GP Contract Services: Supporting Information

This page provides all of the relevant information regarding the GP Contract Services publications. 

It should be used in conjunction with the GP Contract Services statistical publications produced by NHS England. 

Background information on the GP Contract Services as a whole is included here for users who are unfamiliar with these types of programmes. This background information covers what GP Contract Services are, as well as how and why GP Contract Services data are collected.

Caveats surrounding the GP Contract Services data are included in this document as it is vitally important that users fully understand these caveats when referring to the data in the GP Contract Services publications, or when carrying out their own analysis on the GP Contract Services data.

In previous years these data were published as Official Statistics however they will now be published as management information in line with the other GP contract services. A full review as to why this decision was taken can be found on the INLIQ review page. The data are subject to some minor data quality checks as in previous years, but the outcomes of these are not published alongside the data. 

In previous years, these data were released along with all other GP Contract Services data as a Primary Care Network (PCN) Extract. In response to user feedback, these data are now published on a monthly basis. The data are subject to some minor data quality checks as in previous years, but the outcomes of these are not published alongside the data. NCDES data releases can be found at NCDES publication page.

In previous years, these data were released along with all other GP Contract Services data. In response to user feedback, these data are now published on a monthly basis. 

The data are subject to quality checks, the results of which are included with the data in the .csv file. These checks, and the data items themselves, are defined in the accompanying Data Dictionary. Learning disabilities health checks data releases can be found at LDHC publication page.

From 2022-23 these services form part of the Core GP Contract data collection and are published together in one csv file. The data is subject to some minor data quality checks as in previous years, but the outcomes of these are not published alongside the data.  In 2022-23 this was released as an annual publication, however in response to user feedback publication in 2023-24 will be quarterly.

Prior to the 2017-18 reporting year, data for all GP Contract Services were published as Experimental Statistics. For these releases, the data for each service were subject to a series of data quality checks. The results of these checks were published along with the data in Excel workbooks, in addition to the underlying data in .csv files.

Feedback from these releases highlighted that the majority of the GP Contract Services data received little usage. Therefore, from 2017-18, the data for these services have been released in .csv format only, and have not undergone any data quality checks. These data should therefore be interpreted and used with caution, as the quality of the data is unknown. Users of these data may wish to refer to the 2015-16 publications for details of the quality checks that were conducted previously.

GP Contract Services are primary care programmes that have data collected by NHS England. They form part of the General Medical Services (GMS) contract, which is agreed between NHS England and the British Medical Association’s (BMA) General Practitioners Committee (GPC). The GMS contract covers the delivery of primary care services across England.

GP Contract Services include:

• Core contract components: These are core contract requirements that are part of the GMS contract; GP practices are required to carry out these services as set out in the GMS Regulations.
• Enhanced services (ES): These are voluntary reward programmes that cover primary medical services; one of their main aims is to reduce the burden on secondary care services.
• Vaccination and immunisation programmes: These are programmes that are commissioned for delivery by GP practices following recommendations from the Joint Committee on Vaccinations and Immunisations (JCVI).
• Quality and Outcomes Framework (QOF)[1]: This is a voluntary annual reward and incentive programme for all GP practices in England.
• Indicators no longer in QOF (INLIQ): These are indicators that have been removed from QOF but are still used for management information purposes.

Data on each of the GP Contract Services are collected by NHS England via the Calculating Quality Reporting Service (CQRS) and General Practice Extraction Service (GPES). There are two methods of collecting the data: 1) via automated data collections and 2) via manual data collections.

GP practices record patients’ medical information (e.g. diagnoses, procedures of care, prescriptions, etc.) in their GP practice clinical systems. This information is gathered in the form of data, which can then be collected by GPES. The data for each GP Contract Service are specified in the form of Business Rules, which are the technical specification documents that underpin how each GP Contract Service count/indicator works. These specifications are used so that only the necessary data within the appropriate time period are collected; this is referred to as a data extraction. The data collected by GPES are displayed in the CQRS so that GP practices can check that the data collected are the same as the data held in their clinical system. 

Automated data collection example: Learning disabilities health check scheme 2018-19

This ES consisted of one payment count and 14 management information counts.

Using the payment count (i.e. LD001) as an example, this was:

“Quarterly count of registered patients aged 14 years or over, at the 31 March 2019, on the practice’s Learning Disability register who have received a learning disability health check by the GP practice and have not received a health check in a previous quarter in this financial year.”

A GP practice had 100 patients aged 14 years or over, as at 31 March 2019, on their learning disability ES register. Within the reporting period (i.e. quarter) in question, 25 of these patients received a learning disability health check; note that these patients had not received a learning disability health check in a previous quarter. This information would be stored in the GP practice’s clinical system. Therefore, a specification would be used to extract only the necessary data over the necessary time period. This would result in the number 25 being returned for the LD001 count for the quarter in question. The GP practice would then be able to view this number on the CQRS to check that it is the same as the data that is held in their clinical system. This number would then appear in the data for this ES.

This involves GP practices logging in to the CQRS and manually entering the specified activity that has been carried out on the patients registered at their GP practice. Once the information has been entered in to the CQRS, it is processed in the same way as the automatically collected data. GP Contract Service data are collected manually if the GP Contract Service has not been set up for an automated data collection or if the GP practice is unable to have their data automatically collected. Manual data collections are also available until GPES comes on line to support relevant services.

Manual data collection example : Season influenza vaccination programme 2018-19

This vaccination and immunisation programme consisted of five payment counts and 17 management information counts.

Using the payment count (i.e. SFLU001) as an example, this was:

“Monthly count of patients aged 65 and over on 31 March 2019, who have received a seasonal influenza vaccination by the GP practice, within the reporting period.”

A GP practice had 100 patients aged 65 years or over, as at 31 March 2019, registered at their practice. In the reporting period (i.e. month) in question, 50 of these patients were administered a seasonal influenza vaccination by the GP practice. Therefore, the GP practice would have to log in to the CQRS and enter the number 50 for the above SFLU001 count for the month in question. This number would then appear in the data for this programme.

GP Contract Service data are primarily used for payment and management information purposes; however, the data may be more widely used to help support commissioning, planning and policy decisions.

Some GP Contract Services are associated with payments, which are used to reward GP practices for the activity that has been carried out. An example of this type of GP Contract Service is the QOF, which is an annual incentive scheme that rewards GP practices for the provision of 'quality care' and helps to standardise improvements in the delivery of primary medical services.

Other GP Contract Services are not associated with a payment as the main purpose of these GP Contract Services is for management information. An example of this type of GP Contract Service is the Alcohol related risk reduction scheme for all patients 2018-19, which is a contractual requirement set out in the GMS Regulations.

The frequency of when data are collected for each GP Contract Service varies. Some GP Contract Services require data being collected on a monthly basis, whereas others require data being collected on a quarterly or annual basis. This means that the number of data collections for each GP Contract Service also varies. If a GP Contract Service has data collected on a monthly basis and the service runs for a one year period then this will result in 12 data collections in total. In contrast, if a GP Contract Service has data collected on a quarterly basis then this will result in four data collections during the course of a one year period.

The start date and end date of each GP Contract Service also varies, so a GP Contract Service that has data collected on a monthly basis may not necessarily result in 12 data collections in total. For example, a GP Contract Service such as Seasonal and Childhood Flu Vaccinations with a start date of 1 September 2018 and end date of 31 March 2019 would result in seven data collections in total if the data were collected on a monthly basis.

For GP Contract Services that have an associated payment, the frequency of when each payment is made to GP practices varies. The payment frequency and data collection are the same for some GP Contract Services. For example, a GP Contract Service may have data collected on a monthly basis and GP practices may also receive payment on a monthly basis. Alternatively, the payment frequency and data collection frequency may be offset for some GP Contract Services. For example, a GP Contract Service may have data collected on a monthly basis but GP practices may only receive payment on a quarterly, biannual or annual basis.

GP Contract Services are comprised of either indicators or counts;

Many GP contract services use indicators, of which there are two types: 1) register indicators and 2) non-register indicators.

Register indicator

This is a Boolean indicator meaning that there are only two possible options: true or false. No numerator and denominator are required for this type of indicator.

Register indicator example: INLIQ 2018-19

An example of a register indicator is one of the indicators in the Thyroid INLIQ 2018-19 ruleset (i.e. THY001):

“The contractor establishes and maintains a register of patients with hypothyroidism who are currently treated with levothyroxine.”

The only two possible options available for this indicator are either true (the contractor does establish and maintain a register) or false (the contractor does not establish and maintain a register). GP practices have to ensure the data in their clinical systems are in line with the THY001 indicator criterion to support the data collection in order to ‘achieve’ this indicator.

Non-register indicator

This consists of a numerator and denominator. The numerator is the number of patients who have received the care described in the indicator, whereas the denominator is the number of patients eligible for the care described in the indicator minus those who have been excluded, exception reported, or have had a personalised care adjustment (PCA) applied.

Non-register indicator example: INLIQ 2018-19

An example of a non-register indicator is one of the indicators in the Thyroid INLIQ 2018-19 ruleset (i.e. THY002):

“The percentage of patients with hypothyroidism, on the register, with thyroid function tests recorded in the preceding 12 months.”

The numerator for this indicator is the number of patients on a GP practice’s thyroid register (note that being on the thyroid register denotes having hypothyroidism) who have had thyroid function tests recorded in the preceding 12 months. The denominator for this indicator is the number of patients on a GP practice’s thyroid register who have either had thyroid function tests recorded in the preceding 12 months or do not have a recorded reason for not having had thyroid function tests recorded in the preceding 12 months.

Exceptions, personalised care adjustments (PCAs) and exclusions

Exception reporting, or PCAs apply to indicators where the achievement is determined by the percentage of patients receiving the specified level of care (i.e. non-register indicators). In contrast, exclusion reporting can apply to any indicator.

Exceptions / PCAs

This is where patients who would ordinarily be included in the denominator of an indicator (after exclusions) are excepted from the indicator on the basis of one or more exception/PCA criteria.

Patients are removed from the denominator (and numerator) for an indicator if they have been both excepted/PCA'd and they have not received the care specified in the indicator wording. If the patient has been excepted/PCA'd but subsequently the care has been carried out within the relevant time period, the patient will be included in both the denominator and the numerator (i.e. achievement will always override exceptions/PCAs).

When an appropriate exception/PCA code has been added to the patient record, it applies only to the service year in which it was added. If the timeframe defined to deliver the care described in the indicator wording spans two service years, the exception/PCA would need to be added for each of these years.

Exception example: INLIQ 2018-19

An example of exceptions is in one of the indicators in the Thyroid INLIQ 2018-19 ruleset (i.e. THY002):

“The percentage of patients with hypothyroidism, on the register, with thyroid function tests recorded in the preceding 12 months.”

Following on from the above non-register indicator example, the recorded reasons for not having had thyroid function tests recorded in the preceding 12 months for this indicator are:

• • If a patient recently registered at the GP practice within the last three months of the financial/INLIQ year; the rationale behind this exception is that the GP practice may not have had sufficient time to treat the patient since they registered at the GP practice.
  • If a patient had a hypothyroidism exception reporting code entered in their patient medical record within the last 12 months; this exception reporting code would be entered in a patient’s medical record if the patient was unsuitable for the treatment referred to in this indicator.
  • If a patient was recently diagnosed with hypothyroidism within the last three months; the rationale behind this exception is that the GP practice may not have had sufficient time to treat the patient since they were diagnosed with the condition.

These three types of exception (i.e. recently registered, a specific exception reporting code and recently diagnosed) are used in many indicators that make up the GP contract services.

See the NHS Employers GMS Contract QOF Guidance for further information on exception reporting/PCAs.

Exclusions

This is where patients who do not qualify for an indicator for a definitive reason are excluded from the indicator. Such reasons may include patients not being the necessary age/sex, or not meeting a specific status. Exclusions are specific to each indicator; patients may still be included on the overarching disease register(s) that indicators are applied to. It is important to note that patients are excluded due to ineligibility before having chance to be included in the indicators (see the example below).

Exclusion example: INLIQ 2018-19

An example of an exclusion is in one of the indicators in the Diabetes INLIQ 2018-19 ruleset (i.e. DM016):

“The percentage of male patients with diabetes, on the register, who have a record of erectile dysfunction with a record of advice and assessment of contributory factors and treatment options in the preceding 12 months.”

This indicator only includes male patients; therefore, all female patients are excluded from this indicator.

Core contract components, ESs and vaccination and immunisation programmes use counts, of which there are two types: 1) counts of patients and 2) counts of specific criterion.

Count of patients

This is a count of the number of patients who meet a specific criterion, such as the number of patients with a certain diagnosis or the number of patients who have received the specified care/medication.

Counts of patients are used for all automated core contract component, ES and vaccination and immunisation programme data collections via GPES. Once a patient is included in a count, they will not be added to the same count.

For example, an automated count created to count the number of patients that received a vaccination given by a GP practice in a reporting period (i.e. month). The GP practice administered 50 vaccinations to 40 patients during the month in question; 30 of these patients received one vaccination each and the remaining 10 patients received two vaccinations each. Since GPES is only able to count the number of patients who meet a specific criterion, this automated count would return the number 40 (i.e. the number of patients who received the vaccination, rather than the total number of vaccinations that were administered).

Count of specific criterion

This is a count of the number of events/actions that meet a specific criterion, such as the number of vaccination doses administered by a GP practice.

Counts of the number of specific criterion are only used in manual core contract component, ES and vaccination and immunisation programme data collections via the CQRS (note that counts of patients are also used in many manual data collections). GP practices have to log in to the CQRS and manually enter the specified activity. In the example above, the GP practice would log in to the CQRS and enter the number 50 (i.e. the total number of vaccinations that were administered, rather than the number of patients who received the vaccination).

The learning disabilities health check scheme includes counts and indicators; the indicators include personalised care adjustments (PCAs) and exclusions. Explanations of these counts, indicators and their components can be found on the LDHC supporting information page.

All counts include activity that takes place from a point in time up to and including the end date of a reporting period. This timeframe in which activity takes place (window of activity) varies across each GP Contract Service and is dependent on whether a count is cumulative, non-cumulative or a combination of cumulative and non-cumulative.

Please note that the term ‘cumulative’ is not used as it is defined in the dictionary[1], which is “increasing or growing by accumulation or successive additions.” Instead, a cumulative count is where the window of activity increases in duration for each subsequent data collection that makes up the count. The point in time from which activity is included is the same for all of the data collections, whereas the end date of the reporting period increases for each subsequent data collection. This increase depends on the data collection frequency of the count. If the count involves monthly data collections then the window of activity will increase in duration by one month for each subsequent data collection; if the count involves quarterly data collections then the window of activity will increase in duration by one quarter for each data collection.

As a general rule of thumb, cumulative count titles usually end with “up to the end of the reporting period.”

Cumulative count example: Learning disabilities health check scheme 2016-17

An example of a cumulative count is one of the counts in the Learning disabilities health check scheme 2016-17 ruleset (i.e. LDHCMI005):

“Quarterly (cumulative) count of registered patients aged 18 years or over, as at the Quality Service End Date, identified as having a QOF diagnostic learning disability, as at reporting period end date.”

This ES was set up to have data collected on a quarterly basis. Each data collection for the LDHCMI005 count included activity that had taken place on or after 1 April 2016 and up to and including the end date of the reporting period (i.e. quarter) in question.

The data collection for the first quarter would have included activity from 1 April 2016 to 30 June 2016  then they would be included in the LDHCMI005 count for the first quarterly collection.

The data collection for the second quarter would have included activity from 1 April 2016 to 30 September 2016  Being identified as having a QOF diagnostic learning disability on 15 May 2016 would still qualify the patient to be included in the LDHCMI005 count for the second quarterly collection.

The data collection for the fourth quarter would have included activity from 1 April 2016 to 31 March 2017 . As with the second and third quarters, being identified as having a QOF diagnostic learning disability on 15 May 2016 would still qualify the patient to be included in the LDHCMI005 count for the fourth quarterly collection.

As you can see, regardless of the quarter, the start date has remained the same (i.e. 1 April 2016), whilst the end date of the reporting period increases for each subsequent collection. This results in the window of activity increasing in duration for each subsequent data collection. 

This is a count where the window of activity remains the same duration for each subsequent data collection that makes up the count. The point in time from which activity is included and the end date of the reporting period both increase in parallel for each subsequent data collection. If the count involves monthly data collections then the point in time from which activity is included and the end date of the reporting period will both increase by one month for each subsequent data collection; if the count involves quarterly data collections then the point in time from which activity is included and the end date of the reporting period will both increase by one quarter for each subsequent data collection. This means that the duration of the window of activity remains the same for all of the data collections that make up the count.

As a general rule of thumb, non-cumulative count titles usually end with “within the reporting period.”

Non-cumulative count example: Learning disabilities health check scheme 2016-17

An example of a cumulative count is one of the counts in the Learning disabilities health check scheme 2016-17 ruleset (i.e. LDHC001):

“Quarterly count of registered patients aged 14 years or over, as at the Quality Service End Date, on the practice’s Learning Disability register who have received a learning disability health check and have not received a health check in a previous quarter in this financial year.”

This ES was set up to have data collected on a quarterly basis. Each data collection for the LDHC001 count included activity that had taken place on or after the start date of the reporting period and up to and including the end date of the reporting period (i.e. quarter) in question.

The June 2016 data collection would have included activity from 1 April 2016 to 30 June 2016 . If a patient on the GP practice’s learning disability register received a learning disability health check during this period then they would be included in the LDHC001 count[1] for the June data collection.

The September 2016 data collection would have included activity from 1 July 2016 to 30 September 2016.   The learning disability health check that was completed on 15 May 2016 would not qualify the patient to be included in the LDHC001 count for the September 2016 data collection. This is because this activity did not take place in the window of activity for this data collection.

As you can see from the above, the start date of the reporting period and the end date of the reporting period both increase for each subsequent data collection. This results in the window of activity remaining the same duration for each subsequent data collection that makes up the LDHC001 count in the Learning disabilities health check scheme 2016-17 ruleset.