its suggestion that the populations of prisons and psychiatric hospitals are interchangeable. The closure of many psychiatric beds over the last 30 years has not resulted in a large scale movement of women with schizophrenia into the sentenced prison population.
The low numbers should not obscure the tragic situation of women within prison who suffer from severe mental disorder, whether their diagnosis is psychosis or the severe end of the range of disorders grouped under the blanket labels of personality disorder or learning difficulties. These women make up the 12 for whom transfer to hospital was recommended (Table 10). Reasons for recommending transfer included the rejection, exploitation, and victimization of these women by other inmates. They did not merge with the |ordinary' prison population and required some form of special management, whether in a sick bay or disciplinary unit (see Maden 1992 for a fuller description). They shared many of the characteristics of male prisoners who are difficult to place and risk rejection by psychiatric services, including non-compliance, substance abuse and recidivism (Cheadle and Ditchfield 1982; Coid 1991). Most of these women were in prison because of a lack of suitable NHS provision.
Rates of mental handicap/learning difficulties are higher among female prisoners but intelligence tests were not used and the figures represent no more than an approximate, administrative prevalence. Descriptions of the cases are contained in Maden (1992); it appeared that idiosyncratic practice at one prison was partly responsible for this finding. Some of the women with learning difficulties also had severe personality disorders and mental illness. Their hospital records suggested a gap in service provision, as they described rejection by services for those with learning difficulties and by services for the mentally ill.
Women from other countries
Rates of psychosis, substance abuse, and personality disorder were much lower than for UK resident women, although the rate of neurotic disorder was high. If anything, cultural and language difficulties would mean that our figures are an underestimate of levels of anxiety and depression in these women. There is nothing to suggest that any cases of psychosis or other severe disorder were missed. The usual criticism of psychiatrists dealing with people from other cultures is that they tend to over-diagnose psychosis, but this is not even a theoretical possibility, as we report no cases in women from other countries.
The different pattern of psychiatric disorder in the two groups vindicates the decision to analyse them separately and to exclude overseas residents from the comparison of rates of psychiatric disorder by gender. Had they been included in the comparison, the effect would have been to reduce the apparent rate of most psychiatric disorders in women.
Apart from a single case of drug dependence, the only diagnosis found in overseas women was neurotic disorder, mainly depression or an adjustment reaction. The likelihood of developing depression is increased by adverse social circumstances and a lack of social supports--two characteristics of overseas women in prison, who often have language difficulties and rarely receive visitors. These women are not the central focus of the present report but they form a substantial minority within the female prison population and present a challenge to prison psychiatric services. The social isolation and language problems that contribute to their high rate of neurotic disorder may also act as obtacles to their receiving the counselling help and medical care that they need. Green (1991) provides a fuller description of the background, motivation, and problems within prison of overseas drug couriers.
Treatment needs of women serving a prison sentence
We have described above the needs of a handful of severely disordered female prisoners who require specialist treatment within the NHS. Many more women in prison have unmet treatment needs in relation to neurotic and personality disorders and drug dependence. Prisoners with these diagnoses are less likely to be referred to or accepted by hospital services (Dell et al. 1991). An earlier study (Dell and Gibbens 197 1) showed that only 9 per cent of women remanded to Holloway for medical reports were accepted for treatment whilst 53 per cent of those identified as suffering from personality disorder were rejected. A central problem is psychiatry's ambivalent attitude towards patients with this diagnosis, which is often used in a pejorative way (Lewis and Appleby 1988). Carlen (1985b) criticises doctors' readiness to attach the diagnosis whilst refusing to provide any treatment.
The Criminal Justice Art 1991 stresses the importance of considering alternatives to imprisonment, and expands the range of treatment options for drug users. Nevertheless, it is likely that many women with drug and personality problems will continue to receive prison sentences. Guidelines circulated by the Directorate of the Prison Medical Service (1991) are welcome for their emphasis on throughcare but it is too early to judge their success. There is no female equivalent of the therapeutic regime at Grendon prison. Whilst the small population would not justify a whole prison being given over to this purpose, there is a need for a trial of a wing running on therapeutic community principles, possibly aimed at women who repeatedly self-harm.
Rather than a global need for |treatment', as envisaged by those planning the new Holloway, the female prison population contains several important groups with disparate needs. The most severely disturbed, a small minority, need to be in hospital and their continued presence in prison is an indictment of NHS psychiatric services. A much larger group of women could benefit from a range of alternatives to custody. The main unmet need for treatment within prison is from women with substance abuse and personality problems, who require a range of services, available on a voluntary basis.
The |Reed' review of services for mentally disordered offenders identifies women as a group with special needs and re-asserts the principle that care should be provided mainly by health and social services and in a community setting whenever possible (Department of Health and Home Office 1992). Many of women's mental health problems can be readily detected within a prison setting. The fact that they are highly visible there does not mean that prison is the best place to tackle these problems. The presence of women with mental disorder in prison reflects failings in services elsewhere and it follows that the solution also lies elsewhere, in the type of multi-agency cooperation proposed by Reed.
[TABULAR DATA 7 OMITTED]
Allen, H. (1987), |Official Statistics on Psychiatric Disposals', in Justice Unbalanced. Gender, Psychiatry and Judicial Decisions. Milton Keynes: Open University Press. Altman, D. G., and Gardener, M.J. (1989), |Calculating Confidence Intervals for Proportions and Their Differences', in Gardner M. J. and Altman D. G., eds, Statistics with Confidence. London: BMJ. Bebbington, P. E, and Hill, P. D. (1985), A Manual of Practical Psychiatry, p. 38. London: Blackwell. Benn, M. (1983), |Women in Prison ... Breaking the Silence', Spare Rib, November, 51-5. Callaghan, J. (1968), Hansard, 16 December. Carlen, P. (1983), Women's Imprisonment. A Study in Social Control. London: Routledge and Kegan Paul. Carlen, P., ed. (1985a), Criminal Women. Cambridge: Polity Press. Carlen, P. (1985b), |Law, Psychiatry and Women's Imprisonment. A Sociological View', British Journal of Psychiatry, 618-21. Cheadle, J., and Ditchfield, J. (1982), Sentenced Mentally Ill Offenders. London: Home Office Research and Planning Unit. Coid, J. W. (1991), |"Difficult to Place" Psychiatric Patients: The Game of Pass the Parcel Must Stop', British Medical Journal, 302: 603-4. Cookson, H. M. (1977), |A Survey of Self-injury in a Closed Prison for Women', British Journal of Criminology, 17: 332-46. Dell, S., and Gibbens, T. C. N. (1971), |Remands of Women Offenders for Medical Reports', Medicine, Science and the Law, 11:117-27. Dell, S., Grounds, A., James, K., and Robertson, G. (1991), Mentally Disordered Remanded Prisoners. Report to the Home Office. Department of Health (1990), On the State of the Public Health for the Year 1989--The Annual Report of the Chief Medical Officer. London: HMSO. Department of Health and Home Office (1992), Review of Health and Social Services for Mentally Disordered Offenders and Others Requiring Similar Services. Final Summary Report, Cmnd. 2088. Directorate of the Prison Medical Service (1991), Caring for Prisoners: A Multidisciplinary Resource for People Working with Prisoners. London: DPMS, Home Office. Dobash, R. P., Dobash, R. E., and Gutteridge, S. (1986), The Imprisonment of Women. Oxford: Blackwell. Dooley, E. (1990), |Prison Suicide in England and Wales 1972-87', British Journal of Psychiatry, 156: 40-45. Edwards, S. S. M (1984), Women on Trial. Manchester: Manchester University Press. Gibbens, T. C. N. (1971), |Female offenders', British Journal of Hospital Medicine, September, 279-86. Glick, R., and Neto, V. (1977), National Study of Women's Correctional Programs. Washington, DC: Government Printing Office. Goldberg, D. P., Cooper, B., Eastwood, M. R., Kedward, H. B., and Shepherd, M. (1970). |A Standardized Psychiatric Interview for Use in Community Surveys', British Journal of Preventive and Social Medicine, 24: 18-23. Green, P. (1991), Drug Couriers. London: Howard League for Penal Reform. Grounds, A. (1991), |The Transfer of Sentenced Prisoners to Hospital 1960-1983', British Journal of Criminology, 31: 54-71. Gunn, J., Robertson, G., Dell, S., and Way, C. (1978), Psychiatric Aspects of Imprisonment. London: Academic Press. Gunn, J., Maden, A., and Swinton, M. (1991a), |Treatment Needs of Prisoners with Psychiatric Disorders', British Medical Journal, 338-41. -- (1991b), Mentally Disordered Prisoners. London: Home Office. Hafner, H., and Boker, W. (1982), Crimes of Violence by Mentally Abnormal Offenders. The Psychiatric Epidemiological Study in the Federal Republic of Germany. Cambridge University Press: Cambridge. Heidensohn, F. (1985), Women and Crime. London: Macmillan. Home Office (1986), Report on the Work of the Prison Department 1985-86: 93. London: HMSO. -- (1989a), Prison Statistics: England and Wales 1988. London: HMSO. -- (1989b), |Statistics of Drug Addicts Notified to the Home Office, United Kingdom, 1988', Home Office Statistical Bulletin 13189. London: Government Statistical Service. -- (1990a), Criminal Statistics England and Wales 1989. London: HMSO. -- (1990b), Prison Statistics: England and Wales 1989. London: HMSO. Kennedy, H. (1992), Eve was Framed. Women and British Justice. London: Chatto and Windus. King, R. D., and McDermott, K. (1989), |British Prisons 1970-1987. The Ever-Deepening Crisis', British Journal of Criminology, 2: 107-128. Kozuba-Kozubska, J. and Turrel, D. (1978) |Problems of Dealing with Girls'. Prison Service Journal, 29: 4. Larkin, E., Murtagh, S, and Jones, S. (1988),|A Preliminary Study of Violent Incidents in a Special Hospital', British Journal of Psychiatry, 226-31. Lewis, G., and Appleby, L. (1988), |Personality Disorder: The Patients Psychiatrists Dislike', British Journal of Psychiatry, 44-9. Maden, A. (1992), Psychiatric Disorder in Women Serving a Prison Sentence. University of London MD thesis. Maden, A., Swinton, M., and Gunn, J. (1992), |The Ethnic Origin of Women Serving a Prison Sentence', British Journal of Criminology, 32: 218-21. -- (1993), |Psychiatric Disorder in Women Serving a Prison Sentence', British Journal of Psychiatry, in press. Mandaraka-Sheppard, A. (1986), The Dynamics of Aggression in Women's Prisons in England. Aldershot: Gower. Mayhew, H., and Binny, J. (1862), The Criminal Prisons of London and Scenes of Prison Life. London: Griffin, Bihn & Co. Moorehead, C. (1985), |The Strange Events at Holloway', New Society, 11: 40. Morris, A. (1987), Women, Crime and Criminal Justice. Oxford: Blackwell. NACRO (1986), Offences Against Discipline in Women's Prisons. London: NACRO. Penrose, L. S. (1939), |Mental Disease and Crime: Outline of a Comparative Study of European Statistics', British Journal of Medical Psychology, 18: 39-41. Quinton, R. (1910), Crime and Criminals. London: Longmans, Green & Co. Regier, D. A., Boyd J. H., Burke, J. D. et al. (1988), |One-Month Prevalence of Mental Disorders in the United States. Based on Five Epidemiologic Catchment Area Sites', Archives of General Psychiatry, 45: 977-86. Sim, J. (1990), Medical Power in Prisons. The Prison Medical Service in England 1774-1989. Milton Keynes: Open University Press. Walker, N. (1965), Crime and Punishment in Britain. Edinburgh: Edinburgh University Press. West, D. J. (1965), Murder Followed by Suicide. London: Heinemann. Wilkins, J., and Coid, J. (1991), |Self-Mutilation in Female Remanded and Prisoners', Criminal Behaviour and Mental Health, 1: 247-67. Wooton, B. (1959), Social Science and Social Pathology. London: Allen and Unwin. World Health Organisation (1978), Mental Disorders: Glossary and Guide to Their Classification in Accordance with the Ninth Revision of the International Classification of Diseases. Geneva: WHO.
A. Maden, M. Swinton, and J. Gunn(*) (*) Department of Forensic Psychiatry, Institute of Psychiatry, London. The study was funded by the Directorate of the Prison Medical Service at the Home Office. Dr Graham provided help and advice throughout.
Some of the material forms part of an MD thesis submitted to the University of London (1992).
|Printer friendly Cite/link Email Feedback|
|Author:||Maden, A.; Swinton, M.; Gunn, John|
|Publication:||British Journal of Criminology|
|Date:||Mar 22, 1994|
|Previous Article:||Conditions of successful reintegration ceremonies: dealing with juvenile offenders.|
|Next Article:||Homosexuality as deviance: the case of the gay police officer.|