Skip to main content

Patients Registered at a GP practice Data Quality Statement

Data quality statement for the Patients Registered at a GP Practice statistical publication.

Background

Patients registered at a GP practice is a monthly publication showing counts of patients at NHS Regions, Integrated Care Board (ICB), sub-ICB location, Primary Care Network (PCN) and GP practice by gender and age. In addition, data are presented by GP practice at Lower Super Output Area (LSOA) level by gender on a quarterly basis.


Data source

Prior to April 2024, the data source for the publication was National Health Applications and Infrastructure Services (NHAIS). This service was decommissioned.

From April 2024, the data source for the publication is Personal Demographic Service (PDS). The information about a patients registered GP is used to create aggregate data extracts and publication outputs. In the April 2024 release of the publication, outputs were produced using both data sources to allow comparison and transition. 

Comparison analysis between the data sources has been completed to assure it is fit for use however there are expected differences. The reasons we expect differences include:

  • PDS is a real time system which is different to NHAIS
  • PDS is the national master database and can be updated by methods other than general practices
  • There are differences in the way registrations and demographic data are managed in the PDS system compared to NHAIS
  • Information about practices may be applied differently, for example practice closures or mergers

Please raise any data quality queries about the Personal Demographic Service using the details for the National Service Desk

The information on this page applies to the PDS data source

 


Relevance

Data is aggregated at GP Practice-level and by age and gender. This data is used by many NHS and local authority organisations as a denominator for other local analyses and are considered essential to maintaining this level of information.

NHS England encourages feedback to develop our user knowledge and improve our publications, please use the feedback survey within the publication, or send feedback to [email protected]


Accuracy

There is a known discrepancy between the estimated size of the England population and the number of people registered at GP practices. This difference is is known as list inflation, over-coverage or ghost patients.

List inflation occurs when the National Patient Register continues to record individuals that should have been de-registered. This can happen for a variety of reasons, including - but not limited to:

  • Patients move away and do not de-register when they leave
  • Young, healthy people have moved to a new address but did not prioritise de-registering with their former GP practice
  • Patients have left the country but did not de-register
  • Shared custody of children with split residence is not flagged in the medical record resulting in double-counting
  • Duplicate records, for example due to use of different names, particularly surnames
    • For example, children are recorded against their mother’s and their father’s surname, resulting in double-counting
  • Patients in nursing or residential care are registered for that address as well as their original place of residence
  • Death

In addition to list inflation, which incorrectly increases the count of patients, there are also issues associated with under-coverage. For example:

  • Individuals move to a new area, de-register from their former GP and do not register with a new practice in a timely fashion, if at all
  • Members of the armed forces and their families are included in population estimates but are not covered in the National Patient Register. However, from a statistical perspective, adjustments are made to mitigate the effect of this
  • Babies may not be included in the National Patient Register until formally registered
  • New (or returning) migrants may delay registration with a practice
  • Individuals may be inappropriately removed from a GP list under the “no-contact” criteria and may need to be restored

The impact of the above is believed to be reduced with the data extracted from PDS (Personal Demographic Service).

 


Comparability and Coherence

This publication is a snapshot of the registered patient counts as at the first day of the month. Counts are prone to daily changes as this is a live system and counts can also differ greatly between months. In addition, NHS England’s health geography structure changes to some extent on 1 April each year, consequently comparison with previous months should be viewed with caution. For more information on these changes see the Organisation Data Service (ODS) page.

 

Mapping organisations

Output areas (OAs) are the lowest geographical level at which census estimates are released. Output areas for England and Wales were created from 2001 Census data. Changes were made in these considering the up-to-date population estimates in the 2011 Census and the 2021 Census.

From July 2024, the 2021 LSOA mapping will be used for the quarterly data files.

The publication contains a mapping CSV to allow for mapping to other data sets of GP practices (on PRACTICE_CODE) to ICB sub-location, ICB, NHS Region and GP system supplier. This will be updated each month to give the most relevant organisational mapping.


Primary Care Networks

Primary Care Networks (PCNs) were introduced into the National Health Service in England as part of the NHS Long Term Plan, published in January 2019, and form the building blocks of Integrated Care Systems. A PCN consists of a group of general practices working together with a range of local providers (pharmacy, mental health, social care, community and voluntary sector), serving a population of at least 30,000 and not tending to exceed 50,000. They build on the core of current primary care services and enable greater provision of proactive, personalised, coordinated and more integrated health and social care.

The information about PCNs will be published by ODS online, including name, address, codes and GP practice partner information. From the point of publication, all PCN data (including the practice partner relationships) will be maintained by the relevant commissioning organisation informing ODS of changes to PCNs in their area. It is believed this information could change frequently, with ODS releasing the data monthly.

This publication will apply the most recent version of the practice partner relationship to provide an estimate for the number of patients registered within a PCN.


Timeliness and Punctuality

The publication shows the latest set of counts of patients registered at a GP practice. Data was extracted in the month of publication and is an accurate snapshot as at the first day of that month. LSOA data is available quarterly and will be reported on that basis, that is each quarter.


Accessibility and Clarity

This publication is published on the NHS Digital website every month and all the data tables are available free of charge.


Assessment of User Needs and Perceptions

A consultation was carried out in April 2016 to find out from users how beneficial this publication is. It was found that the provision of GP practice populations by single year of age; 5-year age group, gender and LSOA were essential to the performance of some of their analysis requirements as these data are not available elsewhere. These analyses help inform commissioning and service requirement amongst other things.


Performance, Cost and Respondent Burden

Data are extracted as a monthly snapshot in time from the Personal Demographic Service (PDS) maintained by NHS England.


Confidentiality, Transparency and Security

Unlike previous versions of this publication (October 2015 and earlier), there is no suppression of data for GP Practices with 100 or fewer registered patients.

The original reason for the suppression was possible identification of individuals when data are linked to other datasets. It was thought that the risk of linkage to other datasets requires the other dataset to contain potentially disclosive data, as this report only contains counts of patients at a practice.

Many of the smaller practices are those that are opening or closing, and registrations are therefore at a low level due to the removal or addition of patients. In these cases, identification by linkage would be almost impossible due to the fluctuating nature of the number of GP practices.


Further information

Last edited: 9 April 2024 9:59 am