Skip to main content
Creating a new NHS England: NHS England and NHS Digital merged on 1 February 2023. All references to NHS Digital now, or in the future, relate to NHS England. More about the merger.

Publication, Part of

Autism Statistics

Experimental statistics


Data collection

These statistics are produced from the Mental Health Services Data Set (MHSDS) and are published monthly.

The MHSDS is a complex relational data set which collects record-level data on NHS-funded specialist mental health, learning disabilities and autism services. As a secondary uses data set it intends to re-use clinical and operational data for purposes other than direct patient care.

The MHSDS is unique in its coverage, because it covers not only services provided in hospitals, but also in outpatient clinics and in the community, where the majority of people in contact with these services are treated. MHSDS brings together key information from Adult and Children's mental health, learning disabilities or autism spectrum disorder, CYP-IAPT and early intervention care pathway that has been captured on clinical systems as part of patient care.

From April 2019 data, the MHSDS is collected via the Strategic Data Collection Service (SDCS) Cloud service using the MHSDS v4.0 Intermediate Database (IDB). Prior to this, MHSDS data was collected via the Bureau Service Portal.

Unlike the Bureau Service Portal, the SDCS Cloud an internet-facing service that does not require an N3 or HSCN connection, making it easier for providers to submit data and therefore enabling better coverage in published statistics. This new service stores MHSDS data in cloud-based infrastructure.

Detailed guidance is available to support providers in making submissions via the SDCS Cloud. This includes detailed technical and user guidance. New providers must complete a defined process to gain access to the SDCS Cloud. It also uses two-factor authentication as the secure method of confirming user identity using a combination of two different factors.

Data Processing

Data is this release is based on submissions to the OpenExeter Bureau Service Portal. From April 2019 data, the MHSDS data are processed using NHS Digital’s new Data Processing Services (DPS). DPS uses modern technologies and processes to collect, process and access data more efficiently.

Find out more about our Data Processing Services (DPS).

Data validation

MHSDS data is validated in stages.

Firstly, the data is validated at the point of submission, for each provider. If file-level validation checks are not passed the provider will receive a file-level rejection report. If file-level rejection is passed, the data are successfully submitted but the provider will still receive a report containing details of field-level errors and warnings. These can be investigated, corrected and further submissions can be made within the submission window.

For submitted data, NHS Digital publishes two types of data quality reports in this publication series. By publishing data from both ‘primary’ and ‘refresh’ submissions, providers can review issues identified in a ‘primary’ submission and investigate and resolve them for a ‘refresh’ submission.

The first type of report is a coverage report. This shows the number of records successfully submitted by each provider, for each data table in the MHSDS. When viewed in time-series format in our Power BI reports, this also provides intelligence on the consistency of submissions and enables outliers to be identified.

The second type of report is a ‘VODIM’ report. This classifies each record into five categories; Valid, Other, Default, Invalid, Missing. NHS Digital is expanding the number of items for which VODIM reporting is available. This file also includes a number of Integrity measures. This information is also available as a Power BI report.


These statistics are disseminated via the NHS Digital website.

These quarterly statistics are based on final data. Data are collected in both ‘primary’ and ‘refresh’ submission windows for each month. The ‘primary’ data are designated as provisional and the ‘refresh’ as final data.


The statistics presented in this publication are currently designated as experimental statistics and, as such, remain under constant review; although major changes to methodologies are made between reporting years in order to preserve time-series.

The Mental Health Analysis team welcomes any comments or feedback on the publication so please send any such communications to with ‘Autism Statistics’ in the subject.

Last edited: 24 July 2020 3:42 pm