Announcement of methodological change Improving Access to Psychological Therapies (IAPT) reports

This paper announces and describes changes relevant to all IAPT publications based on data from the September 2020 Primary submission onwards (October 2020). This document incorporates the changes to the IAPT dataset with the introduction of v2.0 of the dataset.

Although most content has remained the same, the addition of new fields, and changes to some validation rules, acceptable values and processing methods requires an underlying change in the methodology used to produce a number of key measures in our published reports.

At the same time as updating the dataset, the submission method to the IAPT dataset has been changed to use the Strategic Data Collection Service in the cloud (SDCS Cloud). This will provide improved user experience and faster data quality feedback and a secure solution using the cloud technology of the future which will integrate with our improved data processing services.

Improving Access to Psychological Therapies (IAPT) is an NHS programme in England that offers interventions approved by the National Institute for Health and Care Excellence (NICE) for treating people with depression or anxiety.

The IAPT programme is supported by a regular return of data generated by providers of IAPT services in delivering those services to patients. These data are received by NHS Digital and published in monthly and annual reports.

The new IAPT dataset version 2.0 specification can be found on the Information Standards Board (ISB) website, 

SDCS Cloud

We are improving how we collect and process the Improving Access to Psychological Therapies (IAPT) dataset, beginning with the collection of data from October 2020. This will give us the power to process larger volumes of data, faster than ever before, whilst providing the tools to manage incoming data to ensure it is accurate, useful, and secure. It is also an internet-facing service that does not require a specialist health and social care data connection, making it easier for more providers to submit data.

New dataset version

We are updating the content and submission requirements for the IAPT dataset, including introducing new concepts such as those for Internet-Enabled Therapy (IET), integrated care and employment support. We will also be introducing SNOMED CT as a clinical terminology to align with other national and international data collections.

Patient matching methodology

We are changing how we identify people within national datasets which use this new technology, including the IAPT dataset. Using one method for doing this across datasets helps us increase the amount of usable, better quality, linkable data available to support research and planning.

Some measures in the official statistics will be calculated differently

A number of the dataset changes directly affect the official statistics, including key measures such as patient outcomes.

We are more likely to match patient records across submissions

In general, records submitted to the IAPT dataset are more likely to result in a match using the new method than using the previous one.

SDCS Cloud

IAPT submissions were previously made by data providers to NHS Digital via the Bureau Service Portal. From the implementation of IAPT dataset v2.0 onwards (October 2020), submissions have been made to NHS Digital via the Strategic Data Collection Service in the cloud (SDCS Cloud).

SDCS Cloud will provide:

• improved user experience and faster data quality feedback
• a secure solution using the cloud technology of the future which will integrate with our improved data processing services

SDCS Cloud is a completely new tool, and different from other existing NHS Digital collection tools, as it is being designed from the ground up to optimise use in the cloud. It uses a two factor authentication as the secure method of confirming user identity. It is also an internet facing service that does not require an N3 or HSCN connection, making it easier for more providers to submit data. 

Since referrals to IAPT services span multiple months, it is critical that we are able to identify patients consistently across multiple submissions, ensuring that data about the full patient journey is represented.

The Master Patient Service (MPS) helps us increase the amount of usable, better quality data available to support research and planning. It does this by verifying the demographic information contained in a person’s health and care records and matching it to their unique NHS number to confirm their identity.

Patients can visit multiple places where they register to receive care or treatment. At any given time, we store the health and care records of individuals as recorded in various systems around the country. This can create challenges, but MPS aims to match the right person with the right record with a 99% accuracy rate.

MPS uses a four-stage algorithm to provide a single best result for each record against Patient Demographics Service (PDS). Our core Data Processing Service assesses the data quality of submitted data before passing to the MPS. MPS then attempts to match all the records to a single NHS Number held in the Personal Demographic Service (PDS). MPS also checks demographic details supplied in the submitted data file, such as age, gender, and postcode, for their ‘closeness’ to the data held in PDS and produces an associated match confidence score. Any results matched under the confidence threshold will not result in a match. We will issue a new version of this notice if MPS developments result in changes to how you can interpret statistics in the IAPT official statistics series

As we develop MPS we will expand coverage to more datasets and continue to make improvements. We will issue a new version of this notice if MPS developments result in changes to how you can interpret statistics in the IAPT official statistics series.

Since employment support is part of the broader offer available to IAPT patients, a single appointment with an IAPT service could consist of only clinical therapies such as Interpersonal Psychotherapy (IPT) or Cognitive Behavioural Therapy (CBT), only employment support, or both clinical therapies and employment support.

The IAPT dataset v1.5 was not built to distinguish between employment support and clinical therapies. This meant that employment support, where recorded in the Appointment table, would be regarded as a valid therapy type given to a patient during a care contact.

This is problematic for appointments where only employment support was administered to a patient. Therefore, the IAPT dataset v2.0 now captures employment support as a coded procedure concept using SNOMED CT, making the distinction between clinical therapies and employment support clearer.

Anxiety Disorder Specific Measures (ADSMs) are questionnaires issued to patients at each contact with an IAPT service and are used to assess the severity of their condition at a point in time. They are an important mechanism for quantifying patient improvement.

There are many different patient questionnaires in use within the IAPT programme, and patients should be given the one specific to their condition at each contact – for example, patients presenting with Obsessive Compulsive Disorder (OCD) should be asked to complete the Obsessive Compulsive Inventory questionnaire.

Each questionnaire assesses severity using a Likert scale, and scores from individual questions are added together to give a number representing how severe the patient’s condition is. Differencing the first recorded score from the last recorded score is the basis for calculating patient outcomes.

The main changes to ADSMs in IAPT version are summarised below. For further information, refer to the NHS England IAPT manual. 

The Impact of Events Scale questionnaire, used for assessing Post-Traumatic Stress Disorder (PTSD), has been replaced by the PCL-5 questionnaire.

A new ADSM, called the Body Image Questionnaire, has been approved for use in the IAPT programme and will be used to assess Body Dysmorphic conditions.

Note that this questionnaire is not currently being used to assess patient outcomes, though this is planned in future

In IAPT dataset v1.5, an ADSM was only used where the initial severity of the patient’s condition (based on the ADSM total score) was above a defined threshold (known as the ‘caseness’ threshold). Where this was not the case, a Generalised Anxiety Disorder questionnaire (GAD-7) was used instead.

In v2.0, this will no longer be the case, and ADSMs will be used irrespective of initial caseness where two or more appropriate scores are recorded in the course of the referral.

The NHS England IAPT Manual recommends the expansion of service provision into Internet-Enabled Therapy (IET). This involves patients working through materials on the internet with a therapist inputting with encouragement, clarifications, and feedback at key points.

Statistics about patient outcomes in IAPT rely on comparing the first and last scores from Anxiety Disorder Specific Measures (ADSMs) taken from each patient contact with a service.

IAPT dataset version v2.0 introduces a new patient contact concept – Internet-Enabled Therapy (IET). Services must record the time a Clinician has spent with an IAPT patient between two dates. Services must also record any ADSMs that are completed by patients through the IET system. This information is captured in two new dataset tables, which are linked to the IDS101 Service or Team Referral table:

• IDS205 Internet Enabled Therapy Care Professional Activity Log
• IDS606 Coded Scored Assessment (Referral)

An Activity Log recorded on table IDS205 will count as the equivalent of an attended treatment session within the dataset. In v2.0, a referral that has two IET activity logs and/or attended treatment contacts will now be included in the published patient outcome measures like the number and proportion of referrals that have moved to recovery. Previously, only referrals that had two attended treatment care contacts were included in patient outcome measures.

Scores taken from referrals having IET are now incorporated into the calculation of patient outcomes alongside those from care contact-based referrals. This will enable a comparison of the efficacy of IET with other forms of treatment.

Referrals to IAPT services may be offered a mix of IET and traditional care contacts, and so the first and last patient questionnaire scores will be taken from both to determine the first and last scores necessary for the calculation of patient outcomes.

Other published measures that refer to the dates of the first or second treatment sessions, such as waiting times, will also include the dates of any IET activity logs in their calculations.

Employment advisors in IAPT pilot

As part of the Spending Review 2015, investment in the Employment Advisers (EA) in IAPT pilot is being taken forward by the Work and Health Joint Unit (WHU). The additional investment provided is being used to increase the number of Employment Advisers embedded in IAPT services. This supports more people with depression and anxiety to receive combined psychological therapy and employment support to:

• remain in work
• find work
• get back to work from long term absence

The EA in IAPT pilot was monitored and reported using an additional data collection alongside IAPT dataset v1.5 this was a set of tables linked to the appointment table in the IAPT dataset. 

In v2.0, concepts from the additional data collection have been incorporated into the main dataset through the following tables:

• IDS602 Long Term Physical Health Condition
• IDS603 Presenting Complaints

This allows for reporting of integrated care nationally and alters the constructions of particular associated metrics. Further details are given in the IAPT Metadata document.

Integrated care

Previously, mental and physical health care treatment have been provided by separate parts of the NHS. A pilot for co-locating physical and mental health care provision was carried out to integrate IAPT therapy into existing medical pathways and services. These services are referred to as ‘integrated services’.

The integrated services pilot was monitored and reported using an additional data collection alongside IAPT dataset v1.5 – this was a set of tables that linked to the Appointment table in the IAPT dataset.

In v2.0, concepts from the additional data collection have been incorporated into the main dataset through the following tables:

• IDS602 Long Term Physical Health Condition
• IDS603 Presenting Complaints

This allows for reporting of integrated care nationally and alters the constructions of particular associated metrics. Further details are given in the IAPT Metadata document.

Some therapy type codes have been determined as no longer required for submission in IAPT dataset v2.0, as follows:

Behavioural Activation (Low/ High Intensity) – this would typically be given as part of another therapy session (for example CBT) and should be mapped accordingly

Ante/postnatal counselling – patients in this period with relatable issues would be treated with another therapy (for example Guided Self Help).

Employment Support (Low/ High Intensity) – Employment Support is not categorised as low or high intensity and therefore a single 'Employment Support' code is provided for v2.0

Measures relating to these specific types of therapy will no longer appear in the official statistics reports. See the IAPT terminology mapping guidance.  

Additional questions have been added to the Assessment and Treatment Patient Experience Questionnaires (PEQs):

• Treatment PEQ Q6 - Did you receive the employment help that you required?
• Assessment PEQ Q4 - Did your assessment cover your employment needs?

These questions will be reported in the IAPT Annual official statistics for 2020-21. For further information, see the IAPT manual.  

The Institute for Medical Technology Assessment (iMTA) Productivity Cost Questionnaire (iPCQ), also known as the ‘presenteeism questionnaire’, is an additional patient questionnaire that relates only to patients that are in work and working.

Additional measures based on this questionnaire are currently being developed for inclusion in the official statistics.

Resources

For an explanation of all measures in the Monthly and Quarterly Activity Data File CSVs, see the IAPT Metadata Document and the Guide to IAPT data and publications.

For general information about submitting IAPT v2.0 data.

For the specification of the IAPT dataset, see the IAPT v2.0 Technical Output Specification.

For the IAPT Manual, see the NHS IAPT website with supporting information

For resources related to monthly IAPT publications and links to all historical IAPT publications.