General Practice Workforce 31 December 2020

General Practice Workforce statistics in England are compiled from data supplied by GP practices.

• We collect individual-level information on all staff (GPs, Nurses, Direct Patient Care and Administrative/Non-Clinical staff) employed at the practice at the end of the applicable reporting period, which provides a snapshot of the workforce at that date. This snapshot includes any GP locums and other temporary staff working in the practice for regular sessions or on long-term placements.  However, any staff that have left the practice before the extract date are not included in the snapshot counts although they may be counted against a new practice if applicable and are also included in the biannual Joiners and Leavers analysis.
• The GP practices also provide high-level working hours information about any GP locums working for them on a more ad-hoc or short-notice basis. We refer to these as “infrequent locums” and they may cover as little as a single session in the entire reporting period. Some of these infrequent locums may cover for unexpected absence at the same practice multiple times in the same or several reporting periods, while others work for a practice only once.
  We include figures for this cohort of GPs to reflect the scale of their contribution to the overall general practitioner workforce and although we include them in the overall totals, the infrequent locum counts are different in nature and calculated differently to the snapshot of the main workforce.
  • It is important to note that many infrequent locums hold other roles within the GP workforce, for example acting as salaried GPs or long-term locums as well. In such cases, they are counted in the snapshot counts against their other roles as well as in the infrequent locum FTE and headcount figures. Where we are able to identify the same individual in multiple roles, we ensure that we do not report them twice in the overall headcount totals.

We use the data provided by the practices to calculate summary full-time equivalent (FTE) and headcount figures and publish these as Official Statistics.

For several years, we have been publishing these figures every quarter for data extracts effective on 31 March, 30 June, 30 September, and 31 December. In 2020, we were asked to collect data and publish statistics each month rather than on a quarterly basis, and as requested we published the first of these new monthly releases for October 2020’s data extract.  However, it is evident that the transition to the monthly collection during the COVID-19 pandemic had a detrimental effect upon the data quality and coverage and although we will retain the October and November 2020 publications for transparency purposes, we will not include data from these months in any further time series or analysis. We are temporarily reverting to a quarterly collection and publication for December 2020 onwards although we expect the monthly collections to resume during 2021. In addition, some measures have routinely been published only every six months – for data collected at the end of March and September – and we will continue to release these biannually.

We collect information about job role, contracted and working hours, gender, age, and ethnicity, but no information relating to earnings or expenses. 

We work closely with GP practices to improve the quality, completeness and coverage of the data submitted, but responsibility for data accuracy lies with the providing organisations.

We have been collecting this record-level data from GP practices since September 2015 and have made a range of improvements to our processing and estimations methodologies since the collection began. In some cases, we recalculated historical figures back to September 2015 using these processing improvements which means that some earlier publications have been superseded with more up-to-date figures. We recommend that you always use the most recent publication for up-to-date information and time series figures, and to avoid confusion we have archived the superseded publications although they are still available.

Prior to September 2015, data about staff working in GP practices was taken annually from the National Health Authority Information System (NHAIS). Because of this change in data source, figures in this publication are not comparable with those published for September 2014 and earlier and we recommend that you do not make any comparisons between the current time series and the figures prior to September 2015.

This data quality statement relates to the entire current General Practice Workforce series. Any data quality issues specific to a particular release are detailed separately.

There are two main sources of workforce data for this report:

• National Workforce Reporting System (NWRS)
  This is an online system in which GP practices record and update staff details. You can find more information about the NWRS, including guidance documentation. 
  Practices are asked to ensure that the information about their staff is up-to-date at the end of each quarter and NWRS automatically extracts the data without requiring practices to confirm the submissions which reduces the burden on practice staff.
  The NWRS has inbuilt validations to reduce data input errors, such as limiting the job roles to those permitted within the National Workforce Data set (NWD), and ensuring only numbers are entered for numeric fields.
  We automatically extract the workforce data at the end of the last day of each quarter.
• We collect information on GP registrars on General Practice placement from Health Education England’s (HEE) Trainee Information System (TIS) which we have used since June 2018 as it is more timely and reliable than our previous sources of GP Registrar data.
  • Introducing this new data source meant that headcount and full-time equivalent (FTE) counts for GP registrars were not comparable with figures published for previous quarters, so we calculated an estimated difference in GP registrar counts for June 2018 and used this measure to revise the earlier GP registrar figures back to September 2015. 
  • These revised figures allow you to make national level indicative comparisons across reporting periods, although we recommend that you use caution when considering GP registrar figures prior to June 2018.
    • However, due to issues with the data quality, a very high number of registrars in the data for March 2018 and earlier were recorded against an “Unknown” practice. This means that although comparisons back to September 2015 can be made at an England level, this is not possible at a regional level and it is therefore inappropriate to make local-level comparisons of GP registrar counts, and any GP totals that include registrars, any earlier than June 2018.

If practices submit invalid, partial or no data for some or all staff groups, we calculate estimates.  These estimates use registered patient counts, which we collect from the GP Payments system (Open Exeter); the methodology is described in the Accuracy section of this data quality statement. We publish the percentage of practices with fully or partially estimated records in each staff group in Annex A in the Excel bulletin as well as in Table 1 in the Report section of the publication webpage.

All GP practices are contractually required to provide data on their workforce. However, a small number of GP practices do not do so or submit incomplete data. In addition, some records for some practices fail data validation and have to be excluded:

• We remove records where no information is given about the job role or staff group if this cannot be derived from the job role.
• Some otherwise good quality records are missing contracted or working hours, and in these cases, we estimate that staff member’s hours based upon the national average hours for that job role.

There is also a recognised issue where some practices make retrospective changes to their workforce data after extraction. These changes are incorporated into the publication figures after the following data extract but the earlier figures in the time series are not retrospectively amended to reflect these changes as this would result in all figures being continually subject to revision. However, the scale of records affected by such retrospective changes has been found to be generally stable across quarters.

Estimates

There are some practices where all the provided data for a staff group is of poor quality and has to be removed. We calculate estimates for headcount and full-time equivalence for those practices which did not provide valid and/or complete data for one or more staff groups; this could be due to poor data quality or no submitted data.

To produce estimates for incomplete or missing data, we use the valid data submitted by the other practices during the reporting period, along with information about their registered patient population. Our estimation methodology takes the practice patient populations into account to address potential issues that could arise if the practices providing no data or poor-quality data were not of a typical or average size. We know the registered patient counts for 98-99% of practices and use the national average registered patient count for practices where this information is not available.

We produce our estimates as follows:

1. For every job role, we calculate a national ratio of FTE per registered patient.
   This uses the total FTE and registered patient count for all the practices that supplied valid data.
2. We calculate FTE estimates at CCG level by taking the national job role ratio calculated in step (i) and multiplying it by the total registered patient count for all practices in that CCG that did not supply valid data for the applicable staff group.
3. We use the same principles to calculate headcount estimates.
4. We aggregate the estimates from CCG level to provide higher level figures at national and sub-national levels.

We produce estimates independently for each of the four staff groups (GPs, Nurses, Direct Patient Care and Admin/Non-clinical staff). This means that if a practice submits no data, or invalid data for a single staff group, their submission for the other three staff groups is still treated as valid.

Until September 2020, we collected workforce data on a quarterly basis. We are now extracting the data every month. We are aware that seasonal variation affects General Practice workforce figures, but as our estimation process calculates ratios using valid data for the applicable reporting period, any seasonal variation should be mitigated.

It is important to note that we do not produce wholly estimated records at individual practice level. At GP practice level, we only estimate missing FTE where the staff member’s record is otherwise valid and complete and is simply missing contracted or working hours.

We do not allocate estimated records any age, gender, or country of qualification information; for these data items, any estimated records are reported against ‘Unknown’.

We do not produce estimates for GP Registrars as the timely and complete TIS data means this is not required.

Locums

We report information about GP locums as long-term and infrequent locums. However, the concepts of “long-term locum” and “infrequent locum” are artificial constructs created solely for data collection and reporting purposes.

Long-term locums

We define long-term locums as GP locums who work regularly at a practice to cover long periods of time (although we do not specify what constitutes a “long” absence), such as maternity leave, long-term sickness or a vacancy, or locums who cover one or more sessions per week or month on a planned and regular basis. The long-term locums have consistent and predictable working hours, and their presence in the workforce is expected on a planned and/or ongoing basis. Details about these long-term locums are captured in the main part of the NWRS in the same way as permanent practice staff.

We have been collecting information on long-term locums since the launch of the data collection. In the spring of 2017, we revised our guidance documentation to clarify what was needed. This resulted in a notable increase in counts of these GP locums between December 2016 and March 2017 which suggested that these figures had been under-reported in the earlier collections.

Infrequent locums

We define infrequent locums as GP locums who do not regularly work at a practice, may cover very few sessions, and are typically employed on an ad-hoc basis. These infrequent locums have unpredictable working patterns and typically cover for unexpected or unplanned need, often at short notice. Some infrequent locums work as little as a single session at a particular practice, although some provide ad-hoc cover to the same practice in several reporting quarters.

Because these infrequent locums work at the practice for such a short time, it is likely that full details about the individual may not be available to NWRS users and it became evident that practices were not able to provide complete – and in some cases – any infrequent locum information.

To be able to collect information about this important subset of the GP workforce, we introduced a new module to the NWRS tool and now ask practices to provide only GP name, GMC number and the total number of hours worked during the reporting period. We use the working hours information to calculate an average full-time equivalence and are also able to report on the number of distinct individuals working in this role. This means that although these infrequent locums may not have been working for a practice on the date of the data extract, we can nonetheless present valuable figures about GP locum usage and demand. However, the infrequent locum FTE and headcount measures are fundamentally different to the snapshot figures calculated for the rest of the general practice workforce.

We have been collecting information on infrequent locums since September 2017 and have been working closely with GP practices to improve the data quality.

Locum time series figures

Our stakeholders are keen to be able to review a time series back to September 2015 which was complicated by the changes to the long-term locum collection and the introduction of the infrequent locum module. Therefore, following a consultation with our stakeholders, in November 2019 we calculated estimates for FTE GP locums to revise the historical time series for September 2015, March 2016, September 2016 and December 2016 and present a more complete picture of these locum totals.

To do this, we calculated an estimated change in FTE long-term locum GP counts between March 2016 and March 2017 and used this figure to produce an uplift that we applied to the reporting periods prior to the release of the improved guidance.  We produced these GP locum estimates for CCGs using registered patient to locum FTE ratios and incorporated the estimates into the national time series in the General Practice Workforce 30 September 2019 publication and into the regional breakdowns from the General Practice Workforce 31 December 2019 publication.

As with other estimates, we did not allocate these new records any gender, age or other characteristic when producing these historical FTE GP locum estimates.

It is also important to note that it is not possible to produce estimates for GP locums by headcount as the working hours and patterns are so unpredictable. However, this means that any GP headcount totals that include locum counts from March 2017 onwards cannot be compared to counts for earlier reporting periods.

Details of this locum estimation methodology can be found in the publication released November 2019.

Full-time equivalent (FTE) and Headcount Figures

Many primary care staff work in more than one practice, CCG, STP, or region. When we refer to “headcount”, we mean the number of distinct individuals working at a practice, CCG, or other area/regional level. Headcount figures tend to be higher than FTE figures because we may count the same person several times depending on where they work or because the working hours of part-time staff members are added together when reporting full-time equivalent figures.

We calculate headcount separately for every reporting level, for example, GP practice, CCG, STP, and region, and higher-level headcount figures cannot necessarily be calculated by simply adding up the GP practice counts. This is because if the quality of the data is good, we can identify the same person in different organisations so at the higher reporting levels, we count them only once. However, if a record is missing the required identifiable information then we may be unable to correctly determine where else that person has been working.

For example, a staff member works full-time across two Practices within the same region, spending one day (20% of their time or 0.2 FTE) at Practice A and four days (80% of their time or 0.8 FTE) at Practice B. Because the data quality is good, we can identify that it’s the same person in both practices even though they hold two distinct roles or contracts as illustrated in Table 1.

Table 1: Headcount methodology – same region

If, however, the two practices were in different regions, with good data quality, we can still identify that it’s the same individual with two roles. In this case, they would be included in headcount figures for both GP practices and for both regions, but only once at a national level as shown in Table 2.

Table 2: Headcount methodology – different regions

The first two examples apply to a member of staff working in the same type of job role for different practices. However, an individual could also work in different job roles and in these cases, we count them once in each staff group as well as the overall totals.

For example, a GP works three days as a salaried GP in Practice A and two days as a locum GP in Practice B. At a national level, this GP shows in headcount figures for both Salaried and Locum GPs, but only once in the total GP headcount as shown in Table 3.

Table 3: Headcount methodology – different job roles

Note: FTE is based on the proportion of time staff work in a role, with 1.0 FTE equalling 37.5 hours per week.

Contract/role count is the total count of specific posts held/worked in a given organisation and some GPs or other staff members may have multiple roles either within or across organisations.

Impact of COVID-19

There is evidence to suggest that the completeness and coverage of data extracted from the NWRS for March and June was adversely affected by a range of issues, including the COVID-19 pandemic, and that the exceptional pressures on the general practice workforce meant not all practices were able to update their NWRS data in time for the extracts on 31 March and 30 June.

For September, we were confident the practices had returned to normal levels of NWRS data completion in time for the extract on 30 September.

Because seasonal variation can affect workforce counts, we strongly recommend that any historical comparisons be made only for the same point across years (such as September to September) rather than between quarters or months.

As we have noted, changes in our estimation and GP locum methodologies mean that we revised the entire time series for our publication of  September 2019’s data which was released 28 November 2019 and to avoid confusion, we have archived old publications whose numbers have been superseded.

Unless it is explicitly stated, notably by a vertical line in the tables, all published data is comparable with the same point in earlier years back to September 2015. There are two breaks in the headcount time series, for locum GPs that affect the figures for locums, All GPs and All Fully Qualified GPs. There are also two breaks in the headcount and full-time equivalent (FTE) time series for Direct Patient Care staff role-level figures, due to the presence in September 2016 of multiple Direct Patient Care records with a ‘Not Stated’ staff role.

Unknown Data

GP Practices may not have submitted information about age, gender, and country of qualification for all members of their staff. In these cases, headcount and FTE data are reported as ‘Unknown’ for these categories and therefore, this data is not comparable over time.

GP Joiners and Leavers

We have improved our methodology for producing figures on Joiners and Leavers and we revised all figures for Joiners and Leavers to the Qualified Permanent GP Workforce (excludes Registrars & Locums) in the Supplementary Information Tables in the September 2019 publication.

Details of the methodology are in the data quality statement that accompanies that publication.

We define joiners as GPs whose identifying information (GMC registration number; National insurance number; forename, surname and date of birth; or first initial, surname and date of birth) was present in the data set at the end of the relevant time period but was not there at the beginning. Similarly, we identify leavers as GPs whose identifying information was present in the data set at the start of the specified time period but was not there at the end.

Estimated records are not included in the current methodology and therefore there is a risk that current joiner and leaver counts are under or overestimated.

Absences

The NWRS is used to collect information on staff absences and the types of absence recorded include long-term sickness and parental leave. We published these figures biannually in the Supplementary Information Tables, and the most up-to-date figures can be found in the September 2019 publication. This information is currently recorded by only around 20% of practices and we do not estimate for those who have not recorded any data. Due to the low completion rate we advise you to use great caution when considering the historical figures and we are suspending the publication of these tables whilst the quality and suitability of the data are reviewed. Please refer to the Notice of Changes page in the June 2020 publication for more detail about our ongoing changes to the publication.

Vacancies

The NWRS is also used to collect information on vacancies, to capture the unfilled job roles for which practices are recruiting. We published these figures biannually in the previous Supplementary Information Tables, and the most up-to-date figures can be found in the September 2019 publication. This information is currently provided by less than 20% of practices and we do not estimate for the missing data. Due to this low completion rate we recommend that you use great caution when considering the historical figures and we are suspending the publication of these tables whilst the quality and suitability of the data are reviewed. Please refer to the Notice of Changes page in the June 2020 publication for more detail about our ongoing changes to the publication.

The relevance of NHS workforce data is maintained by reference to working groups who oversee both data and reporting standards. Major changes to both are subject to approval by the Data Coordination Board (DCB) which replaced the Standardisation Committee for Care Information (SCCI) from 1 April 2017.

Significant changes to workforce publications (e.g. frequency or methodology) are subject to consultation, in line with the Code of Practice for Statistics.

We publish figures as quickly as possible after extract. Scheduled publications are announced.

We are currently publishing data on a quarterly basis but expect this to become monthly during 2021.  Additional regional breakdowns and additional experimental tables based on those that were published for March and September data will continue to be published biannually as supplements to the standard quarterly release. Please refer to the Notice of Changes page in the June 2020 publication for more detail about changes to the contents of these biannual tables. 

Further information is available at the following links:

1. Primary Care

2. Primary Care Network (PCN) Workforce

3. GP Earnings and Expenses Estimates

4. Healthcare Workforce

Other UK publications

Scotland:

Wales

Northern Ireland