Why don't NHS England want to know about over 1,000 people in inpatient services?

NHS England are doing another consultation of changes they are making to the information they collect in the Assuring Transformation dataset about the number of people with learning disabilities and autistic people in inpatient services. It's for data gatherers and is not accessible https://nhs-digital.citizenspace.com/standards-assurance/assuring-transformation-data-collection-v4/ 

This consultation describes the Assuring Transformation (AT) dataset as the 'definitive source of information on inpatient numbers and activity'. This short blogpost is me, once again, with a reminder that this is not the case. There is another dataset routinely collected by mental health service providers, called the Mental Health Services Dataset (MHSDS), that is much more comprehensive in its coverage. I'll try and show why this matters by just looking at figures at the end of May 2022 (the most recent information we have).

At the end of May 2022, AT reported that there were 2,075 autistic people and people with learning disabilities in inpatient services. MHSDS reported there were 3,205 people, over 50% and 1,000 people more than the AT dataset. If you're redesigning a national plan, would you not be interested in these extra 1,000 people? 

If you look at the figures for specific provider organisations, the biggest discrepancies in reporting are in the organisations with the most people with learning disabilities and autistic people in their inpatient services. So, here's a list of the organisations where there's a discrepancy of 50 or more people between the AT and MHSDS datasets - many of these will be names you recognise:

Cygnet Healthcare Ltd.    AT - 195 people     MHSDS - 360 people 

(so 165 autistic people and people with learning disabilities in their inpatient services are not visible in the AT dataset, and presumably invisible to the commissioners that provide the data for AT)

Elysium Healthcare.        AT - 160 people    MHSDS - 235 people (75 people invisible to AT)

Partnerships in Care Ltd. AT - 95 people       MHSDS - 170 people (75 people invisible to AT)

Priory Group Ltd.            AT - 130 people    MHSDS - 65 people

(this is one of a small number of organisations where there are more people recorded in the AT dataset than in the MHSDS, so presumably there are people in inpatient services according to AT that don't count as mental health services according to the MHSDS???)

St Andrews Healthcare.   AT - 110 people    MHSDS - 170 people (60 people invisible to AT)

The Huntercombe Group.  AT - 90 people    MHSDS - no people (again, seems very odd)

Birmingham and Solihull Mental Health NHS Foundation Trust.

                                     AT - 15 people        MHSDS - 75 people (60 people invisible to AT)

Bradford District Care NHS Foundation Trust.

                                     AT - no people        MHSDS - 80 people (80 people invisible to AT)

Cumbria, Northumberland, Tyne and Wear Foundation NHS Trust.

                                     AT - 75 people       MHSDS - 145 people (70 people invisible to AT)

Greater Manchester Mental Health NHS Foundation Trust.

                                     AT - 20 people       MHSDS - 80 people (60 people invisible to AT)

Mersey Care NHS Foundation Trust.

                                     AT - 95 people    MHSDS - 145 people (50 people invisible to AT)

Nottinghamshire Healthcare NHS Foundation Trust.

                                     AT - 80 people    MHSDS - 155 people (80 people invisible to AT)

For some of the NHS Trusts in particular, some of this discrepancy is likely to be because they run generic mental health inpatient services, where people are usually inpatients for relatively short periods of time, and who the MHSDS dataset is much better at picking up than the AT dataset. And for areas looking to reduce reliance on long-stay 'specialist' inpatient services, this might be a deliberate and potentially defensible strategy. But this is highly unlikely to be the case for the private sector organisations on this list. What is going on there, and when the discrepancies between the two datasets have been known for years (and complained about repeatedly by the National Audit Office), why hasn't NHS England got this sorted out?

And meanwhile, over 1,000 people with learning disabilities and autistic people in inpatient services, at a stroke of a computer key, are rendered invisible.