Safeguarding statistics and adults with learning disabilities - a quick update

Following some recent conversations on twitter, this post provides a quick update on what national statistics tell us about how many adults with learning disabilities are involved in Section 42 safeguarding processes in England (again NHS Digital produce annual reports about these statistics).

The first thing to say is that much less information is reported publicly about safeguarding specifically for adults with learning disabilities than previously, mainly due to concerns about the quality of the information (these are still officially 'experimental statistics'). This also means that there have been a few changes to how the information has been collected from 2010/11 to 2016/17, so it's hard to draw strong conclusions about changes over time.

The first graph below shows how many adults with learning disabilities (vs how many other adults in total) have been recorded in the safeguarding statistics over time. In 2016/17 there were 14,890 adults with learning disabilities involved in safeguarding enquiries across England, a slight uptick from 14,815 people in 2015/16. Although it's hard to draw strong conclusions about trends over time, in general the number of safeguarding enquiries involving adults with learning disabilities has been going down while the number of safeguarding enquiries involving other adults has been going up. We can see this from the fact that in 2010/11 21% of all safeguarding enquiries involved adults with learning disabilities; by 2016/17 this had dropped to 14% of all safeguarding enquiries.

I will leave it to others who know much more about safeguarding than me to offer interpretations (is that uptick a blip or the start of a more general increase?), but there is one more graph I want to show that speaks to how consistent safeguarding processes are across England. The graph below simply shows the safeguarding rate for adults with learning disabilities per 100,000 adult population in each local authority, ordered from the lowest to the highest rate. The variation in safeguarding rates across local authorities looks extreme to me - why is this, and what consequences does this have for people with learning disabilities and others around them?

Now you see me, now you don't - independent sector services and Transforming Care statistics

I’ve gone on about this subject so much in this blog that I’m sure you’re as heartily sick of it as I am, but once again I’ve been worrying away at statistics about people with learning disabilities and/or autistic people in inpatient services. To try and recap, pithily…

• Exhibit A. The Assuring Transformation dataset is produced and reported monthly by NHS Digital for the NHS England Transforming Care programme, based on health service commissioners’ reports on the number of people in specialist inpatient services for people with learning disabilities and/or autistic people.
• Exhibit B. The newer Mental Health Services Dataset (MHSDS) is also produced and reported monthly by NHS Digital, based on mental health service providers’ reports on the number of people flagged as learning disabled and/or autistic in any inpatient mental health service.
• There is some overlap between the two datasets but also some big differences, with Assuring Transformation recording fewer people overall, and a less transient group of people generally spending long periods of time in specialist learning disability inpatient services, about half of which are in the independent sector. The MHSDS records more people overall, including a more transient group of people spending short periods of time in mainstream NHS mental health inpatient wards (probably).

I think this matters, not just because of my itchy-under-the-skin desire for consistency, but in the real world. NHS England reports its progress on Transforming Care to the world in terms of Assuring Transformation statistics, yet these statistics (like the Transforming Care programme itself) are due to stop at the end of March 2019. After that, we will be left with the MHSDS – what will the picture be then?

Handily, the ever-excellent @NHSDigital have for some time been reported direct cross-tabulations between the Assuring Transformation and MHSDS datasets, stripping out people in inpatient services for ‘respite’ (I know – don’t get me started or this post will be even longer) to reduce this source of inconsistency. I’ve blogged on the general picture before – here I want to focus on the details according to specific independent sector inpatient organisations. What stories do the two datasets tell us about independent sector inpatient services over a one-year time period, comparing October 2016 to October 2017?

The first thing is that there are a lot of them – 47 organisations (which may include more than one inpatient unit) listed in October 2016. The number of organisations listed increased to 58 organisations in October 2017.

The second thing is that most of them are small (listing places for 10 or fewer people in either dataset) – 31 out of the 47 organisations in October 2016 fitted into this very small category. All these 31 organisations were still listed in October 2017, with the addition of 10 more organisations with small numbers of people listed (this list now includes two local authorities; Nottingham City and Wiltshire). A list of these organisations’ names is at the bottom of the post.

This leaves 16 organisations in October 2016 (with an extra one in October 2017) listing more than 10 people on either dataset. This is where I start to get seriously confused, so I’m afraid I’m going to share this confusion with you rather than make sense of things.

The table below shows the next eight organisations up in terms of number of people listed in their inpatient services (less than 50 people). Most of these organisations are listed by commissioners in the Assuring Transformation dataset as specialist inpatient services for people with learning disabilities and/or autistic people, but they are not listing themselves as hosting (I don’t know what the right word is) these people as people with learning disabilities and/or autistic people in inpatient mental health services. What happens when Assuring Transformation data collection stops in March 2019 – will all these people become invisible again and will commissioners (and the organisations themselves) feel no further pressure to reduce their numbers?

One organisation in this group (the Jeesal Akman Care Corporation) has added people in their services to the MHSDS dataset between October 2016 and October 2017 so they will appear beyond the end of the Transforming Programme (although confusingly they record more people in the MHSDS than commissioners record in Assuring Transformation). Even more confusingly, one organisation in this group (Livewell Southwest) has gone for the opposite strategy, with them recording increasing numbers of people as people with learning disabilities and/or autistic people in their inpatient mental health services even though commissioners aren’t counting them as such in the Assuring Transformation dataset. Does this mean that these people are invisible right now to the strictures of Transforming Care?

Name of organisation	Number of people in the service at the end of October 2016 according to…		Number of people in the service at the end of October 2017 according to…
	Assuring Transformation	MHSDS	Assuring Transformation	MHSDS
Equilibrium Healthcare	15	*	5	5
Curocare Ltd	20	*	5	*
Ludlow Street Healthcare	15	*	15	*
St George Healthcare Group	20	*	15	*
Livewell Southwest	*	5	*	25
Cheswold Park Hospital	20	*	20	*
Jeesal Akman Care Corporation	40	*	35	40
Brookdale Healthcare Ltd	35	*	35	*

This leaves the eight organisations in 2016 (nine organisations in 2017) with by far the largest numbers of people, according to either or both datasets. These organisations (some of which seem to be ultimately owned by an even smaller number of companies, and have been embroiled in a number of more or less obscure acquisitions) completely dominate – between them they are reported to host around 90% of all people with learning disabilities and/or autistic people in independent sector inpatient services. The number of people in these organisations overall doesn’t seem to have changed much from October 2016 to October 2017.

Some of these organisations (Priory Group, Lighthouse Healthcare, Danshell Group) are only recorded as hosting people in specialist inpatient services for people with learning disabilities and/or autistic people in the Assuring Transformation dataset (with some big changes over time possibly reflecting acquisitions/sell-offs of particular units?). Why aren’t these organisations (which commissioners clearly consider to be specialist units) registering these services as mental health inpatient services for the purposes of the MHSDS, and what do they consider them to be instead? When the Assuring Transformation data stops being collected in March 2019, will these 245 people become statistically invisible?

The new organisation in 2017 (Elysium Healthcare, partly formed through acquisitions from Partnerships in Care and the Priory Group) has gone for the opposite approach, recording themselves as the providers of inpatient mental health services for 125 people with learning disabilities and/or autistic people in the MHSDS with none of these people recorded by commissioners as in inpatient services according to Transforming Care. Are these all people with learning disabilities and/or autistic people in various forms of mainstream mental health inpatient service run by Elysium, even though they have a Learning Disabilities & Autism division?

Four other organisations record people in both datasets in 2017, although the number of people can vary greatly across Assuring Transformation and the MHSDS. For example, in October 2017 Cygnet were recorded as hosting 75 people in the Assuring Transformation dataset but 110 people in the MHSDS, and St Andrews were recorded as having 200 people in the Assuring Transformation dataset but 305 people in the MHSDS. Cambian Healthcare’s figure go in completely the opposite direction, recording 150 people in Assuring Transformation, 120 people more than the 30 people recorded in the MHSDS.

The illustration of how much these figures are a moveable feast is most starkly shown by the statistics for the final organisation in this table, Partnerships in Care. In October 2016, they recorded 280 people in Assuring Transformation and 390 people in the MHSDS. By October 2017, they still recorded 270 people in Assuring Transformation, but any people in the MHSDS had gone.

Name of organisation	Number of people in the service at the end of October 2016 according to…		Number of people in the service at the end of October 2017 according to…
	Assuring Transformation	MHSDS	Assuring Transformation	MHSDS
Priory Group Ltd	50	*	95	*
Lighthouse Healthcare	70	*	55	*
Cygnet Healthcare	75	115	75	110
Huntercombe Group	80	80	95	100
Danshell Group	95	*	95	*
Elysium Healthcare Ltd	n/a	n/a	*	125
Cambian Healthcare Ltd	135	30	150	30
St Andrews Healthcare	205	305	200	305
Partnerships in Care Ltd	280	390	270	*

Looking at this makes me feel distinctly wobbly. It seems clear to me that who gets included in the statistics is to a certain extent the end result of tactical decisions being made by commissioners and independent sector services, and that the services in particular can make decisions that change things quite drastically. If the MHSDS is to become the only source of information on the number of people with learning disabilities and/or autistic people in independent sector inpatient services after March 2019, then this needs to be sorted out urgently.

As it stands, the number of people with learning disabilities and/or autistic people in independent sector inpatient units (dominated by a very small number of organisations) is larger than either the Assuring Transformation or MHSDS datasets describe singly. In October 2017, I estimate this would be at least 1,365 people (looking across both datasets), rather than the 1,185 people reported in Assuring Transformation and the much lower 745 people reported in the MHSDS.

My worry is that over time more and more people in inpatient services (or services that might feel like inpatient services, like re-registered or newly built ‘care homes’ on the sites of existing hospitals) will become invisible in any national statistics. Will these people then be quietly forgotten about in ‘business as usual’ - until the next scandal?

Organisations listing fewer than 5 people in inpatient services (listed by commissioners or the provider):

• October 2016: Partnerships in Care (Hull); St Magnus Hospital; Turning Point (Manchester); The Woodhouse Independent Hospital; Vista Healthcare Independent Hospital; The Lane Project; Alternative Futures Group; Vision Mental Healthcare; Eden Supported Living Ltd; The Atarrah Project Ltd; Coed Du Hall; Choice Lifestyles; Castlebeck Care Teesdale; St Matthews Healthcare; Community Links (Northern) Ltd; Vocare; Making Space; City Healthcare Partnership CIC; Cambian Ansel Clinic Nottingham; Navigo; Virgin Care Ltd; Woodside Hospital; Alpha Hospitals; Modus Care; Breightmet Centre for Autism; Mental Health Care (UK) Ltd; InMind Healthcare; Glen Care; Turning Point; Raphael Healthcare Ltd; Care UK; .
• October 2017: All the above, plus: Shrewsbury Court Independent Hospital; Young Persons Advisory Service; Northorpe Hall Child & Family Trust; Newbridge Care Systems Ltd; Cambian Childcare Ltd; Youth Enquiry Service (Plymouth) Ltd; John Munroe Hospital; Here; Nottingham City Council; Wiltshire Council.

Transforming Care - readmissions update

This blogpost is updating a post I did a few months ago about how many people were being admitted to inpatient units for people with learning disabilities and/or autistic people. In the light of the experiences of people like Eden, who has ended up being sent back to an inpatient unit only recently after leaving several years spent in them, I want to see if the statistics can tell us anything about people going back into inpatient units (otherwise known as readmissions).

This blogpost uses information from the Assuring Transformation dataset, which is updated monthly by @NHSDigital. When reading this blogpost, it’s worth remembering that Assuring Transformation statistics are submitted by commissioners, who tend to focus on people in specialist learning disability inpatient services who spend relatively long periods there. Another dataset, the Mental Health Services Dataset (MHSDS), focuses more on people with learning disabilities and/or autistic people in shorter-term general mental health inpatient services – unfortunately I can’t see any readmission information reported there.

Every month, the Assuring Transformation statistics report how many people have come into an inpatient unit, according to commissioners. The graph below adds these together in six-month blocks over two and a half years (July-December 2015 through to July-December 2017) to see whether there are any changes over time.

  

What do I see in this graph?

First, the overall number of admissions to inpatient units doesn’t seem to showing a clear downward trend over time. The overall number of admissions for July-December 2017 (990 people in six months) is down from figures throughout late 2016 and early 2017, but is still higher than the figures for two years earlier (965 people July-December 2015).

Second, a consistent quarter of ‘admissions’ are actually people being ‘transferred’ from another inpatient unit, a proportion that is remarkably consistent from July 2015 to December 2017. In July-December 2017, this was 25% of admissions, representing 250 people.

Third, nearly one in five people (19% - 180 people) admitted to these inpatient units in July-December 2017 had previously been in an inpatient unit relatively recently. Forty five people (5%) had been out of an inpatient unit for less than 30 days before being re-admitted to an inpatient unit. A further 135 people (14%) had been out of an inpatient unit between 1 month and 1 year before being re-admitted to an inpatient unit. If anything, the proportion of people being readmitted might be increasing over time (it was 14% of people admitted in July-December 2015).

Unfortunately, from the statistics we don’t know any more about the circumstances of people being readmitted to inpatient units. There is information on the ‘source of admission’ - where all people admitted to inpatient units have come from (those admitted for the first time, those readmitted, and those ‘transferred’). The graph below shows this information in the same six-month blocks as the first graph, going back to January 2016.

Like the first graph, where people are coming from before being admitted to an inpatient unit seems to be pretty static over the two years.

In July-December 2017, over two fifths of people (44% - 440 people) were admitted from their ‘usual place of residence’ – this is defined as including living with family, supported housing/living, sheltered housing (as long as there is no ‘health’ element to the support provided), and having no fixed abode (the impact of homelessness on people with learning disabilities and/or autistic people is pretty invisible generally and needs urgent attention).

Fewer people (7% - 65 people) were admitted to an inpatient unit direct from residential care – we don’t know if any of these were people living in places that had been re-registered with the Care Quality Commission from hospital units to residential care homes.

As is to be expected from the ‘transfer’ information in the first graph, substantial numbers of people (120 people – 12%) were ‘admitted’ from an ‘other hospital’ (defined as an NHS or non-NHS hospital ward specialising in mental health and/or learning disabilities) and a further 45 people (5%) were admitted from an NHS ‘secure forensic’ service.

A small but consistent proportion of people (5% - 45 people) were admitted from a ‘penal establishment’ – this is defined as including prisons, young offenders institutions etc, but also police stations and police custody suites.

Well over a quarter of people (29% - 290) were admitted from ‘acute beds’. I looked at the definition of this as I wanted to see if it included mainstream mental health inpatient beds in general hospitals. It turns out it doesn’t – these are people being admitted to inpatient units direct from ‘wards for general patients or younger physically disabled people, or accident and emergency’. My money is on a lot of people going direct from A&E to inpatient units.

Not sure there’s a grand conclusion to be drawn from these statistics, but there’s enough to worry me. With a year to go until the end of the Transforming Care programme, and no national strategy in sight for people with learning disabilities, it at least looks like the Transforming Care is struggling on its own terms. What would I expect to see from these statistics if Transforming Care was meeting its goals? I would certainly expect to see the number of people being admitted to inpatient units going down if pre-admission Care and Treatment Reviews and other measures were having an effect. If people were in the right place for the minimum period of time, then I would also expect to see the number of people being transferred between inpatient units dropping. And if people were leaving inpatient units to go to places with the right support, then the number of people being readmitted to inpatient units should also be dropping.

Having said that, it’s obvious that the Transforming Care programme is trying to swim against a pretty hefty tide – a developing (developed?) catastrophe in education for children with learning disabilities, ever higher hoops to be jumped for more thinly sliced social care support, and signs that community health services generally are disinvesting from supporting people with learning disabilities at any age. Why do people like Eden have to be the canaries signalling imminent explosion in this particular coal mine?

Dismantling the right support

Today (6^th March 2018) @NHSBenchmarking held an event releasing the findings from their most recent round of data collection concerning NHS/health services for people with learning disabilities. This is a project that’s been going for a few years, where volunteer organisations (mainly NHS Trusts) across the UK help NHS Benchmarking collect information about health services for people with learning disabilities. I think 47 organisations took part in 2015/16, and 49 organisations took part in 2017.

I think this project is really important, because it tries to collect information that isn’t available in any other way (although there is a strong case to be made that much of this information should be routinely collected nationally). It also tries to collect information about some of the kinds of community-based health services that the NHS England strategy Building The Right Support says are needed, if people with learning disabilities or autistic people aren’t going to keep being unnecessarily shunted into inpatient units.

Thanks to live tweeters at the event (hashtag #NHSBNLD ) I was able to see some of the findings from this project. NHS Benchmarking also produced a handy summary for 2017 which you can see here, which they also produced in exactly the same format (thank you!) for 2015/16, which I found here and is also shown below. Although the number of organisations taking part changed from 2015/16 to 2017, some of the tweets from the event showed NHS Benchmarking making comparisons over time so I’m assuming this is OK to do.

This post is just a quick, instant reaction to some of the information presented today by NHS Benchmarking. I personally think it makes pretty gloomy reading – rather than building the right community-based support, these figures generally suggest that this support is being dismantled.

Specialist inpatient services

The NHS England Transforming Care programme has set great store on trying to reduce the number of people using inpatient units and reducing the number of these units that exist. The NHS Benchmarking data (from NHS Trusts, I think, so it doesn’t include independent sector organisations that now run inpatient units for half of all people using them in England) shows that the average length of stay for people in a unit has dropped slightly from 244 days to 230 days. This includes some general mental health inpatient units where people stay for much shorter periods of time, and can be contrasted with the national Assuring Transformation data. For January 2018, this dataset reported that people spent an average 987 days in their current inpatient unit, and 1,949 days continuously in inpatient units where they had been transferred directly between units.

More worryingly, the number of places in these units seems if anything to be increasing. In 2015/16, there were 4.8 places in these units per 100,000 of the general adult population – in 2017 this increased to 6.4 places per 100,000 population.

The nursing workforce in these units also changed slightly – in 2015/16 34% of the staff were registered nurses (66% were support workers), compared to 32% of staff in 2017 (and 68% support workers).

Across services for people with learning disabilities, in both 2015/16 and 2017, 14% of the total pay bill was spent on Bank staff. The percentage of the pay bill spent on agency staff increased, from 5% in 2015/16 to 7% in 2017.

Taken together, this suggests a continuing drift towards more inpatient services, with a progressively less skilled and stable workforce working within them.

Community teams for adults

The NHS Benchmarking project also reported some information on community teams for adults with learning disabilities that I don’t think is available anywhere else.

First, the number of contacts with adults with learning disabilities made by community health services has increased, from 2,688 contacts per 100,000 adult general population in 2015/16 to 2,756 contacts per 100,000 population in 2017. But there are signs that community health services are ‘doing more with less’, and are creaking under the strain.

For example, the waiting time from referral to assessment has increased for ‘routine’ referrals from 34 days in 2015/16 to 41 days in 2017. As with inpatient services, the skills of these community teams are also changing – 61% of staff in these services were registered nurses in 2015/16, compared to 53% of staff in 2017.

Community teams for children

The NHS Benchmarking project also reports a couple of bits of information on community services for children with learning disabilities, and here the trends look pretty catastrophic. In 2015/16 there were 2,289 contacts with children with learning disabilities made by community health services per 100,000 general child population – a figure already lower than the equivalent for adult teams. By 2017 this had dropped to 1,471 contacts with children per 100,000 population, a drop of over a third (36%). Over the same year, the average waiting time for a routine appointment for children with learning disabilities increased from 32 days to 72 days.

Dismantling the right support

The ambition and work of the NHS England Transforming Care programme (due to finish in a year’s time), particularly in terms of building decent community-based support services for people and families, is coming up against the brute reality of disinvestment and cuts to exactly the types of services specified as needed in Building The Right Support. These cuts aren’t confined to one year either – the NHS Benchmarking report for 2016 found that in two years from 2014 to 2016 the number of service contacts with people with learning disabilities had dropped by 18% and the spend on these services had dropped by 23%. These cuts seem particularly savage in services for children with learning disabilities, which is where, if anything, support needs to be front-loaded.

From an inadequate base, the right support is being further dismantled.