NHS Digital Data Release Register - reformatted

NHS England (temporary National Repository) projects

5 data files in total were disseminated unsafely (information about files used safely is missing for TRE/"system access" projects).

Project 1 — NIC-59012-K2Q5T

Type of data: information not disclosed for TRE projects

Opt outs honoured: N ()

Legal basis: Health and Social Care Act 2012

Purposes: ()

Sensitive: Sensitive

When:2016.12 — 2017.02.

Access method: Ongoing

Data-controller type:

Sublicensing allowed:


  1. SUS (Accident & Emergency, Inpatient and Outpatient data)
  2. Mental Health Minimum Data Set
  3. Mental Health and Learning Disabilities Data Set
  4. Mental Health Services Data Set
  5. Improving Access to Psychological Therapies Data Set


To ensure that NHS England can meet its statutory duties (as per NHS Act 2006 and the Health and Social Care Act 2012 s13N,s23) and to meet the requirements of the Five Year Forward View, the objective for processing can be summarized as the provision of an ad-hoc and routine analysis and reporting service to support the work of NHS England (NHSE) in the following responsibility areas:

1. Proactive management of commissioned services – including contract management, performance management, needs and inequalities analysis, benchmarking, service review and development, planning, budgets and allocations and general commissioning assurance activities
2. Analysis and reporting to support QIPP (Quality, Innovation, Productivity and Prevention) programme activities
3. Data quality analysis and data quality management, to ensure data processing has been carried out effectively
4. To engage the Health Foundation to provide their analytical expertise to the Health Data Lab project using SUS data only.

The objectives for processing the Mental Health (MHMDS, MHLDDS, MHSDS) and Improving Access to Psychological Therapies (IAPT) data are specified in more detail below:

Mental Health (MHMDS, MHLDDS, MHSDS)

Despite previous initiatives such as the 2011 mental health strategy, challenges with system-wide implementation coupled with an increase in people using mental health services has led to inadequate provision and worsening outcomes in recent years, including a rise in the number of people taking their own lives. NHS England and the Department of Health published Future in Mind in 2015, which articulated a clear consensus about the way in which NHS England can make it easier for children and young people to access high quality mental health care when they need it. The 2016 Five Year Forward View for Mental Health report from the Mental Health Taskforce builds on this strategy and sets out the start of a ten-year journey for the transformation which clearly states the role that NHS England has to play.

The Mental Health data is crucial in monitoring progress against the Five Year Forward View. In particular, it will help:

• Understand current patient pathways, what care is available now and what level of referrals to mental health services are anticipated to ensure 70,000 additional children and young people each year will receive evidence-based treatment.
• Ensure that there will be the right number of CAMHS T4 beds in the right place reducing the number of inappropriate out of area placements.
• Support at least 30,000 additional women each year to access evidence-based specialist perinatal mental health treatment.
• Ensure that appropriate services are being commissioned to reduce the premature mortality of people living with severe mental illness (SMI); and 280,000 more people having their physical health needs met by increasing early detection and expanding access to evidence-based physical care assessment and intervention each year.
• Ensure people with SMI can access evidence based Individual Placements and Support (IPS)
• Ensuring that at least 60% of people with first episode psychosis starting treatment with a NICE-recommended package of care with a specialist early intervention in psychosis (EIP) service within two weeks of referral.
• Support a comprehensive programme of work to increase access to high quality care that prevents avoidable admissions and supports recovery for people who have severe mental health problems and significant risk or safety issues in the least restrictive setting as close to home as possible.
• Improve the quality of services commissioned, the case-mix of patients in treatment, population needs, the differences in success of treatment and care at practice, CCG, provider level and other geographies (e.g. regions) as well as the impact on other parts of the healthcare system, e.g. A&E.
• Improve outcomes and tackle inequalities of people with MH problems.
• Provide insight into suicide by looking at those with prior mental health problems, the severity and length of the problems and how many of those committing suicide also had wider physical health problems to help reduce the number of people taking their own by lives.
• Enable the robust quality and performance planning and monitoring at a local and national level.
• Make availability of home treatment visible in every part of England as an alternative to hospital
• Check provision of all-age mental health liaison services to meet the national commitment that at least 50% will meet the service standard

MHSDS data has also been expanded to include extensive information on people with learning disability and/or autism. The annual learning disability provider census, which ran from 2013-15 has been stood down, and all relevant content is now included within MHSDS. In addition, the content of the commissioner-based Assuring Transformation (AT) data collection has been included within MHSDS, with a goal to stand down AT when MHSDS data quality and completeness reach acceptable levels. Both the census and AT cover only inpatient care. There is currently no other data set which gives details of specialist community and outpatient services used by people with learning disability and/or autism.

NHS England therefore needs to be able to monitor the quality and completeness of Mental Health data, so that the data can become the single, definitive source of information about people with learning disability and/or autism using NHS-funded services. As there is a requirement for further segmentation beyond the existing Data Quality reporting by NHS Digital, patient-level data is required. This is also true for other elements of Mental Health data (e.g. early intervention in psychosis) where NHS England have set-up aggregate data collections from providers until the quality of MHSDS can be improved. This increases burden and causes confusion.

Detailed patient-level data is also required to compare Assuring Transformation and MHSDS inpatient data. This is necessary to identify under- and over-reporting in MHSDS (compared to AT) and to identify where patient records are inconsistent across the two data sets. Assuring Transformation is currently being used to monitor inpatient trajectories as part of the three-year national transformation plan ‘Building the right support’. If the monitoring data set switches to MHSDS before the end of this three-year period, NHS England needs to have absolute confidence that the two data sets are comparable and compatible.


The Improving Access to Psychological Therapies (IAPT) programme began in 2008 and has transformed treatment of adult anxiety disorders and depression in England. Over 900,000 people now access IAPT services each year, and the Five Year Forward View for Mental Health committed to expanding services further alongside improving quality. IAPT services provide evidence based treatments for people with anxiety and depression (implementing NICE guidelines).

The use of IAPT data will support the following priorities for service development:

• Expanding services so that at least 1.5m adults access care each year by 2020/21. This means that IAPT services nationally will move from seeing around 15% of all people with anxiety and depression each year to 25%, and all areas will have more IAPT services.
• Focusing on people with long term conditions. Two thirds of people with a common mental health problem also have a long term physical health problem, greatly increasing the cost of their care by an average of 45% more than those without a mental health problem. By integrating IAPT services with physical health services the NHS can provide better support to this group of people and achieve better outcomes.
• Supporting people to find or stay in work. Good work contributes to good mental health, and IAPT services can better contribute to improved employment outcomes.
• Improving quality and people’s experience of services. Improving the numbers of people who recover, reducing geographic variation between services, and reducing inequalities in access and outcomes for particular population groups are all important aspects of the development of IAPT services.

In addition, there is a strong policy need to understand the linkage between physical and mental health. Physical and mental health are closely linked – people with severe and prolonged mental illness are at risk of dying on average 15 to 20 years earlier than other people – one of the greatest health inequalities in England. Two thirds of these deaths are from avoidable physical illnesses, including heart disease and cancer, many caused by smoking. In addition, people with long term physical illnesses suffer more complications if they also develop mental health problems.

To better understand the relationship between physical and mental health, NHS England intend to link SUS, Mental Health data and IAPT record level data that has been anonymised in accordance with the ICO Anonymisation Code of Practice using a consistent pseudonym which has been derived for commissioning purposes.

This is an area where the evidence is currently relatively weak. Linking SUS, Mental Health and IAPT data will ensure commissioners can understand full patient pathways for their patients and plan their care, for example NHS England cannot currently answer questions such as whether patients with MH issues are at a higher risk of particular outcomes (e.g. admissions, readmissions, increased lengths of stay).Therefore linking data is an important requirement.

Expected Benefits:

1. Analysis and reporting will help ensure that NHS England meets its statutory duties (as outlined above) to commission effective and efficient services in line with NHS England’s Five Year Forward View.

2. tNR to act as a proving ground for the Commissioner Assignment Methodology (CAM) and to convert the CAM methodology to a system algorithm.

3. Support analysis of development and monitoring outcomes for new models of care.

4. Developing improved methodology for calculation of commissioner budget allocations.

5. Provides robust findings on which complex changes to care are most effective, enabling large transformation programmes to improve the effectiveness of their interventions. For example, SUS data has been used extensively (monitoring trends in acuity of cases, investigating the characteristics of attenders, understanding the relationship between attendances and admissions, etc.) in the development of the recent A&E Plan.

6. Enable NHS England to make better use of existing data, without compromising data security and by using data that is anonymised in line with the ICO Anonymisation Code of Practice to mitigate the risk of compromising patient privacy.

7. Reduced resources whilst delivering robust assessment of national programmes.

8. There have already been significant benefits realised from the use of activity data derived from SUS. NHS England now share a common understanding of activity levels across the system, which has enabled better local and regional performance management, as well as the development of national policies e.g. new demand and capacity plans for elective care. Better activity data has also enabled a more robust national planning process, and so improved the allocation of funds across the system.

9. Increased access to Mental Health and IAPT data are widespread given the relative lack of evidence (as compared to measuring physical health), despite £34 billion being spent each year on mental health (source: MH FYFV). The data will allow us to better monitor (for example by looking at local variation or the links with physical health) progress against some of the priority actions identified in the MH FYFV, such as waiting time standards for early intervention in psychosis. Data access will facilitate the development of new standards e.g. on eating disorders or out of area placements (where patient-level data will allow us to monitor the impact of various thresholds). To monitor progress against policy programmes NHS England need high quality data, and access to Mental Health and IAPT will allow the Data Controller (NHS England) to assist in driving up quality, and cease the aggregate data collections which are currently in place (so reducing burden on providers and administrative costs).


1. Access to this data allows NHS England to meet its ongoing statutory duties under the NHS Act 2006 and the Health and Social Care Act 2012 s13N, s23. Specifically – ‘to exercise its functions ensuring that health services are provided in an integrated way where this would improve quality and outcome of services and reduce inequalities’.

2. Data quality improvements initiatives including reports to ensure that NHS England data processing has been carried out correctly (e.g. expected volume of specialised activity service line codes derived).

3. Provision of an aggregate activity and finance report which will be used to populate an NHS England integrated activity and finance report for the monthly NHS England Executive Group Meeting. This has now been introduced (the benefits from this, and related SUS analyses included in the following section).

4. Analysis of the impact of changes to NHS commissioning business rules (e.g. tariff changes, commissioner assignment, specialised services identification rules, HRG grouping).

5. Gap and reconciliation analyses between monthly activity returns versus SUS data.

6. Gap and reconciliation analyses between aggregate contract monitoring reports submitted to DSCROs versus SUS data.

7. Analysis to facilitate proactive management of NHS England directly commissioned services using pseudonymised or aggregate data only. (This is dependent on the analysis requirement as to whether the output used is pseudonymised or aggregate data.)

8. Enhanced statistical analysis to facilitate proactive management of transformation programmes by local health systems on behalf of NHS England.

9. A Mental Health Five Year Forward View (5YFV) dashboard; delivered in response to the recommendation in the 5YFV. NHS England recently published a first version of this dashboard, which will allow us to hold national and local bodies to account for implementing the 5YFV strategy. The dashboard is structured around the core elements of the MH programme as set out in the 5YFV implementation plan, and include perinatal mental health, children and young people’s mental health and elements across the common, crisis and secure adult mental health pathway including health and justice and suicide prevention. NHS England require improved Mental Health/IAPT data to further develop some of the indicators in the dashboard.

10. To use the Mental Health data to support contract payment and clinical case management (and develop a reliance in this data flow akin to acute services and their use of SUS data).

11. Regular monitoring reports of commissioners (inpatient services) to meet NHS England’s statutory duties and to demonstrate the delivery of NHS England’s Learning Disability Programme by cross-referencing relevant activity with Assuring Transformation data, due to end in 2018.

12. Monitoring and analysis of outpatient and community services; alternatives to inpatient care.

13. Monitoring and analysis of new patient care pathways introduced to support the transformation of services for people with learning disability and/or autism. Access to data will specifically allow:
- Analysis of inpatient services and activity for people with learning disability and/or autism
- Analysis of outpatient and community services and activity for people with learning disability and/or autism
- Analysis of patient pathways as patients move between services

14. Analysis of factors that result in high service usage.

15. Analysis of the usefulness of diagnosis coding. Analysis will firstly focus on an understanding of the completeness and quality of coding in the dataset to provide a basis for any further analysis. NHS England would like to understand the completeness and validity of this data item, as well as identifying any geographical trends or particular providers which show problems with coding completeness. Access to the data would enable further discussion of coding practices in providers for casemix complexity. The intelligence can be shared through commissioning routes to help drive up coding completeness and accuracy to make any subsequent analysis more meaningful.

16. Analysis of the spread of diagnoses geographically and demographically, to identify any trends as well as diagnoses recorded over time (given a robust starting point for coding accuracy and completeness). Admissions and readmissions and activity could also be analysed by diagnosis to better understand these trends and potential differences in provider models to inform commissioning decisions and service improvement.

17. Provision of intelligence to commissioners to support the reduction of unnecessary restraint and potentially abusive restraint. An analysis of restraint to identify any trends or outliers across providers, CCGs and sub-regions. The analysis will also include the frequency of restraint per patient and by ward type. This will highlight any areas for concern in the use of restraint to inform further discussions with commissioners. As the restraint type is added to the MHSDS in v2.0 this will provide further insight and areas for focus in discussions with commissioners. The aim of this is to provide intelligence to commissioners to support the reduction of unnecessary restraint and potentially abusive restraint.

18. To support ongoing updates to the Mental Health Quality Dashboard using quality measures derived from the MHMDS and MHLDDS. (The current dashboard is under review to focus the measures further on quality and utilising the dataset will enable a wider availability of measures as well as robust data. The dashboard can be used by QSG, commissioners and providers for benchmarking and identifying areas for service improvement as well as to inform commissioning decisions.)

19. To support the development of Clinical Services Quality Measures (CSQMs) that provide an at-a-glance indication of how well services are performing. They have been/will be developed as composite measures for Psychosis and Dementia specifically as a series of metrics that, for example, will allow for comparisons between services such as units within hospitals; providing better information for patients clinicians and citizens. Supressed numbers currently available in the published reports do not allow annual aggregation to be input into the composites. The measures will be developed according to statistical principles and will be assured by clinical and technical experts. (NHS England is involving patients, the public, service providers and clinicians in the development of these measures with aggregate – service level information to be available via NHS Choices and My NHS.)

20. To achieve the service improvements required, in association with the findings from the report “The commissioning of specialised services in the NHS” by the National Audit Office (NAO), whereby the findings suggested that NHS England does not have sufficient information to drive service improvement in specialised commissioning.

The target commencement date for the above outputs is December 2016, with the aim to monitor changes on a monthly basis going forward.


Data processed by DSCRO Arden and GEM is anonymised in accordance with the ICO Anonymisation Code of Practice before being securely transferred to Arden and GEM Commissioning Support Unit (CSU) where NHS England’s tNR is hosted.

Under strict access controls, NHS England’s analysts will use remote access arrangements to query the pseudonymised record level data which is held within the tNR in order for them to analyse the data. The data can be accessed remotely from multiple locations in England using secure VPN or the N3 network, depending on where NHS Analysts are based. Access is secured via two personal user IDs and passwords; one to login in the terminal services server giving access to the Arden GEM network domain and then a further login into the SQL Server environment where the user is given read-only access to the data.

NHS England has implemented strict access controls to limit the amount of data which is made available to analysts. In order for users to access data on the tNR they have to be approved by the Information Asset Owner (IAO). Once access to data on the tNR is granted, depending on the role and need of the user, access is secured by using 2 factor authentication via VPN and on the N3 network. This, together with the restrictions outlined in the NHS Digital data sharing framework contract and data sharing agreement, minimise the risk of inadvertent or malicious re-identification. NHS England believe that the wider benefits of ensuring that patients receive the most appropriate care therefore outweigh the low, and managed, risk of harm from the processing activities.

The processing activities are described in more detail below.

1. A national feed of SUS identifiable data will be transferred from SUS to Data Services for Commissioners Regional Office (DSCRO) AGEM who will complete data quality, pseudonymisation and validation of the SUS data.

2. A national feed of Mental Health (MHMDS, MHLDDS and MHSDS) and Improving Access to Psychological Therapies (IAPT) data will be made available by NHS Digital to Data Services for DSCRO AGEM who will undertake data quality checks, pseudonymisation and validation of the Mental Health and IAPT data.

3. The DSCRO will apply the same pseudonymisation key used for the SUS data to the MHMDS, MHLDDS, MHSDS and IAPT in order to enable linkage by the Data Processor (within the tNR).

1. DSCRO AGEM transfer the data (anonymisation in line with the ICO Anonymisation Code of Practice) to Arden and GEM CSU who act as NHS England’s data processor and where the data will be stored on a repository server, known as the temporary national repository (tNR).

4. The data will be processed in the tNR on behalf of NHS England (as recipient data controller) to meet the reporting requirements, by adding value to the data (e.g. adding a tariff and grouper) to support integrated patient care analysis.

5. The pseudonymised SUS, Mental Health (MHMDS, MHLDDS, MHSDS) and IAPT data is made accessible to NHSE analysts via two personal user IDs and passwords; one to login in the terminal services server giving access to the Arden GEM network domain and then a further login into the SQL Server environment where access is given read-only to the data.

6. Data processing staff NHSE contract with (The Health Foundation (THF) - will access the system in exactly the same way as NHS England analysts, to monitor and evaluate NHS England funded transformation programmes set out on the Five Year Forward View on SUS data only. THF cannot access the Mental Health or IAPT data held within the tNR. Access to Mental Health and IAPT data are only available to NHSE analysts.

7. A robust user registration process is in place involving sign-off by the analyst’s line manager to ensure that all users have a suitable level of knowledge about SUS, SQL Server and tNR processed data. The pseudonymised SUS data is made accessible to THF analysts working closely with NHS England analyst using the same access methodology as described in 7 above. The data will be processed by THF on behalf of NHS England to monitor and evaluate NHS England funded transformation programmes set out in the Five Year Forward View.

8. As part of the monitoring and evaluating of the transformation programmes, it will be necessary for the processed data to be enhanced by linking in publicly available contextual information on aggregate level. Examples of publicly available data include GP patient survey result aggregated to GP practice level (source: https://gp-patient.co.uk/surveys-and-reports), measures of deprivation aggregated at LSOA level* (source: https://data.gov.uk/dataset/english-indices-of-deprivation-2015-lsoa-level) and disease prevalence, again geographically aggregated (source: https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases).

9. Data linkage between SUS, Mental Health (MHMDS, MHLDDS, MHSDS) and IAPT will be undertaken on pseudonymised record level data (anonymised in accordance with the ICO Anonymisation Code of Practice) held within the tNR by NHS England data analysts operating under strictly controlled conditions and any inadvertent or malicious re-identification of data subjects will be recorded and reported in line with the NHS England’s incident (disciplinary) management process and appropriate action taken.

Data will only be shared with or processed by the parties listed in this application and will only be used for the purposes stipulated.

Any reports sent beyond the data controller and processors listed in this application would contain aggregate data only, which would be subject to the disclosure controls of the relevant datasets.

*a LSOA is a small geographical area typically covering about 1500 people