Through the keyhole - (still) trying to understand Transforming Care from the statistics

The Transforming Care policy/programme, led by NHS England, is designed to reduce the number of people with learning disabilities and autistic people in specialist inpatient services, and to increase and improve the support people get such that the option of putting people into inpatient services does not arise. Transforming Care as a programme, with its predecessors, has been going since 2012, and is due to wrap up as a programme at the end of March 2019. The effects of the Transforming Care programme should by now be visible in the statistics. As @MarkNeary1 has rightly pointed out, each cold number represents people, and we need to keep this in mind as our eyes glaze over at the numbers to follow. I also want to say hats off to @NHSDigital, who have unobtrusively been working to collect sensible information (no mean feat in the circumstances) and have been steadily improving the information they release.

I’ve done blahg after blahg on what the statistics might be telling us about how Transforming Care is going and I’m afraid this is going to be another one (there may well be more to follow…) that basically repeats what I’ve already said with updated information. This one is about the overall numbers of people in inpatient services.

There are currently two main sources of statistics about people with learning disabilities and autistic people in inpatient services, updated on a monthly basis by @NHSDigital. The first of these are monthly statistics from the Assuring Transformation collection http://content.digital.nhs.uk/article/7860/Reports-from-Assuring-Transformation-Collection . For these statistics, health service commissioners (both CCGs and NHS England, who themselves commission inpatient services at the secure end of things) report every month on how many people with learning disabilities and autistic people are in inpatient services, and various aspects of what’s happening to people in these inpatient services. The second are Mental Health Services monthly statistics (MHSDS) http://content.digital.nhs.uk/mhldsreports . This information comes from NHS and independent sector mental health service providers, and (among other things) reports how many people with learning disabilities or autistic people have been in inpatient services that month – inpatient services can include specialist learning disabilities inpatient services, but also includes mainstream mental health inpatient services.

Do these different sources of information tell us the same story about what’s happening with Transforming Care? Well, let’s go…through the keyhole. First up, what kind of an inpatient service system does Assuring Transformation show us?

The most obvious question is what’s happening to the number of people with learning disabilities and autistic people in inpatient services? Is it decreasing at the rate specified in the Transforming Care programme? The Assuring Transformation dataset gives a set of total numbers of people in inpatient services that does seem to be steadily reducing over time, from 2,775 people in March 2016 through 2,710 people in September 2016, 2,605 people in March 2017, to 2,445 people in September 2017. This would be a reduction of 11.9% in 18 months, and some people in NHS England and journalists quote this and similar figures to show the scale of progress and to say that fewer than 2,500 people are now in inpatient units. If you include the dark purple bars in the graph below, this is what you see.

However, to claim this scale of reduction is at best mistaken and at worst mendacious. As I mentioned above, this information comes from commissioners. As well as reporting figures at the end of every month, they can also update their figures from previous months to include people they hadn’t known about at the time. The longer back in time the statistics are, the more time commissioners have had to update their figures. And these updates add quite significant numbers of people (this is what the dark purple bars are). For example, the statistics for March 2016 reported in September 2017 include 2,615 people reported at the time, and an additional 160 people reported in later updates. The September 2016 statistics include 2,565 people reported at the time, and an additional 145 people reported in later updates. Obviously, by the time you get to September 2017 statistics, there has been no time for commissioners to add updated figures later on, so comparing March 2016 (with 18 months of updates) with September 2017 (with no time for updates) is not comparing like with like. Adding 160 people to September 2017 figures (around the number of people that seem to be added retrospectively) would give a figure of 2,605 people in inpatient services. On these figures there has still been a reduction in the number of people in inpatient services from March 2016 to September 2017, but this reduction is 6.1% rather than 11.9%.

 The number of people in inpatient services reported in the Assuring Transformation data collection is also much smaller than the number of people reported in the MHSDS statistics. The MHSDS statistics in their current format haven’t been going very long and they take longer to collate than the Assuring Transformation data, so I can’t compare the information across the two datasets over the whole time period I’m looking at. But comparing them at the most recent possible point in time, August 2017, shows big differences between the two datasets.

Assuring Transformation reports 2,465 people with learning disabilities and autistic people in inpatient services at the end of August 2017, whereas the MHSDS reports 3,265 people with learning disabilities and autistic people in inpatient services at the end of the same month. And this is only a snapshot of people in inpatient services at any one point in time – over the course of a year, how many people with learning disabilities or autistic people pass through an inpatient service? The only certainty is that it’s way more than 2,500 people.  

I think this big discrepancy also points to important differences across the two sets of information that are important to understand.

In contrast to Assuring Transformation, where the information comes from commissioners, the MHSDS statistics are from mental health service providers, which seem to be much better at identifying people with learning disabilities and autistic people in mainstream mental health inpatient services. For example, according to MHSDS statistics for August 2017, less than half of people with learning disabilities and autistic people in inpatient services at the end of June were in learning disability inpatient services (1,135 people; 34.6%). More people (1,420 people; 43.2%) were in adult mental health inpatient services, with other people in children and adolescent mental health or paediatric inpatient services (80 people), older people mental health inpatient services (130 people), or in non-mental health wards (15 people). The type of inpatient service was unknown for 515 people (15.7%).

The two datasets also report very different patterns of use of these inpatient services. In the Assuring Transformation dataset, only 10 people had been admitted and discharged within the calendar month of September 2017. In the MHSDS statistics 1,250 people had been admitted and discharged within the calendar month of August 2017. Of these 1,250 people admitted and discharged, for 425 people this was for the purpose of ‘respite care’ (in a mental health inpatient service???).

So, the Assuring Transformation statistics seem to be missing out a huge number of people with learning disabilities and autistic people who are using mainstream mental health inpatient services, often for very short periods of time. Because this information is very recent, we don’t know whether the number of people with learning disabilities and autistic people in mainstream mental health inpatient services has increased, decreased or stayed the same alongside the Transforming Care programme. We also don’t know how well mainstream mental health inpatient services are working for people, compared to ‘specialist’ learning disability inpatient services. Finally, we don’t know if the circumstances of the people going to mainstream mental health inpatient services are different in some way to those people who end up in learning disability inpatient services.

In contrast to this, it also seems that the MHSDS may be underestimating the number of people with learning disabilities and autistic people in specialist learning disability inpatient services. Why would commissioners be identifying more people than service providers? In the MHSDS, some of the big independent sector service providers like St Andrews are saying that there are substantial numbers of people with learning disabilities and autistic people in wards not identified as learning disability wards, such as children and adolescent mental health wards, adult mental health wards and older people’s mental health wards. There are also some people who independent sector service providers don’t seem to be identifying as people with learning disabilities or autistic people at all, where service commissioners are identifying them (there are around 500 more people in independent sector inpatient services identified in the Assuring Transformation dataset than in the MHSDS). It’s very hard for me to get my head round what is going on here. Are some people with learning disabilities and autistic people being put in places that do not ‘count’ for Transforming Care purposes, but that are still inpatient services?

Whatever is happening, it is clear that there are many more than 2,500 people with learning disabilities or autistic people using inpatient services of various kinds, particularly if you look at more than snapshots. It’s also obvious that the numbers in Assuring Transformation do not show large falls in the number of people in inpatient services, and that Assuring Transformation information isn’t taking into account what’s happening to people in general mental health inpatient services. It feels to me like this isn't even the end of the beginning...