Postcode lottery? Hospital transfers from one London prison and responsible catchment area

Abstract

Aims and method

To consider the link between responsible commissioner and delayed prison transfers. All hospital transfers from one London prison in 2006 were audited and reviewed by the prisoner's borough of origin.

Results

Overall, 80 prisoners were transferred from the audited prison to a National Health Service (NHS) facility in 2006: 26% had to wait for more than 1 month for assessment by the receiving hospital unit and 24% had to wait longer than 3 months to be transferred. These 80 individuals were the responsibility of 16 different primary care trusts. Of the delayed transfer cases (n=19), the services commissioned by three primary care trusts were responsible for the delays.

Clinical implications

There are significant differences in performance between different primary care trusts related to hospital transfers of prisoners, with most hospitals able to admit urgent cases within 3 months. This suggests that a postcode lottery operates for prisoners requiring hospital transfer. Data from prison services may be useful in monitoring and improving the performance of local NHS services.

Growing dissatisfaction with the state of the UK prison medical service throughout the 1970s and 1980s ¹ led to the publication by the government of The Future Organisation of Prison Health Care. ² This document set out plans for a formal partnership between the National Health Service (NHS) and the Prison Service. Various prisons, including the one in our study, piloted mental health in-reach teams in 2002, part of a process which ended with the NHS taking over the commissioning of health services for all prisons in 2006. The principle of equivalence of care states that prisoners are entitled to the same standard of healthcare as they would have were they not in prison. ^2-4 Those prisoners who have mental disorders of a nature or degree such that their needs can only be appropriately met by in-patient treatment are meant to be transferred from prison to hospital for treatment. ⁵ Prison in-reach teams take referrals from a wide range of sources and generally support a high volume of prisoners with stable mental illnesses. The prisoners described in this paper represent the most acutely unwell.

Although the principle of equivalence is meant to underpin the working of a mental health in-reach team, there is no equivalent of a prison healthcare wing. ⁶ A prison healthcare wing is not a hospital, and treatment under the Mental Health Act is not possible there. ⁷

The delays in transferring prisoners to hospital are well known. ^8-11 We have looked in more detail at a potential cause of the delay, comparing the performance of the responsible primary care trusts.

Method

The study was conducted in a London category B (medium security) local prison. Data were collected using monthly returns that were submitted to the mental health team between 1 January and 31 December 2006. All those transferred from the prison to an NHS mental health facility during that period were included. Information was collected regarding time to assessment and transfer, responsible primary care trust, relevant Mental Health Act section used and the security category of the receiving unit (general adult ward, psychiatric intensive care unit, medium secure unit or high secure hospital). At the end of the year, the available data were analysed.

Results

In 2006, 80 prisoners were transferred from the audited prison to hospital under the Mental Health Act, almost exclusively because of severe mental illness. Table 1 describes the section of the Mental Health Act under which they were transferred and their destination by level of security (low secure units are not widely available in London NHS hospitals and no patients were transferred to a low secure unit in this study). The prison team can, of course, only refer to the responsible NHS catchment area unit, but they may then choose to purchase a bed in the independent sector if one is not available in the NHS locally. We have included in our analysis patients transferred to both NHS and independent sector units.

Table 1 Prisoners transferred to hospital in 2006

	Adult ward	Psychiatric intensive care ward	Medium secure unit	High secure hospital
Section of the Mental Health Act 1983	n
Hospital Order (s37)	1	7	4	0
Interim Hospital Order (s38)	0	2	4	1
Hybrid Order (s45A)	0	0	1	0
Transfer Warrant for Sentenced Prisoner (s47)	2	7	10	0
Transfer Warrant for Unsentenced Prisoner (s48)	1	18	22	0

Prisoners who need such transfer for hospital treatment under the Mental Health Act are the responsibility of their borough of origin, determined by their home address, their general practitioner's address, or where their (alleged) offence was committed. ¹² This required the prison mental health in-reach team to liaise with services commissioned by 16 different primary care trusts: 8 in Inner London, 3 in Greater London, and 5 from other parts of England.

All 80 prisoners received a Mental Health Act assessment within 24 hours of seeing the prison in-reach team's psychiatrist, and all were referred to the appropriate NHS service within 1 week (80% within 3 days). Current practice is that the receiving hospital then arranges its own assessment of the prisoner: 39 individuals (49%) were assessed in this manner within 2 weeks of being referred, 20 (25%) were assessed within 2-4 weeks, and 21 (26%) waited longer than 1 month for such an assessment. Of these 80 prisoners, 70% were the responsibility of the services commissioned by four Inner London primary care trusts and we considered their performance in more depth (Table 2).

Table 2 Performance of the services commissioned by the four Inner London primary care trusts^a (PCTs) responsible for 70% of the transfers (with comparison to the average for all PCTs) in terms of assessment

	Assessed, %
	<2 weeks	2–4 weeks	>4 weeks
PCT 1	60	34	6
PCT 2	17	50	33
PCT 3	32	31	37
PCT 4	50	6	44
Average for all 16 PCTs	49	25	26

In terms of time from referral to transfer to hospital, 39 prisoners (49%) were transferred within 1 month and 22 (27%) were transferred within 1-3 months of referral, but 19 (24%) waited longer than 3 months. Of the 19 delayed (>3 months) cases, the services commissioned by one Inner London primary care trust (PCT3, Table 2) were responsible for 11 (58%) individuals. The services commissioned by another Inner London primary care trust (PCT4, Table 2) were responsible for 7 (37%). With the exception of one case (the responsibility of a Greater London primary care trust), the services commissioned by all the other primary care trusts were able to transfer their patients within 3 months of referral.

Discussion

This is a small sample from one prison in London and it may not be generalisable. When looking at delays we considered all patients transferred and it may have been that the courts were responsible for some of the delays for those transferred under court orders (Sections 37, 38, and 45A of the Mental Health Act). However, in our experience, these orders can often work more quickly as the court will not necessarily insist on a secure unit in the way that the Ministry of Justice tends to do for Sections 47 and 48. Also, if a court-related delay arose, it would still remain possible to transfer the patient under Section 48.

Delays

The delays for prisoners requiring hospital transfer are well known. ^8-11 A greater proportion of the transferred prisoners in this study were considered appropriate for forensic settings (medium security) than general adult units. We would argue that prison psychiatry is neither forensic nor general, but has unique skills of its own, covering the entire range of mental health need, from primary care to high security.

Why were there delays? There were none from the prison team - all prisoners were referred to the appropriate hospital within 1 week, and most within 3 days. Nonetheless, 24% of the group waited more than 3 months for transfer to hospital, a lengthy delay for a substantial number of severely mentally ill prisoners and out of step with the National Service Framework requirement of timely access to a hospital bed (there is now an expectation of transfer within 2 weeks). ^13,14 Our audit shows significant delays in the accepting units carrying out initial assessments, with 26% prisoners waiting more than 1 month. Given the extensive development of mental health in-reach services, and the fact that referrals are made by senior psychiatrists, it seems surprising that it has become routine for receiving units to undertake their own assessment, apparently duplicating work. We would advocate a more thoughtful and discriminating approach to considering whether another assessment is needed in each case.

This is the first description in the literature of the wide range of responsible NHS units that a prison in-reach team has to deal with - in this case, the range of services covered by 16 different primary care trusts (who may each commission several services). This may be a particular problem for London prisons. The development of responsive care pathways, good working relationships across services and standards of best practice in such circumstances is highly challenging.

This is also the first attempt to look at delayed transfers by responsible primary care trusts. (We believe that the Ministry of Justice is unlikely to have contributed to the differences between services commissioned by different primary care trusts, because it deals with cases alphabetically and not geographically.) Although the study sample is small and the activity of one prison is described, clear differences in the performances of the services commissioned by various primary care trusts were identifiable. It is a common experience for those working in prisons to feel their heart sink when they discover that the address of a prisoner requiring hospital treatment lies in the catchment area of a poorly performing primary care trust. We discovered that the services commissioned by one Inner London primary care trust were responsible for 58% of all delayed (>3 months) transfers, whereas the services commissioned by the neighbouring primary care trust (covering a similarly morbid borough) were able to admit all prisoners within 3 months.

Is there a postcode lottery?

We believe our data indicate that there is a postcode lottery for prisoners requiring hospital transfer. How can this be addressed? Although there is now much clearer national guidance, ¹⁴ this has not become embedded within primary care trusts. There appears to be a systemic bias against prisoners in some areas, perhaps for understandable reasons. Many NHS in-patient units are routinely expected to run at 100% bed occupancy, making timely responsiveness for prison referrals, particularly for slow-stream forensic units, almost impossible. A hospital's performance in respect of speedy transfer of prisoners does not bring with it any financial incentives, and indeed may be more costly, encouraging a policy of delay. The Ministry of Justice could make more use of its powers to direct prisoners to hospital. Prison mental health in-reach teams have a unique helicopter perspective on the functioning of a very wide range of NHS services, and better advantage could be taken of this to monitor and improve performance.