Audit of transfers under the Mental Health Act from prison -- the impact of organisational change

AIMS AND METHOD

To describe the change in the number of referrals of prisonersand the delay in transfer to hospital under the Mental HealthAct following a change in the prison health care provision.The transfer time (time from referral to transfer to psychiatrichospital) of prisoners has been audited previously over 1996and 1997.

RESULTS

There has been an increase in the number of prisoners transferred.Both transfers under Section 48 of the Mental Health Act andthe proportion of transfers to high security have increased.The average delay in transfer remains lengthy and there isa trend of increasing delay with increasing level of placementsecurity.

CLINICAL IMPLICATIONS

Despite Government policies to facilitate the transfer of mentally disordered offenders, we found an increase in the delay to hospitalcompared with previous audits.

Introduction

TOP

Introduction

The aim of this audit was to describe and compare the transfersof prisoners to psychiatric services for the year 1 December1998 to 1 December 1999, following the partnership betweenOxleas and HMP Belmarsh, with prisoner transfers in previousyears.

Method

TOP

Method

Patient information was collected on the database at HMP Belmarsh,with a specific subsection designed for the referral and transferprocess from 1 December 1998 to 1 December 1999. Previous auditfigures were available from 1 June 1996 to 1 June 1998. Broadlysimilar data were available over the period of the study. Thedata collected comprised demographic details, the nature of placements and the frequency of sequential referrals. The timesto assessment and outcome were noted. The transfer time ofthe patient is the number of days from the point of first referralto subsequent transfer to psychiatric hospital. Because thedata did not meet the requirements of a parametric test, theanalysis used was the Mann—Whitney test.

Background to the audit

TOP

Background to the audit

It is Government policy that, wherever possible, mentally disordered offenders (MDOs) should receive care and treatment from healthand social services. In the Circular 66/90 Home Office andDepartment of Health (1995) report it was recommended thatthe clinical criteria for urgent psychiatric transfer from prison should not differ from those that apply to a person inthe community requiring urgent admission to hospital. The HomeOffice has been actively encouraging the use of the emergencyprovisions of the Act.

Numerous studies over recent years have reported the prevalenceof mental disorder in prisoners and their treatment needs (Gunn et al, 1991; Birmingham et al, 1996; Brooke et al, 1996). Other studies have reported difficulties in the referral process of prisoners to health services (Coid, 1988; Smith et al, 1992).The HM Inspectorate of Prisons report for 1997 (Prison Service, 1999) highlighted the shortfall in the provisionof health care to prisoners and laid out clear recommendationsfor immediate and future changes and further concerns in relationto mental health needs. The Department of Health (1999) recommendeda formal partnership between the NHS and the prison serviceto improve the health and care of prisoners. More recently,others have drawn attention to the mental health needs of prisoners(Smith, 1999; Reed & Lyne, 2000).

The Bracton Centre, Oxleas NHS Trust, has had some involvementwith providing psychiatric care for HMP Belmarsh since 1991,although the extent of provision has varied over time. In December1998 the trust and prison developed a partnership agreementto provide psychiatric care to the prison, with a consultant-ledteam being based at Belmarsh.

Overall picture of the patient group

TOP

Overall picture of the...

The prisoners referred and transferred to hospital under theMental Health Act 1983 for the period of 1 year from 1 December1998 are described. Forty-eight patients were transferred duringthis period, ranging in age from 20-58 years. In terms of ethnicity,28 (58%) were White European, 11 (22%) were Black Caribbean,6 (13%) were Black African and 5 (10%) were of other origin.

In relation to the alleged offence, 25 (52%) were violent innature, including 12 murder charges, 8 (17%) were sexual offences,6 (13%) were criminal damage offences and 8 (17%) were acquisitiveoffences. A total of 19 cases (40%) were the responsibilityof neighbouring mental health trusts and 31 (60%) were theresponsibility of other mental health trusts, including 5 (10%)from outside the London region.

A further group of 20 prisoners who had been referred duringthis period but had not been transferred consisted of fiveprisoners rejected for transfer, two released prior to transfer,seven prisoners awaiting assessment outcome and a further sixprisoners who had already been accepted by an outside servicebut were still awaiting transfer. Some prisoners in this last group had already experienced lengthy delays (range 10-180 days).One prisoner died in custody while awaiting a bed.

In all but five cases (10%) the working diagnosis was a primarypsychotic illness. Transfers were under the category of mentalillness in 46 cases (96%), psychopathic disorder in anothercase and learning disability for the remaining case.

Audit

TOP

Audit

During the reported 1-year period (audit two) there has beenan increase in the annual number of prisoners transferred.Figures over the 2-year period 1996-1998 (audit one) show 80transfers in this period. The age range was similar to previousaudits (20-58 years compared with 20-61 years).

Section of Mental Health Act
In the most recent year (1998-1999) there has been a large increasein the proportion of prisoners transferred under Section 48of the Mental Health Act, which was used in 60% of cases. Thenumber of prisoners transferred on hospital orders (Section37 and Section 37/41) and sentenced (Section 47) has decreased(see Table 1).

View this table:
[in this window]
[in a new window]
Table 1. Section of the Mental Health Act (MHA)

Service type
A greater proportion of prisoners were transferred to high security (Broadmoor, n=6; Ashworth, n=5, Rampton, n=3) during the mostrecent year. Similar proportions of prisoners were transferred to medium security and placements of lesser security (see Table 2).

View this table:
[in this window]
[in a new window]
Table 2. Service type

Delay in transfer
The transfer times across all placement types did not differsignificantly between audit one and audit two (Z=-0.2958, P>0.05).The transfer time was studied by placement type and the resultsindicate a trend of increasing delay with the level of security (Table 3). Of note is the very short transfer time to the psychiatricintensive therapy unit placements in audit two, which werealmost all in private facilities and virtually all the prisonerstransferred there were from an area local to the prison.

View this table:
[in this window]
[in a new window]
Table 3. Delay in transfer

Comment

TOP

Comment

Over the first year of this service there was an increase inthe number of prisoners transferred compared with the averagenumber of transfers over the previous 2 years, but delays intransfer remained lengthy and formed a predictable responsefrom services. We found the unavailability of beds to be themost common reason for delay in the transfer process, with otherreasons being the differences in opinion regarding the levelof security required, legal reasons, diagnostic disagreementsresulting in repeated referrals and assessments, long delaysto initial assessment (more than 6 weeks) and catchment areadisputes.

To address the needs of prisoners requiring transfer, initiativeshave included promotion of the use of urgent transfer orders(Section 48) and the national increase in the number of bedsin medium security, including a sizeable proportion in theprivate sector. However, we experienced lengthy delays in theassessment process and the transfer of prisoners after acceptanceby a service due to a lack of availability of beds across all levels of security. Transfers to local general services werecharacterised by lengthy delays, suggesting that direct communityadmissions take priority with the impression of a preferenceto await completion of court proceedings also playing a part.

A greater proportion of prisoners required high-security placements compared with previous years. This is explained in part by HMPBelmarsh now receiving all prisoners committed to the CentralCriminal Court from the London area from other local prisons.It is noted that the referral process to high security is lengthyand this patient group often have a duplication of assessmentsimpeding the access to appropriate care.

Home Office statistics reporting the annual number of transfersfrom prison to hospital in England and Wales indicate that63% of patients admitted to hospital from prison in 1999 wereunsentenced or untried and only 10% of all the transfers wereto high security (Johnson & Taylor, 2000).

Prison statistics from the Annual Report of the Director ofHealth Care 1997-1998 (Prison Service, 1999) indicate shorterdelays in transfer process. A total of 166 prisoners in Englandand Wales were recorded as awaiting transfer on 31 March 1998and only 33 prisoners experienced a delay of greater than 28days. This may be explained by only the cases where the Sectionpapers have been submitted to the Home Office (in practice,when a bed is shortly available) or an order made in court.Contrary to expectation and clinical need, the use of emergencytransfer under Section 48 in itself has no positive effect onthe delay. An earlier report on transfers under Section 48contrasts with this present finding (Mackay & Machin, 1998).In our study the best expeditor for prisoners is underthe Sections of the Mental Health Act with a time limit imposed,that is Sections 37 and 38.

This audit has implications for NHS providers in prison, bearingin mind that the referrals of this prisoner group were madeby experienced psychiatrists. There are difficulties of treatingprisoners in custody; even those prisoners previously knownto services may remain in prison on no medication while awaitingtransfer if they decline treatment. Those prisoners who accepttreatment may become lower in priority to transfer, while remaining vulnerable in a prison setting. It is clear that considerablework needs to be done on the organisation and coordinationof the referral process. Measures need to be put in place toensure greater continuity of care of prisoners under the CareProgramme Approach.

References

TOP