Mental Health Act Statistics, Annual Figures, 2021-22

This document is a background data quality report for the Mental Health Act Statistics, Annual Figures 2021-22 publication.

This publication contains the official statistics about uses of the Mental Health Act (‘the Act’) during 2021-22. Under the Act, people with a mental disorder may be detained in hospital in the interests of their own health or safety, or for the safety of others. They can also be treated in the community but subject to recall to hospital when necessary for assessment and/or treatment under a Community Treatment Order (CTO).

Prior to 2016-17, this publication series was sourced from the KP90 data collection. These data returns were made by organisations in England that are registered to provide Mental Health Services and make use of the Mental Health Act 1983 legislation. These include high security psychiatric hospitals as well as other NHS service providers and independent hospitals. The data were submitted as aggregate data items for each provider via the KP90 collection, which collected this data for the last time in 2015-16.

From 2016-17, these statistics were primarily sourced from the Mental Health Service Data Set (MHSDS), an administrative data source which contains record-level data about the care of children, young people and adults who are in contact with NHS-funded mental health, learning disabilities or autism spectrum disorder services . Organisations submit data to the MHSDS on a monthly basis and this publication is an annual view of Mental Health Act data compiled from those monthly submissions.

The main MHSDS data is supplemented by data from Acute emergency departments. For acute hospitals which are not in scope for MHSDS but do make use of the Act, this information is now collected via the Emergency Care Data Set (ECDS). For 2016-17 and 2017-18 this information was collected via the ‘Annual uses of the Mental Health Act 1983 in English acute trusts’ (Acute) collection. However, as the ECDS is now operational, this should be used to record uses of the Act in hospital emergency departments from 2018-19. Acute providers should therefore ensure that they are completing the relevant fields in the ECDS.

The change in primary data source from KP90 to MHSDS allows us to:

1. Better meet user needs, as it supports more detailed analysis of uses of the Act, incorporating patient demographics and details of their pathways through services. The aggregate collection did not support such detailed analysis.
2. Reduce burden, as the MHSDS is sourced from routine flows of administrative data. The burden of collection is reduced as organisations do not have to complete a bespoke annual return specifically for the Mental Health Act.

Further information about the MHSDS is available. 

Data quality outputs are produced as part of each monthly MHSDS publication, providing detailed information on coverage, validity and integrity of the data. 

This page aims to provide users with an evidence-based assessment of the quality of the statistical output of the Mental Health Act Statistics 2021-22 publication by reporting against the nine European Statistical System (ESS) quality dimensions and principles.

In doing so, this meets our obligation to comply with the UK Statistics Authority (UKSA) Code of Practice for Official Statistics, which states:

“The quality of the statistics and data, including their accuracy and reliability, coherence and comparability, and timeliness and punctuality, should be monitored and reported regularly. Statistics should be validated through comparison with other relevant statistics and data sources. The extent and nature of any uncertainty in the estimates should be clearly explained”.

For each dimension this section describes how this applies to the publication. We will continue to provide clear and comprehensive information about the methods used in our analysis and the quality of the data to assist users in interpreting our reports.

In addition, as the publication is now compiled from administrative data this document also references and signposts to further information about the assurance of the MHSDS. Quality assurance of administrative data is an ongoing, iterative process to assess the data’s fitness to serve their purpose. More information about the ongoing assurance of MHSDS monthly submissions can be found in the Data Quality section of the Mental Health Services Monthly Statistics publication. These statistics have been assessed as Medium Risk and Medium Profile in accordance with the UKSA Administrative Data Quality Assurance Toolkit and this document also highlights appropriate activities relating to the following UKSA assurance practice areas:

1. Operational context & administrative data collection
2. Communication with data supply partners
3. QA principles, standards and checks applied by data suppliers
4. Producer's QA investigations & documentation

A summary of these can be found in Appendix 3.

In addition, this document contains background information about the Mental Health Act including changes to Mental Health law, which will be of interest to readers who are not already familiar with the Act.

People can be admitted to hospital for the treatment of mental health problems on either a voluntary basis as ‘informal’ patients, or on a compulsory basis as ‘formal’ patients. Formal patients are within the scope of this publication; most have been admitted to hospital on a compulsory basis under the Mental Health Act 1983 (known as being ‘detained’ or ‘sectioned’). Some are subject to a Community Treatment Order (CTO) which allows the patient to leave hospital and be treated in the community when this is appropriate. See the ‘Related links’ section of the main report for further statistics on informal patients.

The Mental Health Act 1983 (‘the Act’) is the main Act of Parliament covering the care and treatment of people with mental health problems. It sets out how and when a person can be admitted, detained and treated in hospital without consent. In order to apply it, certain professionals must agree that this must be done because the health or safety of the individual, or that of other people, is at risk. The individual must be considered to have a ‘mental disorder’; the definition of this term was broadened as a result of the 2007 Act.

Mental health law is about balancing the need to detain people in order to protect them or other people from harm and the need to respect peoples’ human rights and autonomy. Whilst the application of the Act means that a person loses certain rights (such as their liberty and refusal of treatment), it also sets out other rights, such as a right of appeal and help from an advocate, and free aftercare once released from certain Sections of the Act. These are outside the scope of data collections made as part of this release, but more information can be found in the Code of Practice for the Mental Health Act.

The various Parts and Sections of the Act have different purposes, durations and other features but the majority of sections used to detain patients fall under Parts II and III of the Act, which respectively cover ‘civil sections’ and ‘forensic sections’ (those applied under criminal law). More information about individual uses featured in the report is provided in the Metadata file.

The 2007 changes to the Mental Health Law

The Mental Health Act 2007 made some major amendments to the existing 1983 Act. These included the introduction of Community Treatment Orders (CTOs), which came into effect in November 2008 and replaced Supervised Discharge. They allow people who meet the criteria to be treated in the community rather than under detention in hospital and were intended to address the problem of ‘revolving door’ patients (those that end up being repeatedly detained in hospital), although a person does not need to have been readmitted in order to be placed on a CTO. Information about the use of CTOs was introduced in the 2008-09 annual publication.

The 2018 independent review of the Mental Health Act

In 2017 the government announced that an independent review of the Act would be carried out. The review was set up to look at how the legislation in the Act is used and how practice could improve. In December 2018, the review published recommendations in the report ‘Modernising the Mental Health Act: increasing choice, reducing compulsion’. In responding to the review, the government said it would introduce a new Mental Health Bill. 

The Mental Health Act and the Mental Capacity Act

The Mental Capacity Act allows, among other provisions, the restriction of freedom for individuals who do not have capacity to agree to decisions regarding their freedom, finances, and choices about health assessments, treatment and visitors.

The 2007 Mental Health Act made changes to the Mental Capacity Act 2005 including the introduction of Deprivation of Liberty Safeguards (DoLS) from 2009 which are used to deprive a person of their liberty.

Information about DoLS applications under the Mental Capacity Act is not included in this publication; these data are published separately by NHS Digital. 

There is no legal framework required for restriction of liberty of individuals who are not detained in hospital under the Mental Health Act (those people either not subject to the Mental Health Act or on a CTO or Guardianship order).

Where a person lacks capacity, the Mental Capacity Act should usually be used in preference to the Mental Health Act to admit and detain a person in hospital for treatment, provided the person doesn’t object to or resist the admission or treatment. The Mental Health Act is required if the deprivation of liberty in hospital is to give treatment for mental disorder and the treatment could not otherwise be given because of a valid and applicable advance refusal or refusal from a welfare attorney.

Deprivation of liberty orders are applied for by the responsible care home or hospital and are authorised (or not) by the responsible supervisory body. The safeguards are intended to ensure that this is only done when it is in the best interests of the individual and also to provide a framework to determine whether a deprivation of liberty is already occurring for existing cases, and whether or not this is appropriate, as well as reviewing or monitoring existing arrangements.

The degree to which the statistical product meets user needs in both coverage and content.

This publication is the source of official statistics about uses of the Mental Health Act in England. It presents the number of new detentions under the Mental Health Act which occurred during the annual reporting period, broken down into sections and scenarios, and also the number of people detained under the Act at the year-end. It also includes counts of short-term orders and Community Treatment Orders.

The data are primarily sourced from the MHSDS, the key source of national information about the use of NHS-funded mental health, learning disabilities and autism services. To ensure that the relatively small number of uses of the Act that occur in the emergency departments of acute hospitals are also covered, data from the ECDS are also included in this publication.

This publication only covers uses of Section 136 where a health-based place of safety was used. The Home Office publishes data based on all places of safety, including police stations. 

Content of this publication

This publication includes the following statistical outputs:

• An Easy Read version of the summary report to make summary information more accessible
• Interactive data visualisations using Microsoft Power BI which illustrate variations in uses of the Act across different groups of people and also information about data quality
• Excel Data Tables containing aggregate counts of detentions and people detained under the Act, in the same format as previous years
• Machine-readable CSV data file which includes the data in the Excel tables at national level and additional information at sub-national level (based on high-level measures presented in the Excel Tables)
• Metadata document describing the measures in the publication and explaining how they have been derived from record-level administrative data.

Definitions

The following measures are included in this publication:

• Detentions under the Mental Health Act 1983, by legal status
• Uses of Short-Term Detention Orders under the Mental Health Act 1983
• Uses of Community Treatment Orders
• Uses of section 2 and section 3 of the Mental Health Act 1983
• People subject to the Mental Health Act 1983 on 31st March 2021
• People subject to repeated detention under the Mental Health Act 1983
• Discharges from hospital when previously detained under the Mental Health Act 1983
• Length of Detention of patients discharged from hospital when previously detained under the Mental Health Act 1983

Using the MHSDS administrative data source has made it possible to add further breakdowns including age, gender, deprivation decile and ethnic group. Detentions, uses of section 136 and CTOs are therefore also presented as rates per 100,000 population. For detentions, the rates are also presented by Sustainability and Transformation Partnership (STP) areas, to meet the need for monitoring of the Act at this NHS administrative geography.

From 2016-17, the methods used to calculate the measures in this publication have changed, due to the change in the way the figures are sourced and produced. This change was described in an Announcement of Methodological Change.

The figures are now produced from referral and patient-level information rather than from aggregate numbers submitted by individual hospitals. The new method produces these statistics from analysis of individual changes in legal status and the sequence of hospital admissions recorded for each person in the MHSDS. Essentially the information is derived by re-using data collected in the course of delivering patient care, rather than from aggregate numbers submitted to a bespoke data collection.

All providers registered to use the Mental Health Act are required by law to keep records of uses of the Act. This is done using official forms, so definitions of terminology and local recordkeeping are well established. Records are liable to inspection by the Care Quality Commission in its regulatory role. However, the change in data source means that the route for supplying information for national monitoring of the Act has changed. Details of arrangements for data collection and submission are provided in the Completeness section below. Further information about the logic used to construct individual measures is provided in the Metadata file.

National Statistics

From 2017-18, the data in this publication are classified as National Statistics. These statistics have been assessed by the UK Statistics Authority as being fully compliant with the Code of Practice for Official Statistics. They have been judged to meet the highest standards of trustworthiness, quality and value. The Authority judged that the change in data source has improved the statistics by providing greater insight into uses of the Act and that NHS Digital should continue to innovate on this basis.

Meeting user needs

From 2016-17, NHS Digital made the following changes to this publication to better meet user needs:

• Change in primary data source: the use of more granular referral-level data from the MHSDS allows NHS Digital to perform new analyses of the data that were not possible from the aggregate KP90 collection. This includes presenting rates of detention by age, gender and ethnicity. Over time, additional insights will be gained from further investigations of the data.
• Publishing information about uses of the Act more frequently: several measures are included in the ‘Monthly Data’ CSV file in each Mental Health Services Monthly Statistics publication (see Appendix 1 for further details).
• Improved reporting in this publication: in 2016-17 we changed the format of the summary report and improved the clarity of explanations. Additional Excel tables are again provided this year which provide rates of detention by increased gender values such as "Non-binary" and "Indeterminate". Also, we have now provided more granular rates of detention by Sustainability and Transformation Partnership (STP) combined with age group, gender, ethnicity and deprivation. We have also produced analysis of people subject to repeated detention, which could not be produced from the KP90. We have produced Microsoft Power BI reports, allowing users to access interactive data visualisations. In addition, we have produced an Easy Read version of the summary report to improve accessibility.

Completeness

We have assessed completeness at three levels:

• Have all eligible organisations submitted data to MHSDS?
• Have the organisations that use the Act included this information about in their MHSDS submissions?
• Is the data submitted about the Act complete? For example, does it cover all relevant hospital sites and services and all required elements of the Act (e.g. uses of section 136 as well as uses of CTO)? Have monthly data been submitted consistently and at the expected volumes?

Completeness: Have all eligible organisations submitted data to the MHSDS?

Not all eligible organisations are yet submitting data in the MHSDS. There are difficulties in correctly identifying all those organisations that should do so. This is particularly the case in the independent sector, which includes a large number of small organisations. Organisational change including mergers and changes in registration makes it difficult to maintain a complete list of all eligible healthcare providers.

NHS Digital is working with partners, including NHS England, NHS Improvement and the Care Quality Commission (CQC) to ensure that all eligible organisations submit data to the MHSDS. We are working on a comprehensive data quality improvement plan for mental health data. We are contacting non-submitting organisations to ensure that they have plans in place to begin submitting to the MHSDS and providing additional support to new submitters.

Information about organisations submitting data to the MHSDS is included in the monthly publication ‘Mental Health Services Monthly Statistics’. This information is shown both in the Data Quality outputs published each month and also in data visualisations at the Mental Health Data Hub.

These reports show that the number of submitters to the MHSDS has continued to increase in 2020-21, from 106 in April 2018 to 171 in March 2019 to 267 in March 2020 to 314 in March 2021 to 344 in March 2022. It should be noted however that some of the increase in recent years is attributed to smaller independent healthcare providers who generally would not be expected to record any detentions.

The Mental Health Services Data Quality page at the Data Hub provides interactive data visualisations which can be used to investigate aspects of data quality further. 

Completeness: Have the organisations that use the Act included this information in their MHSDS submissions?

See also: Are all organisations submitting data about the Act? page

Not all organisations that are required to submit MHSDS will make use of the Mental Health Act, so the number of organisations eligible to submit MHSDS is greater than the number registered to use the Act.

The majority of uses of the Mental Health Act in health services occur in specialist mental health, learning disability and autism services, including those commissioned by the NHS but provided by Independent Sector Providers (ISPs). These organisations are mandated to submit MHSDS in accordance with the information standard published under section 250 of the Health and Social Care Act 2012.

For the purposes of assessing that all eligible providers who use the Mental Health Act submitted data, we have also drawn on comparisons to providers that submitted data to the last KP90 collection in 2015-16. However, users should note that some relevant organisations may have opened, closed or merged since then.

In 2015-16, eight independent service providers accounted for 90 per cent of detentions in this sector. Seven of the eight providers submitted MHSDS data in 2016-17 and all eight providers submitted some MHSDS data in 2017-18 and 2018-19. However, poor data quality from Priory Group, including the subsidiary organisation Partnerships in Care, has the greatest negative impact on completeness in 2019-20, 2020-21 and 2021-22. Neither organisation made twelve MHSDS submissions in 2019-20, 2020-21 and 2021-22; and in both cases none of their submissions contained Mental Health Act data (in Table MHS401). To provide an indication of the scale of the shortfall this creates in the national figures, these two organisations together recorded nearly 2,500 detentions in 2015-16, the last year of the KP90 aggregate collection. For 2019-20, 2020- 21 and 2021-22 only five of the eight major independent providers submitted any detentions data.

In terms of completeness, these large providers tend to have many sites which are geographically dispersed. Further work is required to identify which sites are in scope for the MHSDS and NHS Digital is working with partners in this regard. In addition, although there are many small ISPs that have not provided data to the MHSDS, it is the quality of that submitted by the large organisations that has a greater impact on completeness.

To assess the completeness of data about uses of the Act, we have grouped providers into three categories. This reflects the different contribution each group makes to these statistics and the different issues affecting data quality. These groups are:

• NHS Mental Health and Learning Disabilities Trusts – most uses of the Act occur in these settings and these providers are required to submit data to the MHSDS.
• Acute Trusts – a large number of organisations which each typically account for a small number of uses of the Act, if any. From 2018-19 these organisations can submit this data via the new Emergency Care Data Set (ECDS). In both 2018-19 and 2019-20 the ECDS data proved to have significant data quality issues, with the result that there was a significant decline in detentions from this group of providers compared to previous years. However, in 2021-22, there has been a slight improvement in detentions from this group (19.8% when compared to last year) although this is still lower than detentions prior 2017- 18. This partly reflects improving data quality in in ECDS collection. Please refer to the tables in Appendix 2 Table 8 for details of providers submitting to the ECDS this year.
• Independent Service Providers – a large number of organisations typically providing inpatient services including some secure settings likely to provide services for detained patients. These providers are required to submit data to the MHSDS. There is a wide variation in the size and capacity of ISPs, and these are not all registered with the NHS Organisation Data Service (ODS) service in a way that supports consistent identification.

Table 5 of this publication’s Excel Data Tables reflects this grouping.

Table 1a below shows that the number of organisations submitting Mental Health Act data has declined this year largely as a result of poor coverage from the ECDS. The number of Acute providers submitting data declined to around half last year’s number. Another area of shortfall remains independent providers of specialist mental health services. The number of ISP submitters to the MHSDS was 17 in 2020-21 compared to 46 in the last year of KP90.

Table 1a: Number of organisations submitting data about the Mental Health Act, by provider type, 2015-16 to 2021-22

Sources: KP90 2015-16, MHSDS & MHA Acute 2016-17 & 2017-18, MHSDS & ECDS 2018-19, 2019-20, 2020-21 & 2021-22

Table 1b shows the detentions that were derived from submissions made by these providers. Overall, the number of known detentions was up by 0.2 per cent compared to last year, but there were variations looking at provider types.

Detention counts for specialist mental health and acute providers increased, partly reflecting improving data quality in both collections. We are continuing to provide feedback to previous Acute submitters about the need to submit Mental Health Act data to the ECDS to continue improving the data quality.

Table 1b: Number of detentions under the Mental Health Act, by provider type, 2015-16 to 2021-22

Sources: KP90 2015-16, MHSDS & MHA Acute 2016-17 & 2017-18, MHSDS & ECDS 2018-19, 2019-20, 2020-21 & 2021-22

The number of ISPs presented in Table 1a excludes a group of smaller independent providers from 2015-16 which were shown as ‘other’ ISPs in Table 5 that year. In 2015-16 data was collected from 225 independent sector hospitals. This includes 33 hospitals that were not reported separately in past years' CSV file as they either had no ODS code and/or represented very small numbers of detentions. These providers were aggregated into an ‘other’ group and accounted for 95 people detained on 31st March 2016.

Users interested in exploring the data submitted by organisations should refer to the Power BI reports released with this publication and also Appendix 2 in this data quality page. These contain detailed breakdowns of organisation-level data which are intended to assist users in a more detailed interpretation of completeness, and to support ongoing data quality improvement initiatives.

The interactive Power BI reports also allow users to select individual providers and to compare between providers and over time. Users can view the number and consistency of records submitted in each of the data tables used to produce the statistics in this publication.

Further details of the specific data quality issues encountered by organisations trying to submit Mental Health Act data can be found at the Data Quality page at the Mental Health Data Hub. This information is available in the ‘Coverage’ visualisation (shown under the ‘Coverage for Mental Health Services Data Set’ heading).

The CQC monitors the use of the Act, including the keeping of records about its use by service providers. We are supporting the CQC in their work to ensure that previous submitters to the KP90 collection are all submitting data to the MHSDS. In their regulatory role, the CQC will also include an assessment of the data submitted about the Act when making decisions about whether a care organisation is ‘well-led’.

Completeness – Are the data submitted about the Act complete?

See also: Are the MHSDS data complete? page

Our analysis of the data, including comparisons to KP90 returns in previous years, indicate that the data are incomplete at national level.

Our assurance processes looked both at the submission and consistency of data across the 12 months of the reporting year and used comparisons with the snapshot count of patients subject to the Act at 31st March (as shown in Table 5 in the Excel Data Tables) to assess the completeness of submissions.

Users interested in reviewing monthly data volumes can explore organisation-level counts on the ‘Provider-level coverage’ visualisation in the Microsoft Power BI reports provided on the Mental Health Act statistics page of the Mental Health Data Hub. Please note that there is no direct relationship between the number of records submitted in any one, or all, of these tables and the measures presented in this publication, which are derived from a combination of these items.

Where mergers and acquisitions have occurred in the independent sector, the resulting mix of different IT systems in a single new organisation makes it is difficult to assess whether data are being submitted for all sites in scope or if the data being submitted are limited to a subset of MHSDS-compliant IT systems. Further investigations are required, particularly for large independent providers that have multiple sites, to determine which sites are in scope for the MHSDS and to ensure that action plans are in place to submit data for any missing sites. NHS Digital is working with partners in this regard.

As part of monthly MHSDS publications, national and organisation level data quality measures are shown that validate a selection of key data items by provider. The Data Quality csv files in each monthly publication contains a variety of data quality information relevant to the Mental Health Act. This information is also updated each month at the Data Quality page of the Mental Health Data Hub as interactive data quality visualisations:

• Provider Feedback: a qualitative summary of any data quality issues, by provider, resulting from validation and investigation of each month’s data by NHS Digital and the provider.
• Coverage: the number of records submitted each month by organisation and also by MHSDS table. This is a useful resource for users due to the high volume of detailed information it contains. The March 2022 Submission Tracker contains this information for the full 12-month period covered by this report, whilst the Power BI reports contain data from April 2016.
• Validity Count / Percentage: for each organisation, the numbers and proportion of records which have ‘valid’, ‘other’, ‘default’, ‘invalid’ and ‘missing’ values. This includes the Mental Health Act measure ‘MHS-DQM07 Mental Health Act Legal Status Classification Code’.

The ‘Data Quality CSV’ file in each monthly publication also includes the number of records submitted by each organisation for the measure ‘MHS-DQM07’ in machine-readable format.

Table 2 shows the number of providers submitting data to the ‘MHS401: Mental Health Act Legal Status Classification Code’ data table for each month in 2021-22. During this period, the number of submitters fluctuated between 68 and 77 providers. Users interested in exploring this data further can use the ‘VODIM status for MHSDS records by Measure and Provider’ visualisation at the Data Quality page of the Mental Health Data Hub to investigate trends in the number of records submitted and validity by provider.

Table 2: Number of organisations submitting any records to Table MHS401 (Mental Health Act Legal Status Classification Period), by month and organisation type, 2021-22

Source: Monthly MHSDS Data Quality Reports (Data Quality Measure MHS-DQM07 (Valid))

To support providers in making accurate monthly submissions to the MHSDS, NHS Digital provides feedback at different stages in the data flow (including reports at the point of submission). These include detailed Data Summary Reports about coverage, volume, code validity and data consistency.

In 2019-20 the model for submissions to MHSDS changed. For the majority of the year providers had the opportunity to make an initial ‘primary’ submission followed later by a ‘final’ submission. Where there were concerns about data quality, NHS Digital contacted providers directly so that any issues with local data extraction processes can be addressed for a future submission. Where issues had been identified in a primary submission, the process allowed organisations to correct them in a subsequent final submission. Note that these checks are currently limited to key elements of the dataset.

In April 2020 a new submission model was introduced. For the 2019-20 reporting period this acted as a “Refresh” of the data for the year. This allowed providers to submit any periods which they might have missed through the year or to resubmit any period of data within the 2019-20 period. This new approach is intended to allow providers some flexibility to improve their submissions by amending incorrect data or adding previously missing data. For the 2021-22 period this submission model has allowed providers to update data for any reporting period which has passed in the financial year. The provisional submission window replaces the current primary window and the performance window replaces the current refresh window. Data can then be resubmitted in each update window. The final window will be the last chance to amend data for the financial year. The ‘last good file’ submitted for each month will be used to generate final statistics at the end of the year.

Accuracy is the proximity between an estimate and the unknown true value. Reliability is the closeness of early estimates to subsequent estimated values.

Methodology

From 2016-17 these statistics about uses of the Mental Health Act are primarily produced from referral-level information sourced from the MHSDS rather than from aggregate numbers submitted by individual providers to the KP90 collection. The ECDS is used as an additional data source for Acute providers who do not make submissions to the MHSDS (see next section).

Therefore from 2016-17 these statistics are derived from administrative data, which are data produced in the course of delivering patient care, rather than from a bespoke aggregate collection. Their use for statistical purposes is secondary. Using this referral-level data allows a much greater range of analysis to be produced, including new breakdowns, for example by age or ethnic group, and for different patient pathways (for example, those on an Early Intervention Pathway, or those who enter the services via a crisis pathway). It supports analysis or repeated detention for individual patients within and across reporting periods.

The MHSDS v5 contains 60 data tables, only some of which are required to produce measures in this publication. These tables include:

• MHS001: Master Patient Index (provides the age, gender and ethnicity for each person, enabling analysis by these characteristics)
• MHS401: Mental Health Act Legal Status Classification Period (provides details of detention periods for people detained under the Act)
• MHS403: Conditional Discharge (provides details of each separate period of conditional discharge for the patient)
• MHS404: Community Treatment Order (provides details of CTO periods)
• MHS405: Community Treatment Order Recall (provides details of periods where a person on a CTO was recalled to hospital for treatment under section 17a of the Act)
• MHS501: Hospital Provider Spell (provides details of periods in hospital). Please note that each record has an associated referral in table MHS101: Service or Team Referral and that data cannot be submitted without an associated referral.

NHS Digital has developed methods for accurately producing counts of the specific scenarios that are monitored in the official Mental Health Act statistics, by analysing individual changes in legal status and the sequence of hospital admissions recorded for each person in the MHSDS. From this information, the number of uses of individual sections of the Act can be determined.

Therefore, whilst the methodology can accurately produce counts of events with a high degree of accuracy from the data source, issues with completeness affect the accuracy of outputs produced using these methods.

The metadata file contains further details including definitions for measures included in this publication and technical descriptions of the constructions used to produce these measures from the data sources.

Accuracy of Acute data from the ECDS

Although uses of the Act in hospital emergency departments comprise a relatively small proportion of the total uses, accurate data is required from these services to ensure the accuracy of the national Mental Health Act figures.

Previously these providers submitted their Mental Health Act data via the KP90 collection along with specialist mental health providers. When the KP90 was decommissioned, a smaller Acute collection was set up to collect these uses as an interim measure (like the KP90, this collected aggregate data). To reduce burden and increase utility, providers were expected to submit data about their uses of the Act to the new Emergency Care Data Set (ECDS) once it became operational; the Acute collection would be decommissioned at this point.

As noted previously, from this year Acute providers were able to submit data about their uses of the Act via the ECDS (and the interim Acute collection was closed). However, not all emergency departments are yet submitting data to the ECDS. The Mental Health Act data is recorded in a dedicated table, but our analysis shows that this table also has significant coverage issues. At national level this results in shortfalls in the numbers of detentions and short-term orders shown in this publication.

To support ongoing data quality improvement initiatives, Table 8 in Appendix 2 lists the providers who have successfully submitted data about the Act to the ECDS whilst Table 9 shows former KP90 submitters who did not submit to either the MHSDS or ECDS (as appropriate) in 2021-22. A comparison of Acute providers using Table 5 of the Data Tables will also show those whose submitted data has changed significantly from last year, indicating potential data quality issues.

We are also working with our Acute provider networks to ensure that the recording of Mental Health Act data in the ECDS improves.

Sources of error and bias

The main source of error and bias in these statistics is the completeness of the MHSDS data (see ‘Relevance’ for a summary of these issues). At national level, this results in a downward bias to counts and any derived rates.

Therefore, although we are presenting the same measures as in previous years, due to missing data the figures derived from MHSDS submissions since 2016-17 are not directly comparable to those published in earlier periods from the KP90 collection.

The impact of coverage issues on the reliability of individual figures varies according to what is being counted and at what level the data are being used. The following general guidance to bias in this publication is provided alongside detailed guidance below for each Excel table:

• Counts of events (such as new detentions) are understated in these statistics due to shortfalls in coverage (also see ‘Relevance’). Counts of events at sub-national level are also understated, but to varying degrees, depending on the completeness of the data in each area. This publication includes the first breakdowns of detentions by Sustainability and Transformation Partnership (STP) areas (Tables 1d, 1i, 1j in the Data Tables). As an aggregate data collection, the KP90 did not support further analyses by geography. However, counts of detentions and associated rates by STP areas are each affected by coverage and completeness issues to a different extent. Please note that these breakdowns are based on the patient’s place of residence, not place of treatment, and therefore do not reflect treatment that occurs away from the area of residence. A further source of downward bias is that this detailed analysis can only be produced from the record-level MHSDS data; this information could not be derived from the ECDS this year.
• Counts of people detained at year-end (see Table 5) are particularly affected by missing data from Independent Service Providers (ISPs) and from Acute providers submitting data to the ECDS. Due to the type of services they provide, ISPs tend to have longer stay patients and these are more likely to be picked up in a snapshot end of the year count than those detained for shorter periods of time. Therefore missing data from ISPs will have a greater impact on the count of people detained (Table 5) than for new detentions (Table 1). This is because the count of people detained at the end of the period will include not only some people detained in the current reporting year, but also in hospital as a result of continuing detentions that commenced in earlier periods. In addition, we have been unable to obtain counts of people from the ECDS this year. We will be able to add this information to future publications once NHS Digital has completed development of new analytical methods to reliably count patients from this new dataset.
• Population-based rates for age, gender, deprivation and ethnicity are derived using counts of events. Therefore they are also understated due to the numerator (the number of events) being incomplete whilst the denominator (population) is complete (based on either Census data or ONS estimates of the complete figures). There is an additional downward bias resulting from records where a valid age, gender or ethnicity could not be obtained. However, completeness for these fields is good in the MHSDS and the counts used are shown in each Excel table. A further source of downward bias is that this detailed analysis can only be produced from the record-level MHSDS data; this information could not be derived from the ECDS this year.

Although we cannot know whether the missing data causes bias in term of age, gender or ethnicity, we can segment the data that we did receive to see if there are large variations in characteristics. We can compare demographic information for NHS and independent providers that did submit data to see if there are large variations in age, gender or ethnicity which could result in bias to derived rates, due to variations in completeness across these types of providers.

Table 3 shows the counts for age, gender and ethnicity split between NHS and Independent Sector providers. These counts show that despite the differences in services provided, the percentage splits of gender, age and ethnicity are broadly similar across NHS and Independent Sector providers and as such we believe the calculated rates of detention are reliable.

Table 3: Detentions by gender, age and ethnicity, 2021-22, by provider type

Source: MHSDS

Table 4 below provides detailed guidance on errors and bias from the MHSDS, as the main data source for this publication. It highlights issues specific to each of the Excel Data Tables.

Table 4: Sources of errors and bias, by Excel Data Table, 2021-22

Transfer of patients between providers

Our analysis of the record-level MHSDS data shows that where patients are transferred between hospitals whilst subject to the Act, the same uses of the Act may be recorded in two places, leading to a potential double counting of uses. It suggests that there was some double counting of this type in previous years where providers submitted aggregate figures to the KP90. This means that detention figures submitted in KP90 returns may have been overstated.

NHS Digital has developed methods for ensuring that each use is only counted once in the analysis of MHSDS data; this is because the data shows where the same use of the Act has been recorded by more than one provider and we count each use where it originally occurs – and further occurrences can be identified as transfers whilst the patient was ‘on section’.

This count of transfers is not part of the existing time series, but we believe it provides important context (transfers ‘on section’ were not recorded in the KP90 collection). Transfer numbers for 2021-22 at provider level are included in the CSV file. Using the record-level MHSDS, our analysis of 2021-22 data shows that in addition to the 34,844 detentions that occurred at the point of admission, a further 5,776 admissions occurred where the patient was transferred from another provider whilst already detained (‘transferred on section’). In the KP90 returns, we believe that some of these transfers were submitted as new detentions on admission to hospital. Although we can derive the number of transfers on section using the MHSDS, we do not know to what extent this double counting occurred prior to 2016-17 as ‘transfers on section’ were not included in the KP90 collection.

Composition of detentions on and following admission to hospital

These statistics have always shown detentions that occurred on admission to hospital separately to those that followed admission (See Table 1a in the Data Tables). In the KP90 return, providers were responsible for classifying detentions into these categories and submitting aggregate figures. In the MHSDS, both the date and time of events are recorded so we classify detentions based on the date and time of hospital admission and change in legal status. We can therefore identify detentions that occurred following admission to hospital but on the same day.

This year we find that the number of detentions on admission to hospital has decreased whilst those following admission to hospital have increased. Table 1a in the Data Tables shows that the main contributors to this trend are Independent providers. Examining the breakdown of sections, the main factor is an decrease in detentions under Section 2 occurring on admission, with an increase in detentions under Section 3 following admission where initially admitted as an informal patient. There are several possible reasons for this trend, and this may be the result of a combination of factors which would require further investigation and analysis. These include:

• Improved coverage – as more providers submit data about the Act to the MHSDS, this affects the distribution of national figures and may contribute to the observed pattern.
• Change of secondary data source – the change in composition of detentions is driven by NHS providers and a contributing factor may be the change from the Acute aggregate collection to the ECDS, an administrative data source. Coverage is lower from the ECDS this year than the Acute collection last year.
• Accuracy of time recording – there may be a greater proportion of matching start times between the Mental Health Act period and the hospital provider spell. This could be the result of either improved recording, or greater use of rounded times matching across these tables. This is covered further in the next section.

Time recording in the MHSDS

Our analysis indicates that the accuracy of time recording in the MHSDS is a factor in the proportion of detentions shown as occurring either on or following admission. Specifically, it appears that overall submissions include an unusually high proportion of rounded times and this leads to difficulties in accurately determining whether a detention occurred on admission to hospital or following it.

The start time for the hospital provider spell and Mental Health Act episode are used to calculate whether a detention occurred ‘on admission’ or ‘following admission’. Possible scenarios are below used to illustrate potential issues when times are rounded.

• Scenario 1: A person is detained on admission to hospital at 10:37am. The hospital reports both the time of admission and the start time of the detention accurately as 10:37am. From the data reported to the MHSDS, this event is correctly interpreted in these statistics as a detention on admission.
• Scenario 2: A person is detained on admission to hospital at 10:37am. The hospital reports the time of admission accurately at 10.37am. The Mental Health Act episode is reported with a lower degree of precision than the hospital admission. It is reported as occurring at 11.00am. From the data reported to the MHSDS, this event is incorrectly reported as a detention following admission.
• Scenario 3: A person is admitted to hospital at 1:22pm and is later detained at 5:25pm. The start time of the hospital spell is not reported by the provider, but the Mental Health Act episode start time is reported. NHS Digital cannot compare the missing times, so the reporting is based on the dates submitted. As these events are both submitted as occurring on the same day, this event is reported as being a detention on admission when it was actually a detention following admission.

Table 5 shows the degree to which rounding of times, as shown in Scenario 2, occurred in the 2021-22 data.

Table 5: Percentage of Mental Health Act episode and hospital provider spell start times recorded as occurring at specified intervals, 2021-22

Source: MHSDS

About a quarter of Mental Health Act episodes were recorded as starting on the hour, alongside over a sixth of hospital provider spells. Times were also rounded to the nearest 30- minute and 15-minute intervals. Only 45.8 per cent of Mental Health Act episodes were not reported at starting at any of these intervals, which suggests a significant proportion of times were rounded. This may be the result of limitations within provider IT systems used or the configuration of those systems.

The start time for a Mental Health Act episode is a ‘mandatory’ field and must be completed in MHSDS submissions when data are submitted for the MHS401 data table. As such all records successfully submitted to the MHSDS relating to a Mental Health Act episode in table MHS401 of the MHSDS contain a start time. This does not however guarantee that the time is completed accurately.

The start time of the hospital provider spell is a ‘required’ field rather than a ‘mandatory’ field. These should be reported to the MHSDS where they apply but it is possible to make a successful MHSDS submission with ‘required’ items missing. This results in the type of issue illustrated in Scenario 3.

Since 2017-18, the quality of time recording in both the MHS401 and MHS501 tables has been largely the same. There are some small differences but given the change in the number of submissions and the quantity of data submitted it cannot be suggested that the changes seen are significant.

Comparability is the degree to which data can be compared over time and domain. Coherence is the degree to which data that are derived from different sources or methods, but refer to the same topic, are similar.

Time series data are presented as part of these statistics, as in previous years. However due to the change of data source in 2016-17 and the issues of accuracy identified above, it is not possible to make direct comparisons to periods up to 2015-16. However, we have produced some more nuanced analysis to assist users in understanding the differences observed in the number of detentions compared to previous years.

The impact of COVID-19 must also be considered. The COVID-19 pandemic has had a large impact on society and health services around the world. Due to the coronavirus illness (COVID-19) disruption, this is now starting to affect the quality and coverage of some of our statistics, such as an increase in non-submissions for some datasets. We are also starting to see some different patterns in the submitted data. For example, fewer patients are being referred to hospital and more appointments being carried out via phone/telemedicine/email. Therefore, data should be interpreted with care over the COVID-19 period. This publication covers 2020-21 and as such the impact of COVID-19 may largely be seen in the figures.

However, in England the national lockdown began on 23 March 2020. The time series data for people subject to detention does show a decrease in people subject to detention in March 2020 so the context of COVID-19 should be kept in mind.

Change in primary data source

The change in primary data source from 2016-17 follows the recommendation in the Secretary of State’s Fundamental Review of Returns 2013 that the KP90 collection would be retired once the same information could be produced from administrative data sources.

The scope of the primary administrative data source has gradually increased from covering only NHS mental health services for adults, to including Independent Sector Providers in 2011 together with changing the format to permit analysis of individual uses of the Act, adding learning disability services in 2014 and, with introduction of Children and Young Peoples’ services and the MHSDS in January 2016, coverage of children and adolescents.

This means it now covers most services where the Act is used. A small number of uses of the Act occur in NHS acute hospitals each year and remain out of scope for MHSDS. This information was collected separately in a bespoke return in 2016-17 and 2017-18 but has now been replaced by the ECDS, another administrative data source. Although the number of uses of the Mental Health Act that occur in these settings is small compared to the MHSDS, accurate data are required to provide a full national picture. Services should be delivered in the least restrictive setting possible, and this cannot be monitored without information about uses of detentions and other sections of the Act. This includes any changes in the way the Act is used, particularly use of Section 136 in Acute emergency departments.

Comparisons between statistics from KP90 and MHSDS

As the new MHSDS-based analysis counts the same events as were previously produced using the KP90 collection, in theory it is possible to compare outputs from both data sources. However, at present the high-level numbers are not comparable to those produced in previous years due to organisational coverage and any local issues with supplying comprehensive, accurate data via the MHSDS. Therefore, statistics produced from the MHSDS should not be directly compared to those produced from the KP90 collection.

Estimating the change in detentions from 2020-21 to 2021-22

Although the number of reported detentions has increased by 0.2 per cent from 2020-21 to 2021-22, part of this increase is due to changes in data quality.

Improvements in MHSDS data quality were offset by the poor data quality from some Acute providers. Although the number of providers submitting to ECDS was greater that last year, fewer Acute providers submitted data to the ECDS than the previous Acute collection, and some of the data that were submitted were incomplete.

To estimate the change in detentions between 2020-21 and 2021-22, we have made an assessment that the quality of data for a smaller group of providers supports a like-for-like comparison, based on the following criteria:

• All providers submitted data to the KP90 in 2015-16 (this assists in the assessment of data quality) and MHSDS in 2016-17 to 2018-19. Providers submitting to the ECDS or Acute collection were excluded from this analysis.
• All providers submitted 12 months’ data to the MHSDS in each year.
• Our ongoing data quality investigations did not reveal any significant MHSDS data quality issues reported relating to detentions.

There were 23 NHS and one independent provider that met these criteria (see Appendix 2). Our analysis was based on data from these 24 providers. It excludes ‘detentions following use of section 136’ and ‘revocation of community treatment orders’. Although these form part of the total number of detentions (as shown in Table 1a of the Data Tables), they are relatively small components of the total and completeness for these measures is affected by different factors than the two main measures (‘detentions on admission’ and ‘detentions following admission’). Therefore excluding these categories allows for the production of more reliable estimates of trend changes.

Table 6: Detentions figures for 24 selected providers, 2020-21 to 2021-22

Source: MHSDS

The data in Table 6 show that when we restrict our analysis to a smaller group of 24 providers with good data quality, there is an decrease of 5.7% per cent in detentions from 2020-21 to 2021-22.

Geographical comparisons

Using the new MHSDS data source, we have produced geographical analyses and this publication includes breakdowns of detentions by Sustainability and Transformation Partnership (STP) areas (Table 1d, 1i, 1j in the Data Tables) and Clinical Commissioning Group (CCG) of GP registration or residence (Table 1f). Comparison between STP areas and CCG areas should be made with caution, as each area is affected by coverage and completeness issues to a different extent. The numbers and rates shown do not allow a true comparison of detentions between different STP or CCG areas. However, they can be used to indicate areas where MHSDS coverage is low and therefore require further data quality investigations.

Comparisons using population-based rates

We have presented population-based rates in this publication, which allow comparisons to be made between different groups of people, even though the data are incomplete (see also ‘Accuracy and Reliability’ section). We have presented crude rates for gender, age groups and deprivation but also standardised rates for ethnicity, which are split into smaller groups of people and are therefore subject to greater variation. These ethnicity rates use direct standardisation to adjust for the different gender and age profile of each ethnic group, allowing us to make comparisons on a like-for-like basis. We have also calculated 95 per cent confidence intervals which are presented alongside the rates.

Population figures do not match across the Data Tables as we have used the latest figures available in each case, at the time of production. For gender and age, we have used ONS mid-year population estimates 2020. CCG breakdowns use the ONS mid-2020. The STP breakdowns use the CCG population data mapped to STP. Whilst this may not map exactly, the benefit of more timely population data outweighs the fact that some CCGs may not map exactly. Ethnicity data uses 2011 Census as mid-year estimates are not produced for this breakdown. This Census data will not reflect any changes in the ethnic structure of the population in England which have occurred since 2011. When deriving rates, this creates a mismatch as the numerator is based on 2020-21 data, but the denominator uses 2011 data; this mismatch will increase until the next Census data are published.

Changes in processing

This is the second Annual Uses of the Mental Health Act publication that uses data collected through the Strategic Data Collection Service in the cloud (SDCS Cloud). It is also the second time data are linked using the Master Person Service (MPS). The change to using MPS from the original MHSDS Person ID means that data will likely change slightly but the impact is thought to be very small with the impact estimated to be less than 0.1% difference at national level. The published number from 2018-19 using MHSDS Person ID was recorded as 48,868. Using MPS Person ID the total would have been 48,856 (-0.02% change).

At provider level the differences are very small. The percentage difference ranges from -13% to 40% difference but the actual differences between numbers only range from -14 (-0.8%) difference to +4 (0.2%) difference. As such whilst these issues do impact the comparability of the data between 2016-17 to 2018-19 and 2019-20 onwards, they should not stop people from making comparisons across the years on this basis alone.

Other Mental Health Statistics

The summary report contains links to related information, including to:

• Monthly publications of mental health statistics produced by NHS Digital. These cover about secondary mental health, learning disabilities and autism services. The statistics in this Mental Health Act Statistics annual publication are derived from the MHSDS and monthly reports are also published from the MHSDS. The Mental Health Services Monthly Statistics publication includes several measures relating to the number of uses of the Mental Health Act (see Appendix 1). These measures are broken down by age and service type, by provider and CCG of the patient’s GP practice. The counts of people in the March 2022 Final publication are the same as those shown in Table 5 of the Data Tables in this publication as they both relate to a snapshot as at 31st March 2021. However, as noted in Table 4, data for one provider has been removed due to data quality issues and as such will not match to the published data
• The Mental Health Bulletin, an annual publication produced by NHS Digital about people who used adult secondary mental health and learning disabilities services during the period. A link is provided to the 2020-21 publication; statistics for 2020-21 have not yet been published and are also sourced from the MHSDS.
• Statistics on police uses of sections 135 and 136 of the Act. This Mental Health Act Statistics publication only covers uses of section 136 where the place of safety was a hospital. Since the police are involved in all uses of section 136, statistics on police uses also include uses where the place of safety was another location e.g., police station. These statistics are published by the Home Office in the ‘Police Powers and Procedures’ publication series.
• The use of Guardianship under Sections 7 and 37 of the Mental Health Act; these uses of the Act are reported separately by NHS Digital.
• Deprivation of Liberty Safeguards applications under the Mental Capacity Act 2005 (see the Background information to the Mental Health Act section of this document for further information).

Timeliness refers to the time gap between publication and the reference period. Punctuality refers to the gap between planned and actual publication dates.

This report has been produced within seven months of the end of the reporting period. It was produced within six months of the May deadline for providers to submit final data for March 2022, the last month of the annual reporting period.

Accessibility is the ease with which users are able to access the data, also reflecting the format in which the data are available and the availability of supporting information. Clarity refers to the quality and sufficiency of the metadata, illustrations and accompanying advice.

For the first time ever we have converted the summary report into HTML format to improve the clarity of explanations, present the information in a format suitable for a wider range of users. This data quality section will be of more interest to the expert user as it provides detailed explanations of the methods and our assurance of the data.

From 2017-18 we introduced an Easy Read version of the summary report to improve accessibility to a wider range of users, including service users.

An expanded range of interactive data visualisations are provided using Microsoft Power BI. These will be of interest to all users and can be used to further explore local data quality issues.

Data are provided in both formatted Microsoft Excel tables and machine-readable (CSV) formats. The Excel Data Tables retain the same format used last year for consistency.

Re-use of these data is subject to our terms and conditions.

Definitions for measures included in this publication are available in the accompanying metadata file. Terminology is defined where appropriate and we provide a summary of the methods used to produce these measures from record-level data.

Full details of the way that MHSDS returns are processed, which will be of use to analysts and other users of these data, are provided in the MHSDS User Guidance,

This refers to the extent to which different aspects of quality are balanced against each other

Although NHS Digital is committed to expanding the scope of reporting about the Act, the main trade-off was to adjust the scope of analysis to be achievable within NHS Digital resources and production time. Last year, we widened the scope of this annual publication by including new analysis of repeated detention, a new Easy Read version of the summary report and made improvements to data quality reporting in the Power BI reports.

By publishing this information, we hope to promote a virtuous cycle of improving data and service quality through more frequent monitoring, data quality feedback and use.

This refers to the processes for finding out about users and uses, and their views on the statistical products.

A key purpose of this annual publication is to provide the Department of Health and Social Care with information about the number of uses made of Mental Health Act 1983 legislation (except for Guardianship cases under Sections 7 and 37) as amended by the Mental Health Act 2007 and other legislation. It is intended to inform policy development in relation to the Act. Consultation with customers and stakeholders is undertaken through a variety of channels to ensure that developments introduced to the publication meet their requirements. The UK Statistics Authority has supported us and provided guidance as we develop these statistics. This includes working with partners to design statistics that support the NHS Long Term Plan22 . NHS Digital last undertook a consultation on Mental Health Act statistics in 2012 and made a series of changes to the publication as a result. Further changes were made in 2017 following feedback from the United Kingdom Statistics Authority. These changes resulted in this publication being designated as National Statistics (please refer to the Relevance section for further details).

NHS Digital welcomes all feedback and suggestions relating to this publication. Feedback can be provided to NHS Digital via email to [email protected] (quoting ‘Mental Health Act Statistics’ in the subject line) or via telephone on 0300 303 5678.

This refers to the effectiveness, efficiency and economy of the statistical output.

Prior to the introduction of the MHSDS, providers making use of the Act were required to submit this data to the KP90 collection each year. NHS Digital has estimated that the annual KP90 collection required 1,170 working days of effort from submitting organisations, equivalent to 5.85 person-years. The cost was estimated at approximately £367,000 per year. This burden on providers has now been removed as it is produced from the MHSDS, which is also the source for both the monthly Mental Health Statistics and the annual Mental Health Bulletin. A smaller additional burden on Acute providers has also been removed this year as we are no longer requiring an additional submission from these providers. Data on uses of the Act by NHS acute hospitals in emergency departments is collected via the Emergency Care Data Set (ECDS) from 2018-19. Acute providers should be already submitting data to the ECDS and therefore do not need to make an additional submission.

This refers to the procedures and policy used to ensure sound confidentiality, security and transparent practices.

Data submissions to the MHSDS are processed in line with the rules described in the Technical Output Specification for the dataset using a fully assured system that pseudonymises individual identifiers. As with all NHS Digital publications, the risk of disclosing an individual’s identity in this publication series has been assessed and the data are published in line with the disclosure control method for the MHSDS dataset approved by the NHS Digital Disclosure Control Panel. Figures from MHSDS presented at sub-national geographies are rounded to the nearest five, except for small numbers (those from zero to four) which are suppressed and marked by a “*” symbol. Figures from ECDS presented at sub-national geographies are rounded to the nearest five, except for small numbers (those from one to seven) which are suppressed and marked by a “*” symbol. Zeroes are shown in the ECDS data at sub-national level. Figures at national level are unrounded and unsuppressed. Table 5 of the Excel Tables also identifies organisations that did not submit data with a “-“ symbol and these are also identified in Appendix 2.

Please see links below to relevant NHS Digital policies:

A Guide to Confidentiality in Health and Social Care

How we look after information

Freedom of Information Process

Statistical Governance Policy [Archive Content]

This appendix lists the Mental Health Act measures which are included in the Mental Health Services Monthly Statistics release.

These reports can be accessed via the landing page:

Please note that the information below is also available as interactive data quality visualisations at the Data Quality page of the Mental Health Data Hub. Users can choose whether to investigate data quality issues using our prepared visualisations at the Data Hub or by using the CSV files. From April 2019 onwards the Data Quality files are published in CSV format only and only relate to the specific reporting period.

Data Quality Coverage CSV file: Number of items submitted by each provider for each MHSDS table. The following tables collect data which can be used in Mental Health Act analysis (but not all are used in the analysis presented in this report – see the ‘Accuracy and Reliability’ section):

These tables include:

• MHS001: Master Patient Index
• MHS401: Mental Health Act Legal Status Classification Period
• MHS402: Mental Health Responsible Clinician Assignment
• MHS403: Conditional Discharge
• MHS404: Community Treatment Order
• MHS405: Community Treatment Order Recall
• MHS501: Hospital Provider Spell

VODIM CSV: for the ‘MHS-DQM07: Mental Health Act Legal Status Classification Code’ measure there are show counts and percentages of data items that are ‘valid’, ‘other’, ‘default’, ‘invalid’ and ‘missing’ for each submitting organisation.

Provider Feedback CSV: contains supplementary information about data quality, including some issues that relate to Mental Health Act data.

Please note that in order to get a full picture of the data quality for the entire annual period each separate CSV should be consulted.

The following measures are included in the Data Quality machine-readable CSV file (each monthly report includes data for a single month):

• MHS08 - People in contact with mental health services subject to the Act at the end of the reporting period
  • MHS08a – those aged 0-17
• MHS09 - People in contact with mental health services subject to detention at the end of the reporting period
  • MHS09a – those aged 0-17
  • MHS09b – those aged 18-64
  • MHS09c – those aged 65 and over
• MHS10 - People in contact with mental health services subject to CTO or conditional discharge at the end of the reporting period
  • MHS10a – those aged 0-17
• MHS11 - People in contact with mental health services subject to a short-term order at the end of the reporting period

From the July 2020 publication onwards NHS Digital will also be recording the number of uses of the Act in each reporting period. These measures will use the same methodology as this publication and should help to provide more timely data on the uses of the Act. The July 2020 publication was released on 08 October 2020:

• MHS81 - Detentions in RP
• MHS82 - Short Term Orders in RP
• MHS83 - Uses of Section 136 in RP
• MHS84 – CTOs in RP

This appendix includes information about organisations that submitted data about the Act in 2021-22 and/or have previously submitted to the KP90 return.

The purpose of this information is to:

• support users in the assessment of data quality due to changes in organisational coverage, resulting from the change in data source.
• provide information to support ongoing data quality improvement initiatives.

This year we have again produced an additional data quality visualisation which contains much of the information in these tables. Due to the benefits of interactivity, users may prefer to investigate these trends at the Mental Health Data Hub. Users should refer to the ‘Provider-level coverage’ visualisation at the Mental Health Act page.

Table 7: Organisations that submitted fewer than 12 months' Mental Health Act data to the MHSDS in 2021-22

Table 8: Organisations that submitted data to the ECDS collection in 2021-22

Table 9: Organisations submitting data to KP90 in 2015-16 but not submitting data about the Act to either MHSDS or the ECDS collection in 2021-22

Independent sector hospitals must register with the CQC on an annual basis. As the registration status and ownership of ISPs is subject to frequent change, where organisations are not registered with ODS, local knowledge may be required to determine their eligibility to submit data to MHSDS.

Please note this list excludes 33 small independent hospitals with a combined total of 95 detentions in 2015-16. These were not reported separately in the 2015-16 CSV file as they either had no Organisation Data Service (ODS) code and/or represented very small numbers of detentions. Please also note that some of the organisations listed will have closed since the 2015-16 KP90 collection. Details of the current status of each organisation can be obtained from the ODS portal.

Table 10: Selected providers used in comparison of detentions, 2015-16 to 2021-22

We have provided counts of people detained at year-end for each of the annual periods, as these helped to inform the process for selecting providers whose data was judged to be reliable. We recognise that the number of detained patients is also limited by capacity and organisational changes (e.g., mergers and ward closures) will result in changes in numbers that do not result from data quality issues.