New statistics on restraint, seclusion and segregation - what do they tell us?

The ‘restrictive interventions’ (or more bluntly restraint, seclusion and segregation) being used by inpatient services on people with learning disabilities or autistic people has been gaining some public and political attention, thanks to the doughty work of many campaigners and journalists. The Secretary of State for Health (and Care – remember what that is?) has commissioned the Care Quality Commission to “review the use of restraint, prolonged seclusion and segregation for people with mental health, a learning disability and/or autism”.

Before today (17 January 2019) there have been no publicly available statistics on what inpatient units are doing to people with learning disabilities or autistic people in inpatient units in terms of ‘restrictive interventions’ since information collected from the last Learning Disability Inpatient Census in September 2015 (see this blogpost for details). Recently, both radio (File On 4) and TV (Sky News) gained and made public data from NHS Digital that it was already routinely collecting on these ‘restrictive interventions’ (see this blogpost for more details of the data obtained by File On 4). 

Today, as part of its monthly digest of statistics concerning people with learning disabilities and autistic people in inpatient units, NHS Digital included for the first time additional tables about the ‘restrictive interventions’ experienced by people with learning disabilities and autistic people in these places. This blogpost takes a first quick look at some of the information from these tables (taken from the LDA Monthly Statistics from MHSDS – October 2018: Reference Tables). There is undoubtedly lots more in these tables that I’ve missed, and also it’s more than possible that I’ve made some errors of interpretation/basic maths, so please do say if I’ve got something wrong and I’ll correct it.

So, where do these statistics come from? They come from the Mental Health Services Dataset (or MHSDS), which is collected from mental health service providers (both NHS and independent sector) about everyone using their inpatient mental health services. For a while now, information has been flagged about people with learning disabilities and/or autistic people in these services and reported monthly alongside Assuring Transformation data (a dataset collected by service commissioners that NHS England tend to report when discussing the trajectory of Transforming Care). The coverage of people and services in the Assuring Transformation dataset and the MHSDS overlaps but is not the same:

• the MHSDS does not include everyone in the Assuring Transformation dataset because some big ‘specialist’ inpatient services don’t report to the MHSDS (for reasons that escape me https://chrishatton.blogspot.com/2018/03/now-you-see-me-now-you-dont-independent.html )
• the MHSDS includes a lot of people with learning disabilities and/or autistic people who are in generic mental health inpatient units (rather than specialist learning disability units), often for short periods of time. For example, the MHSDS reports that at the end of September 2018 there were 3,625 people with learning disabilities or autistic people in some form of inpatient unit/ward. During the month of October, 445 people were admitted and were still an inpatient by the end of October, 500 people there at the beginning of the month were discharged at some point in the month, and 900 people were both admitted and discharged within the month of October. So by the end of October, there were 3,575 people in some form of inpatient unit/ward.

Among the many, many things collected in the MHSDS, information is collected from service providers on every instance of ‘restrictive intervention’, broken down into:

·        Physical Restraint (prone);

·        Physical Restraint (excluding prone);

·        Chemical Restraint;

·        Mechanical Restraint;

·        Seclusion;

·        Segregation;

·        a category where a restrictive intervention has been recorded but the specific type of intervention has not been recorded.

Helpfully, NHS Digital report this information in two ways; by the number of people who have been subject to a restrictive intervention at least once in the month of October 2018; and by the number of restrictive interventions that have been used in the month of October (because people can be subject to multiple instances of restrictive interventions).

I’m going to start by going through the statistics on the number of people with learning disabilities or autistic people who have been subject to at least one restrictive intervention through the month of October 2018. In total 435 people were recorded as being subjected to restrictive intervention at some point in the month of October. In terms of specific types of restrictive interventions (these numbers will add up to more than 435 as the same people could have experienced more than one type of restrictive intervention):

·        100 people were subject to prone physical restraint

·        275 people were subject to another form of physical restraint

·        65 people were subject to chemical restraint

·        20 people were subject to mechanical restraint

·        80 people were subject to seclusion

·        5 people were subject to segregation

·        170 people were subject to a restrictive intervention where the form of that intervention was not recorded

I think, taking into account people going into and leaving inpatient units throughout the month, that in total 4,970 people with learning disabilities or autistic people were in some form of inpatient unit at some point in October (3,125 who were there at the start of October and stayed, plus 445 people who were admitted during the month of October and stayed, plus 500 people who were there at the start of October and were discharged during the month, plus 900 people who went in and out during October – I feel I need to show my working so you can tell me if/where I’ve made a mistake).

[As an aside, this number of people is a lot bigger than Assuring Transformation (and doesn’t even include some people in the Assuring Transformation dataset), and personally I think should be an important part of any programme designed  to improve support for people with learning disabilities or autistic people in distress. As well as people who are stuck in terrible conditions for years, there are a lot of people in and out of generic mental health inpatient units and we understand very little about how/if that’s working for people].

So, overall I think 8.8% (435 out of 4,970 people) of people with learning disabilities or autistic people spending some time in an inpatient unit/ward in October 2018 were recorded as subject to some form of restrictive intervention. This information is also broken down by age group, so I have tried the same calculations for each age group listed, showing the percentage of people in each age group who were subject to at least one restrictive intervention:

·        13.7% of people aged under 18 (65 out of 475 people in inpatient units/wards)

·        11.2% of people aged 18-24 (105 out of 940 people)

·        8.5% of people aged 25-34 (115 out of 1,350 people)

·        6.9% of people aged 35-44 (60 out of 870 people)

·        5.3% of people aged 45-54 (35 out of 660 people)

·        6.4% of people aged 55-64 (25 out of 390 people)

·        8.8% of people aged 65+ (25 out of 285 people)

As I mentioned earlier, NHS Digital also report the number of restrictive interventions used on people with learning disabilities or autistic people during October 2018. In total 2,700 instances of restrictive intervention were recorded over the month, broken down according to the following types of restrictive intervention:

·        285 instances of prone physical restraint

·        1,485 instances of another form of physical restraint

·        225 instances of chemical restraint

·        65 instances of mechanical restraint

·        190 instances seclusion

·        5 instances segregation

·        In 445 instances of restrictive intervention (16.5% of all recorded instances), the form of that intervention was not recorded

On average, among those people in inpatient units/wards subject to restrictive intervention, people experienced 6.2 instances of restrictive intervention per person during October 2018 (2,700 restrictive interventions among 435 people). By age group:

·        People under 18 experiencing restrictive intervention were subject to an average 11.7 instances of restrictive intervention per person (765 restrictive interventions among 65 people)

·        People aged 18-24 experiencing restrictive intervention were subject to an average 5.5 instances of restrictive intervention per person (580 restrictive interventions among 105 people)

·        People aged 25-34 experiencing restrictive intervention were subject to an average 5.9 instances of restrictive intervention per person (675 restrictive interventions among 115 people)

·        People aged 35-44 experiencing restrictive intervention were subject to an average 5.7 instances of restrictive intervention per person (340 restrictive interventions among 60 people)

·        People aged 45-54 experiencing restrictive intervention were subject to an average 4.3 instances of restrictive intervention per person (150 restrictive interventions among 35 people)

·        People aged 55-64 experiencing restrictive intervention were subject to an average 4.0 instances of restrictive intervention per person (100 restrictive interventions among 25 people)

·        People aged 65 or older experiencing restrictive intervention were subject to an average 3.6 instances of restrictive intervention per person (90 restrictive interventions among 25 people)

NHS Digital also provide statistics according to individual provider organisations, categorised according to whether they are NHS or independent sector. For obvious reasons I thought this would be worth looking at in more detail. But when I started looking, it became clear to me that a lot of provider organisations are simply not reporting data on restrictive interventions properly, and that this lack of proper reporting is likely to be more common in the independent sector.

For example, there are 14 independent sector providers listed in the MHSDS for October 2018 (although as we know from Ian Birrell and others, many of these different ‘brands’ are ultimately owned by a small set of mainly U.S. companies/hedge funds). Only two of them (Cygnet and St Andrews) record 5 or more instances of restrictive intervention (of any type) happening in the month of October 2018 (below 5 and the numbers are suppressed by NHS Digital to prevent individuals being potentially identifiable). So the following organisations (I’ve just listed the ones with more than 5 people in them at the end of October 2018) are publicly reporting that, in effect, no restrictive interventions were visited on any people with learning disabilities or autistic people in their inpatient units – and there were 1,345 people in independent sector inpatient units at the end of October 2018:

·        CAS Behavioural Health Ltd (160 people at the end of October)

·        Elysium Healthcare (355 people)

·        Jeesal Akman Care Corporation Ltd (50 people)

·        LiveWell SouthWest (25 people)

·        Partnerships in Care Ltd (220 people)

·        Priority Group Ltd (10 people)

·        The Huntercombe Group (80 people)

Even where an organisation is reporting that restrictive interventions are happening, in some organisations the type of restrictive intervention is very commonly not reported. For example, St Andrews reported their staff doing 810 instances of restrictive interventions in October 2018 – but in over half of these (420 instances, 52%) the type of restrictive intervention was not recorded. Looking down the longer list of NHS Trusts in this MHSDS dataset, I could only find one where this was an issue (South London and Maudsley NHS Foundation Trust, where the type of restrictive intervention was not recorded for all 25 instances of restrictive intervention).

It’s great to see this information being made public by NHS Digital, and over time a more complete picture will build up about the restrictive interventions being applied to people with learning disabilities and autistic people in inpatient units. From this first month of information, I think there are a few things we can say already:

·       A lot of people are subject to restrictive interventions, of multiple types, on multiple occasions, in a short space of time. In the month of October 2018, 8.8% of people with learning disabilities and autistic people in inpatient units/wards experienced some form of restrictive intervention (435 out of 4,970 people). If people were subject to restrictive intervention, on average they were subject to 6.7 instances of restrictive intervention during the month (2,700 restrictive interventions among 435 people).

·       Physical restraint, including prone restraint, is commonly used. The most commonly reported types of restrictive intervention were physical restraint other than prone restraint (1,485 instances of restrictive intervention), prone restraint (285 instances), chemical restraint (225 instances) and seclusion (190 instances). For 16.5% of restrictive interventions (445 instances) the type of restrictive intervention was not recorded, although almost all of these are accounted for by one organisation, St Andrews (where the type of restrictive intervention was not recorded for 420 instances of restrictive intervention in October).

·       Young people aged under 18 are the age group most likely to be subject to restrictive interventions (13.7% of young people in inpatient units/wards) and, if they do, to experience twice as many instances of restrictive intervention (11.7 instances of restrictive interventions per person in October 2018) as other age groups.

·       It seems that some organisations, particularly those in the independent sector, are not properly recording or reporting information on restrictive interventions, or are simply not reporting this information at all.