Tuesday, 29 July 2014

How low can you go? How many people with learning disabilities 'should' be in specialist inpatient services?

How low can you go?

In response to the continuing, and continuing, and continuing, and continuing call for the closure of assessment and treatment units, many voices have quite reasonably pointed to the diversity and complexity of inpatient services that have been lumped under the label of ‘Assessment and Treatment Unit’, and have also suggested that some of the services under this label might actually be necessary for the health and safety of some people with learning disabilities (and the health of safety of their communities).

This post tries to set out some of this diversity in inpatient services for people with learning disabilities (the complexity is added by the sources of information!), and to ask the question “How many people with learning disabilities ‘should’ be in specialist inpatient services”?

What do we know about the number of people with learning disabilities in England in different types of specialist inpatient service?

Although the information is typically confusing and inconsistent, there are a few sources of information that look at the number of people in inpatient services by the type of service and by the level of security of the service.

There are several different labels for the variety of specialist inpatient services for people with learning disabilities in England, and the information we have on them does not use a consistent set of these labels.

So, a detailed analysis of the 2010 Count Me In Census of specialist inpatient services for people with learning disabilities in England (Glover & Olson, 2012 http://www.improvinghealthandlives.org.uk/projects/ipbch) reported that there were 3,642 people with learning disabilities in these services, broken down into service type using the following labels:
·         1,344 (36.9%) people in assessment and treatment units
·         1,061 (29.1%) people in long-stay inpatient services
·         466 (12.8%) people in NHS campus services
·         362 (9.9%) people in rehabilitation services
·         85 (2.3%) people in short-stay services (less than a year)
·         132 (3.6%) people in respite services
·         192 (5.3%) in ‘other’ types of inpatient service

These numbers say that in 2010 there were still almost 500 people in NHS campus services, which by 2013 are all supposed to have closed – does this account for the drop in numbers of people with learning disabilities in inpatient units from 2010 to 2013?

Anyway, a survey by the Royal College of Psychiatrists (2013) and the 2013 Learning Disability Census have used similar labels to categorise inpatient services for people with learning disabilities in England, as shown in the table below.

Although there are some worrying differences between these two sources of information, a big proportion of inpatient services are forensic services of high, medium or low security, and of course other types of inpatient service can also operate according different levels of security.

Royal College of Psychiatrists 2013 survey (at 80% ‘bed’ occupancy)
2013 Learning Disability Census
Category 1: high, medium and low secure forensic beds
1,914 (60.5%)
1,361 (41.9%)
Category 2: acute admission beds within specialised learning disability units
651 (20.6%)
466 (14.3%)
Category 3: acute admission beds within generic mental health settings
143 (4.4%)
Category 4: forensic rehabilitation beds
498 (categories 4 & 5 combined) (15.7%)
262 (8.1%)
Category 5: complex continuing care and rehabilitation beds
530 (16.3%)
Category 6: other beds including those for specialist neuropsychiatric conditions
10 (0.3%)
97 (3.0%)
Total number of people
Royal College of Psychiatrists (2013) http://www.rcpsych.ac.uk/pdf/FR%20ID%2003%20for%20website.pdf  

The table below shows information from the 2010 and 2013 census on the number of people with learning disabilities subjected to different levels of security in inpatient services in England.
Across all inpatient services for people with learning disabilities, around half of people with learning disabilities are in ‘general’ (no secure) services, with over a third in low secure services, so less than a fifth of people are in medium or high secure services.

Although the Royal College did not provide this information in their report, they did note that around 70% of the people in their Category 1 (forensic beds) were in low secure forensic services (around 1,340 people, vs 574 people in medium or high secure forensic services, consistent with the 2013 figure below).

2010 Count Me In Census
2013 Learning Disability Census
High secure
47 (1.3%)
73 (2.2%)
Medium secure
399 (11.0%)
512 (15.8%)
Low secure
1,247 (34.2%)
1,195 (36.8%)
General ‘no secure’
1,949 (53.5%)
1,470 (45.2%)
Total number of people
2010 Count Me In Census: Glover & Olson (2012) http://www.improvinghealthandlives.org.uk/projects/ipbch

 How many people need to be living in inpatient settings?

As far as I know, only one organisation has tried to directly answer this question.

The Royal College of Psychiatrists (2013) Faculty of Intellectual Disability Psychiatry report on People with learning disabilities and mental health, behavioural or forensic problems: the role of in-patient services http://www.rcpsych.ac.uk/pdf/FR%20ID%2003%20for%20website.pdf  recommends that there should be specialist inpatient services for 6-7 people with learning disabilities per 100,000 total population. Coincidentally, this works out at just over 3,000 people – the number of inpatient beds there already are!

There are a number of reasons (some of which are contained in the report itself) to think that this number is much too high:
·         First, the report points out that this is a substantial reduction from the figure of inpatient services for 14-29 people with learning disabilities per 100,000 total population recommended in Bailey & Cooper (1997). If such a reduction is possible over the last 20 years, why assume that further reductions will not be possible in the future?
·         Second, the report also states “If, in the absence of significantly improved community services, the less restrictive in-patient facilities (categories 2–5) are further reduced because they are all uniformly described as assessment and treatment beds, then many more people will have unmet needs that compromise their mental health and safety. The result of this could be even more people ending up in far more restrictive forensic beds (category 1).” So, if there are significantly improved community services (and I think there is agreement that better community services are urgently needed) does this mean that fewer inpatient services will be needed?
·         Third, the report also points out that people in forensic inpatient services are paid for by specialist commissioners, so the (to my mind perfectly feasible) option of moving people from ‘low secure’ inpatient services to community-based settings is largely blocked by a money transfer issue rather than anything about how people in low secure settings could be best supported in the community.
·         Fourth, information from the 2013 Learning Disability Census shows massive regional variation in how many ‘home’ people with learning disabilities are currently living in inpatient settings around the country. So the average across England is the Royal College recommended 6.0 people per 100,000 total population in inpatient units. However, one region (the South West) only sends 2.8 people with learning disabilities per 100,000 to inpatient units – less than half the Royal College recommended level – showing that it is possible.

It’s also worth noting that providers in the 2013 Learning Disability Census reported that only 141 people ‘required indefinite inpatient care’ because of physical or behavioural needs, with a further 496 people ‘currently not dischargeable because of mental illness’. Compare and contrast with the 1,702 people where the ‘reason’ for the lack of a transfer plan was a ‘clinical decision’ (NHS England, 2014 http://www.england.nhs.uk/ourwork/qual-clin-lead/wint-view-impr-prog/ ).

Using this logic probably gets you to the point of assuming that inpatient services are only really necessary for offenders with learning disabilities in medium or high secure settings, which is currently around 600 people, if high quality community services were in place. Furthermore, depending on the nature of the offence, the risks to self and others, and the success of interventions in these settings, it is possible that some of the 500 people currently in medium secure settings could move into community-based settings over time. Decent community services might also help people with learning disabilities not to become engaged in serious criminal activity in the first place, further reducing the need for these services.

The Mansell swerve

One notable omission from this  post so far has been the revised Mansell report (2007) http://webarchive.nationalarchives.gov.uk/20130107105354/http://www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/@dh/@en/documents/digitalasset/dh_080128.pdf . This is because this report does not provide any specific recommendations on the number or rate of people with learning disabilities and challenging behaviour likely to need specialist inpatient services.

My reading of this is that these inpatient services are so much of a dispreferred last resort that to suggest an ‘ideal’ number of places in them in effect only encourages them (“if you build it, they will come”), when the aim is to see how near to zero you can get.

The extended quote from the Mansell report below shows quite clearly the weight that should be placed on assessment and treatment units in the overall scheme of things.

“The difficulty of bringing people back home once they have been placed a long way away is so great that every effort should be made to avoid such placement. What is needed is a much wider range of options locally so that the individual needs of the person in crisis can be better met. For example, options for respite can be expanded by:
·         funding a short hotel break for the individual or those they live with
·         providing more help at home
·         staying with staff identified as having a particularly good relationship and skills with the individual
·         spending less time in the house during the day
·         using a local house or flat before it is permanently occupied
·         staying in a designated short break care service

Options for intervention can be expanded by:
·         Extra skilled leadership/support for staff (‘getting back on track’)
·         Extra skilled staff (‘extra pairs of hands’ or ‘new pairs of eyes’)
·         Telesupport; phone or visiting checks
·         Extra help to enable access to ordinary settings
·         Stay in a designated assessment and treatment unit

Options for a breathing space can include:
·         Having services available for the individual when needed through proper person-centred planning
·         Staying with staff identified as having a particularly good relationship and skills with the individual while a new property is found
·         Turning property designated to provide individualised short break care into someone’s home (and developing new short break places)
·         Having spare capacity in anticipation of growth”

So, maybe the question is not “How many inpatient places do we need?” but “How low can you go?”


Bailey NM, Cooper SA (1997) The current provision of specialist health services to people with learning disabilities in England and Wales. Journal of Intellectual Disability Research, 41, 52–9.


  1. This has been a lot of really interesting information! I have been trying to learn a lot about people with learning disabilities. My friend's daughter was just diagnosed with one after taking an assessment. http://www.learning-aids.com

  2. Кредити - ипотечни, потребителски, бизнес, жилищни и бързи кредити. Информация за всички предлагани на пазара кредити. Кредитни карти и заеми за физически и юридически лица.