Restraints used on people with learning disabilities and autistic people in inpatient units through the COVID-19 pandemic - what the statistics say

This blogpost will run through what the publicly available statistics say about what restraints (or 'restrictive interventions') autistic people and people with learning disabilities in mental health inpatient services in England have been subject to through the COVID-19 pandemic to date. My thanks to the Restraint Reduction Network for asking me to talk at their conference this year, prompting me to have a proper look at this information and marshal the many graphs into some sort of argument.

A recent blogpost gives some context on the number of people with learning disabilities and autistic people who have been in inpatient services in England through the COVID-19 pandemic, using the same dataset as the one I will be using in this blogpost. It might be worth you having a quick look at that first, but a couple of headlines from that is that it looks like there were substantial dips in the number of people in inpatient services at times when the COVID-19 pandemic was peaking, with high pressure on health services and nationwide lockdowns. There have been increases in the number of people in inpatient services after these periods, but not to the levels seen before the pandemic. These dips in numbers were proportionally higher for children and young people and people not on formal sections.

The dataset I used in that blogpost and in this one is the Mental Health Services Data Set (MHSDS) - this is a large dataset collected by all NHS-commissioned mental health services in England, and includes information on people flagged in these services as an autistic person or a person with learning disabilities. NHS Digital extracts the information specifically related to people with learning disabilities and autistic people in any mental health inpatient service every month, including both the 'specialist' inpatient services that have largely been the focus of Transforming Care/Building The Right Support, and general mental health inpatient services. Providing information for the MHSDS is mandatory for NHS funded service providers, including independent sector providers.

What does the MHSDS say about restrictive interventions?

The MHSDS includes information on both the number of people subject to any restrictive intervention in any single month, and the number of restrictive interventions people have been subject to in any single month. It also provides information on a range of specific types of physical restraint, and on mechanical restraint, chemical restraints, seclusion and segregation.

What do the MHSDS statistics say about restrictive interventions overall? The graph below shows the number of people with learning disabilities and autistic people in inpatient units who have been subject to at least one restrictive intervention in each month, from January 2020 through to August 2021. There are particular spikes in May and July 2020, which I'll come to later, but the overall picture is of fluctuating numbers of people subject to restraints over time.

Because the number of people in inpatient services also fluctuates over this time period, we can also look at what percentage of people in inpatient units in each month are subject to at least one restrictive intervention. This is what the graph below shows - again there are the spikes in May and July 2020, but overall the trend looks to be increasing over time. In January 2020, 7.5% of people were subject to at least one restrictive intervention. By August 2021 this was 9.1% of people.

As well as the number of people subject to restrictive interventions, we know that people can be subject to multiple restraints, so the graph below shows the total number of restrictive interventions that autistic people and people with learning disabilities in inpatient services were subject to each month. Again, we see substantial fluctuations month by month throughout the pandemic, with potential dips when there were fewer people in inpatient units.

To take this into account, the graph shows the average number of restrictive interventions for every person that was subject to at least one. This again shows fluctuations over time, with a potential increasing trend until the summer of 2021. On average, people were subject to around 10 restrictive interventions in every month - one every three days.

The publicly available information also allows us to look at restraints that people in different age groups have been subject to. To simplify things a bit I've looked at four age groups: children and young people aged under 18; adults aged 18-24; adults aged 25-35; and adults aged 35+. The graph below shows the percentage of people in inpatient units in each age group who have been subject to at least restrictive intervention in each month. As the graph shows, people aged 18-24 are consistently most likely to have been subject to restraint - even setting aside the May and June 2020 spikes, around 13-14% of people in this age group experience restraint in each month.

Following on from this, how many restrictive interventions are people in different age groups subject to? The graph below shows that children and young people aged less than 18 are consistently most likely to experience more restrictive interventions in each month - in some months more than 20 restrictive interventions per person subject to at least one.

What restrictive interventions are people subject to?

What do the statistic tell us about which types of restraint people with learning disabilities and autistic people in inpatient units have been subject to throughout the COVID-19 pandemic? The graph below shows the number of people who have been subject to different types of physical restraint throughout the time of the pandemic, according to the MHSDS.

The highest numbers of people are subject to four types of physical restraint (which may be increasing over the time of the pandemic): standing restraint, supine restraint, seated restraint and 'other' physical restraint. By August 2021, at least 150 people were subject to each of these types of physical restraint (and I don't understand why 'other' restraint should be increasing over time - what types of physical restraint are being put into that mysterious box?). It is also worth noting that prone restraint is still happening and being recorded - 100 people per month experience prone restraint.

How many times are people being subject to physical restraints? The graph below shows the total number of instances recorded of different types of physical restraint. The same four types of physical restraint discussed above are also the ones most commonly used: there are potential decreases over time in the number of times supine and standing restraint have been used, but a definite increase in the recording of 'other' restraint.

We can also look at the same type of information for other forms of restrictive intervention: chemical restraints, mechanical restraint, seclusion and segregation. The graph below shows the number of people subject to these forms of restrictive intervention. Seclusion is experienced by a large and increasing number of people across the pandemic period, up to over 160 people in the month of August 2021. Oral chemical restraint and rapid tranquilisers are each experienced by around 80 to 100 people every month, with the number of people subject to oral chemical restraint more than doubling from January 2020 to August 2021.

In terms of the number of times people have been subject to these forms of restrictive intervention, the graph below shows a similar picture, with high use of seclusion throughout and high and increasing use of oral and rapid tranquiliser chemical restraint over time.

NHS and independent sector inpatient services, and the case of the missing information

Information from the MHSDS about restraints used on people with learning disabilities and autistic people can be broken down by whether the inpatient service is an NHS or independent sector. The two graphs below show the number of people subject to at least one restrictive intervention and the total number of restrictive interventions people are subject to, broken down by NHS vs independent sector provider.

Both graphs seem to show a similar picture. For the number of people subject to at least one restrictive intervention, in NHS inpatient services there is a dip in the first peak of the COVID-19 pandemic then a steady increase ever since. In independent sector services there is the double spike in May and July 2020 I've mentioned already, interrupting a gradual decrease and flattening out over time of the number of people subject to restrictive interventions.

The picture in terms of the total number of restrictive interventions is similar - increasing over time in NHS inpatient services, and (except for the spikes) gradually decreasing over time in independent sector inpatient services.

Is this an accurate picture? The spikes in May and July 2020 (finally, he gets to the spikes!) give us a clue. In these two months, the Cygnet organisation (recorded in the MHSDS in May 2020 as having 360 autistic people and people with learning disabilities in inpatient units) recorded 830 and 775 restrictive interventions respectively. In June 2020, August 2020, and every month since, the number of restrictive interventions is recorded as an asterisk, meaning anything between 0 and 4 restrictive interventions in total for the month. These could mean a complete absence of reporting of restrictive interventions in inpatient services for the whole Cygnet organisation. Elysium (485 people in May 2020), Partnerships in Care (90 people) and Huntercombe (90 people) all report asterisk levels of restrictive interventions throughout the whole pandemic period - St Andrews is the only large (50+) independent sector organisation that reports data to the MHSDS every month on restrictive interventions in inpatient services.

So, the figures on restrictive interventions within independent sector inpatient services are likely to massively undercount the restraints that people are being subject to - even the two months of spikes will be undercounts as they are missing information from some big independent sector organisations.

Summary, and why data matters

So, what do I think the MHSDS information tells us about restraints in inpatient services throughout the COVID-19 pandemic (so far)? A few things:

• Overall, about 9% of people with learning disabilities and autistic people in inpatient services will be subject to an average of 10 restrictive interventions per person in any given month.
• Young adults aged 18-24 are most likely to be subject to restrictive interventions.
• Children and young adults up to 18 years old are moat likely to be subject to multiple restraints.
• Even taking the incompleteness of reporting into account, there is a possible trend towards the increasing use of restrictive interventions over the course of the pandemic.
• Standing, supine, seated and 'other' physical restraints are most commonly reported (at least 150 people are subject to each of these types of physical restraint every month).
• Around 100 people every month are subject to prone restraint.
• An increasing number of physical restraints are being recorded as 'other'.
• Seclusion, oral chemical restraint and rapid tranquilisers are most commonly reported, and possibly increasing over the time period of the pandemic.

Why does data matter?

Accountability - self-advocacy groups are getting together and wanting to use information like this to hold local inpatient services to account. Please take a look at this open letter from My Life My Choice, Sunderland People First, Speak Up, and Sheffield Voices - you can sign it electronically.

Visibility - although the MHSDS has been collecting restraints information for some time, it was only made routinely publicly available after a File on 4 programme reported on restraints information gained through a bitterly contested Freedom of Information process. Without this visibility, any form of scrutiny is much more difficult.

Proactive regulation - CQC inspections of inpatient services are more frequently rating services as Inadequate, only using information on restraints that do not seem to be reported to the MHSDS (see this recent CQC inspection report on Cygnet Views Matlock). If MHSDS data were complete and regularly reported, inspections could be more proactive rather than reactive.

The political uses of data. We need completeness of data collection for a level playing field between organisations that comprehensively report restrictive interventions vs those who don't. I really don't understand how major companies can just not report supposedly mandatory data for months and years on end, with seemingly no consequences.

Policy. It's an obvious point, but if there are national policy ambitions to reduce the use of restraints, you need to have a reasonably complete picture of what people are being subjected to.

That's more than enough - words, graphs, and utter grimness of people's lives in the abstracted form of numbers.

The number of people with learning disabilities and autistic people in inpatient services in England through the COVID-19 pandemic

In this blogpost, I'm going to summarise some of the information collected monthly by NHS Digital about the number of autistic people and people with learning disabilities in inpatient services in England throughout the COVID-19 pandemic (so far). There will be graphs.

I've gone on and on about this plenty of times before, so I'll try and be brief here (see this blogpost for a loooong description of the issues), but NHS Digital publish data from two sources about the number of people with learning disabilities and autistic people in inpatient services.

The first is Assuring Transformation (AT), which is the one the DHSC and NHS England/Improvement like to cite as evidence of the progress of Transforming Care/Building The Right Support. This is information provided retrospectively by NHS commissioners of these inpatient services, and largely (but not exclusively) concerns autistic people and people with learning disabilities in 'specialist' inpatient services assumed to be the preserve of national policy.

The graph below shows the world according to Assuring Transformation, for each month from January 2020 through to August 2021. Pandemic, what pandemic? Assuring Transformation shows the same glacial progress as usual, from 2,305 people in January 2020 through to 2,040 people in August 2021 (this figure will increase by at least 100 people as those reported retrospectively get added). You will notice in this graph there are supposed to be two lines, although in this graph they are so similar as to be merged. The first line records a snapshot of the number of people in inpatient services at the end of each month. The second line records the number of people who have been in an inpatient service at any point in the month, as people can be admitted or discharged within each month. As you can see, the picture provided by Assuring Transformation is of very little movement of people in or out of inpatient services throughout the pandemic. As the BBC and Sky News have recently reported, 100 autistic people and people with learning disabilities have been in inpatient units for 20 years or more.

The second data source reported on monthly by NHS Digital is the Mental Health Services Data Set (MHSDS to its friends). This is collected by all NHS-commissioned mental health services in England, and NHS Digital pull out figures for people in any mental health inpatient service who have been flagged as a person with learning disabilities or an autistic person. The graph below shows the MHSDS figures for the number of autistic people and people with learning disabilities in inpatient services from January 2020 to August 2021, with the Assuring Transformation figures kept in for comparison purposes.

As you can see the picture from the MHSDS is quite different to AT, in a number of ways.

First, the MHSDS records much higher numbers of people with learning disabilities and autistic people in inpatient services than AT: 3,515 people compared to 2,040 people at the end of August 2021, for example. Presumably this is because many people are flagged who are in a wide range of 'mainstream' mental health inpatient units not considered by NHS commissioners to be part of the Building The Right Support programme.

Second, the MHSDS records much higher numbers of people in inpatient units at some point in each month compared to the number of people in inpatient units at the end of each month. This is because according to the MHSDS many autistic people and people with learning disabilities are in inpatient units for very short periods of time (in addition to the people who are in units for years and years). For example, 185 people were admitted and discharged within the calendar month of August 2021 (155 of these people were apparently admitted for the purposes of 'respite').

Third, the MHSDS shows what looks like a big impact of the COVID-19 pandemic on the number of people with learning disabilities and autistic people in inpatient units. At the time of the first peak of the pandemic (with its associated lockdown) the number of people in inpatient units drops quite sharply (with fewer people being admitted into inpatient units for short periods of time). There is only a slight increase after the first lockdown through the rest of 2020, until another drop during the second/third peak of the pandemic in winter 2020. From the spring of 2021 when lockdowns were eased again, the number of people in inpatient units rises, although not to pre-pandemic levels.

Sticking with the MHSDS, did these changes during the pandemic affect some groups of people more than others? The graph below shows the number of people in inpatient services at the end of each month by four age bands (under 18, 18-24, 25-34, and 35+, as children and young people in inpatient services have been a particular concern).

We can see decreases for all age groups through the first two/three peaks of the pandemic, with sustained increases after the end of the winter/spring lockdown of 2021. Because the number of people in different age bands is so different, we can also look at percentage changes within each age band, using January 2020 as the baseline. As the graph below shows, there were proportionally much bigger reductions in the number of children and young people aged under 18 in inpatient services throughout the pandemic (so far), with a possible 'rebound' increase to at or above pre-pandemic numbers of people in the summer of 2021.

I've used a similar format to look at the impact of people by their legal status. Unsurprisingly, the two graphs below show less change in the number of people in inpatient services with Part III or other sections through the pandemic. The number of people under Part II sections reduced in the pandemic peaks/lockdown periods, but have returned to pre-pandemic numbers. The number of people in inpatient units not under section (recorded as 'informal' legal status in the MHSDS, but bearing in mind Lucy Series' point that this isn't an accurate term) dropped considerably in the first pandemic peak/lockdown, stayed at this level through to the spring of 2021, and has not returned to pre-pandemic levels.

One final graph. The graph below looks at the number of autistic people and people with learning disabilities in NHS and independent sector inpatient units (all funded with public money, of course). The number of people in NHS inpatient services dropped sharply during the first pandemic peak/lockdown, and increased to less than pre-pandemic levels soon afterwards. The number of people in independent sector inpatient services seems to decline consistently through to the end of the second/third pandemic peak/lockdown, but increases sharply in the spring of 2021.

It's hard to know what to make of this information, particularly as the COVID-19 pandemic continues and both NHS and social care services in many respects are still operating very differently to how they were before the pandemic. But it is clear that there have been substantial reductions in the number of people with learning disabilities and autistic people in inpatient services at peak pressure points in the COVID-19 pandemic, with a disproportionate impact on children and young people not on a formal section and likely to be in inpatient services for short periods of time (including for 'respite'). The number of people in inpatient services has been increasing again from spring 2021, although the extent of this and how long it will last is unclear. 

Most importantly, as far as I know there is virtually no public information on what it has been like for autistic people and people with learning disabilities in inpatient services through the COVID-19 pandemic, what has happened to people who would before have been admitted to inpatient services, what other forms of support people have had (or not had), and what all this means for people's lives now and in the future.

People with learning disabilities in paid employment in England 2020/21 - what do the social care statistics tell us?

This blogpost is the third in a set looking at various aspects of the social care statistics recently published by NHS Digital for 2020/21, focusing on adults with learning disabilities in England. The first two blogposts gave an overview of trends in social care support and expenditure for adults with learning disabilities, and a more focused look at what kinds of places those adults with learning disabilities getting social care are living in.

This blogpost updates previous posts on what the statistics say about the self/paid employment of working age (aged 18-64 years old) adults with learning disabilities in England. These figures are provided by local authorities every year, and since 2014/15 have only been provided for people getting long-term social care (before 2014/15 it was the larger group of people known to local authorities as a person with learning disabilities, even if they weren't getting long-term social care support). This means that these figures don't include the much, much larger group of adults with learning disabilities who aren't known to local authorities or GPs and who don't get any kind of support related to their learning disability.

It is important to note that these statistics are collected for financial years (April to March), so they will represent what was happening for adults with learning disabilities up to March 2021, through the first two/three peaks of the COVID-19 pandemic in England. It is also really important to say that the quality of this information has been questioned at the best of times – there are massive caveats about the extent to which local authorities would be able to collect this kind of information in the middle of the pandemic, particularly given the complexity of people’s working situations (for example, furlough) which these statistics were not designed to pick up.

Some of the numbers reported here are included in the NHS Digital Adult Social Care Outcomes Framework (indicator 1E, if you're interested), although some of the numbers are more buried in publicly available but less accessible CSV datafiles. 

What do the numbers tell us?

The first graph below shows the number of working age adults with learning disabilities that councils say are in any self/paid employment, broken down by whether people are in employment for 16+ hours per week or less than 16 hours per week (and 1 hour a week can count in these statistics). The gap between 2013/14 and 2014/15 is because of the change in data collection I mentioned earlier, so numbers before and after then can't be compared, and 2008/2009 was the first year that information was collected so it's rather dodgy for that year.

 

According to these figures, in 2020/21 there were 6,870 working age adults with learning disabilities getting long-term social care in any form of self/paid employment. After a couple of years of the number of people in paid/self employment edging towards 8,000, this is now the lowest number of people in paid employment since the first dodgy year of data collection in 2008/09. Consistent across the years is that most of the people in employment (65% of people in 2020/21) are working for less than 16 hours per week.

What do these numbers mean in terms of employment rates? The graph below shows the employment rates for working age adults with learning disabilities over the same time period. The columns show the overall employment rate - in 2020/21 this was 5.1%, compared to 75% for adults generally in March 2021. This is the lowest employment rate for adults with learning disabilities recorded since the statistics were changed in 2014/15.

The lines on the graph show employment rates for men and women with learning disabilities separately. Employment rates for men (the purple line) are consistently higher (5.7% in 2020/21) than employment rates for women (the blue line) (4.4% in 2020/21 with this gender employment gap staying fairly stable over time.

The last graph below sounds an appropriate note of caution about the reliability of information on self/paid employment provided by councils. This shows, for all working age adults with learning disabilities getting long-term social care, whether councils say they are: 1) in self-paid employment; 2) not in employment but actively seeking work (and presumably therefore liable to potential benefit sanctions); 3) not in employment but not actively seeking work. There is also a fourth category, where councils say they don't know the employment status of the person. 

 

Over time, the number of people whose employment status is unknown has decreased rapidly, from 44% of working age adults with learning disabilities in 2014/15 to 21% of people in 2020/21 (21% is still a big chunk, however). It is unclear how councils are making decisions about whether to record someone as actively seeking work or not.

Overall, this update shows, bearing in mind understandable caveats about the quality of the information that could be collected during the COVID-19 pandemic, further reductions in employment rates (from a ridiculously low base) for working age adults with learning disabilities getting long-term social care, with most work being extremely part-time, and a gender employment gap. 

As far as we know, many more than 5.1% of working age adults with learning disabilities want to work. We know that secure, stable paid employment for people with learning disabilities is associated with better physical and mental health, and we know that supported employment is highly cost-effective. Building sustainable, fulfilling paid employment for people with learning disabilities, in our current circumstances of wide-ranging labour shortages, surely has to be an urgent ‘building back better’ priority?