Title

GP workforce analysis by practice and cluster deprivation quintiles

Last update

23 July 2024 23 July 2024

Next update

To be confirmed

Publishing organisation

Welsh Government

Source 1

Contractor Services, NHS Wales Shared Services Partnership

Source 2

Welsh Index of Multiple Deprivation, Welsh Government

Source 3

Wales National Workforce Reporting System (WNWRS)

Contact email

stats.healthinfo@gov.wales

Geographical coverage

Local health boards

Data licensing

You may use and re-use this data free of charge in any format or medium, under the terms of the Open Government License - see http://www.nationalarchives.gov.uk/doc/open-government-licence

Weblinks

https://gov.wales/welsh-index-multiple-deprivation
https://gov.wales/general-practice-workforce
https://gov.wales/general-practice-and-primary-care-cluster-population-and-workforce-deprivation-31-december-2021

Keywords

GP practice, cluster, deprivation quintile, primary care, WIMD, GP workforce

General description

This shows data on the workforce in primary care by deprivation quintile. Four different methods for calculating the deprivation quintile are included.

Data collection and calculation

Practice population data is provided by NHS Wales Shared Services Partnership and is a snapshot taken at a single point in time.
The resident lower layer super output areas (LSOAs) of patients registered to each general practice were matched to the Welsh Index of Multiple Deprivation (WIMD) 2019 to count the number of patients who live in the most deprived 20% of LSOAs as determined by their WIMD ranking. General practice data was aggregated to primary care clusters and rankings were created based on two measures to estimate deprivation at primary care cluster level. These were:
1) The number of patients registered to the general practice within the cluster who live in the most deprived 20% of areas in Wales.
2) The percentage of each cluster population that live in the most deprived 20% of areas in Wales.

In terms of workforce data, the data that most closely aligns with the reference period of the practice population data is used. The workforce data as at 31 December 2021 is linked to each practice in the January 2022 practice population extract. For the April practice population data, the workforce data as at 31 March of the same year is used.

The practice population data for 2023 onwards used in the deprivation analysis will not necessarily match the published practice population data for the same period. The published data uses the latest 2021 LSOA mappings, however as the deprivation analysis relies on WIMD 2019 which uses 2011 LSOA mappings, the deprivation analysis also needed to match the 2023 onwards practice populations to the 2011 LSOAs and this resulted in a small number of patients not being matched.

April 2024 data:
There were 372 general practices. As 372 is not divisible into a whole number by five, the quintiles do not contain the exact same number of practices. There are 76 practices for quintiles 1 and 5, 73 practices for quintiles 2 and 4 and 74 practices for the remaining quintile. Quintiles are used to analyse broadly similar groups of practices, with practice quintile 1 containing the 76 practices with the most deprived populations in Wales.

There were 63 clusters. As 63 is not divisible into a whole number by five, the quintiles do not contain the exact same number of clusters. There are 13 clusters for quintiles 1, 3 and 5 and 12 clusters for the remaining quintiles. Quintiles are used to analyse broadly similar groups of clusters, with cluster quintile 1 containing the 13 clusters with the most deprived populations in Wales.

April 2023 data:
There were 382 general practices. As 382 is not divisible into a whole number by five, the quintiles do not contain the exact same number of practices. There are 77 practices for quintiles 1 and 5 and 76 practices for the remaining quintiles. Quintiles are used to analyse broadly similar groups of practices, with practice quintile 1 containing the 77 practices with the most deprived populations in Wales.

There were 63 clusters. As 63 is not divisible into a whole number by five, the quintiles do not contain the exact same number of clusters. There are 13 clusters for quintiles 1, 3 and 5 and 12 clusters for the remaining quintiles. Quintiles are used to analyse broadly similar groups of clusters, with cluster quintile 1 containing the 13 clusters with the most deprived populations in Wales.

January 2022 data:
There were 390 general practices. These practices were ranked in terms of the relative deprivation of their registered patients and split evenly into quintiles containing 78 practices each. Quintiles are used to analyse broadly similar groups of practices, with practice quintile 1 containing the 78 practices with the largest populations living in the most deprived areas in Wales.

There were 63 clusters. As 63 is not divisible into a whole number by five, the quintiles do not contain the exact same number of clusters. There are 13 clusters for quintiles 1, 3 and 5 and 12 clusters for the remaining quintiles. Quintiles are used to analyse broadly similar groups of clusters, with cluster quintile 1 containing the 13 clusters with the most deprived populations in Wales.

Frequency of publication

Annual

Data reference periods

WIMD 2019 is used for determining the most deprived 20% of areas in Wales.

Data for practice populations refer to January 2022 and April for subsequent years.

Practice workforce data: The 31 December 2021 workforce data is linked to the January 2022 practice population data. For the April practice population data, the workforce data relates to 31 March of the same year.

Users, uses and context

The purpose of these data are to show how general practice workforce (full time equivalents) differs by the relative deprivation of the population in Wales.
Determining which measure is most appropriate to use depends on the purpose of what the measure will be used for. It is recommended to use ‘number’ measures when targeting the greatest number of people living in deprived areas, whereas it is more appropriate to use the ‘percentage’ measures when targeting the highest concentration of people living in deprived areas.
While the results at practice and cluster level are broadly similar, the practice level data will provide a more granular analysis of deprivation. Clusters group practices, so while some may broadly serve similar populations, inevitably there will be some practices which have a more deprived population than others, in the same cluster. Clusters analysis may be most appropriate when targeting broader areas, but some more detailed insight may be masked when aggregating data for individual practices.

Statistical quality

The data used in these tables is of good coverage and quality. All GP practices with General Medical Services (GMS) contracts at the time of data extraction are included. Practice population data is available for all practices.
A small percentage of data was not included in the analysis.
As WIMD is a deprivation measure relative to Wales only, English residents could not be included in the model. As we have data collected for Welsh general practices collected on a consistent basis, English practices are not included, and therefore any Welsh residents registered to English practices will not be included. For context, in the April 2024 extract there were 21,067 patients registered to Welsh general practices who resided in England, and there were 13,339 patients registered to English general practices who resided in Wales.
Where a patient resides in England but is registered to a Welsh general practice, they are also removed from that practice’s population list size in this analysis, to not skew the percentage of patients living in deprived areas calculation.
Patients registered to non-GMS practices are also not included in the analysis.
Patients registered to Welsh practices who had missing LSOA data are also excluded from the analysis; there were 108 patients in the January 2022 data, 43 patients in the April 2023 data and 49 patients in the April 2024 data.

Name

HLTH0464