General practice activity: quality report

Data description

The general practice activity statistics provides information on appointments and selected additional activities in general practices in Wales.   

Statistics for appointments show the number of appointments attended and appointments not attended. Appointments attended are categorised by mode, type of practitioner and reason for consultation.

Statistics for selected additional activities include referrals made, fit notes issued, the total number of calls to practices, calls answered within two minutes, calls not answered, administrative communications issued (letters or emails), text messages sent and received, and total digital requests submitted to practices.        

Statistical concepts and definitions

The main statistics in this release are generated from administrative information contained in each general practice’s appointment book; the primary use of the appointment book is for general practice staff to manage their time effectively. The collection of data for use in official statistics is a secondary use and therefore the data is limited by the appointment book structure and data recording processes. 

As the appointment book has not been developed specifically for statistical purposes and as staff at general practices provide bespoke services for patients’ differing needs, not all activities can be simply coded into distinct categories.  Therefore, general practice staff have been asked to use their best judgement when recording complex information, which may not fit into a single category. 

Some staff may interpret similar activities differently and therefore similar activities may be recorded differently. As a result of the limitations of the data collection process, all appointment statistics presented are estimates. 

The data collected are of sufficient quality to produce 'official statistics in development' and are the best available data on general practice activity. It is likely that these statistics will remain ‘official statistics in development’ while data is collected through existing processes. 

Appointments

The term ‘appointment’ refers to all patient-related activities recorded in a general practice’s appointment book. The majority of activities recorded will be traditional appointments where a patient has a consultation with a GP, nurse or other health professional in the practice, either in-person or over the phone. The term will also capture certain activities which staff perform for patients, but without having a consultation, such as a medicine review, if it is recorded in the appointment book. Any ad-hoc GP clinical activity (for example, a nurse requesting a GP’s assistance during a consultation) will not be counted unless it was recorded in the appointment book. In practice, this is likely to result in an undercount of the estimated activity levels published in this release.  

The overall number of appointments should be recorded in a broadly consistent way across all practices. 

When the overall number of appointments is broken down by different categories, there are inconsistencies in the data collected due to the different appointment book software currently used in practices and in the different ways practices are managed. 

For example, reception staff in some practices may determine whether a face-to-face or telephone consultation is required; while other practices may use a ‘phone first’ scheme where a patient’s initial contact with a GP, for an ‘on the day’ appointment will always be through the telephone, and the GP will determine if a face-to-face appointment is also required during the telephone consultation. This situation may result in two appointments for the same patient being recorded, depending on how the practice records its appointment information.

Each individual consultation carried out through an ‘open surgery’ slot should be recorded as a single appointment. 

It would be unlikely for appointments with non-general practice staff such as health board employed health visitors and midwives, which took place at the general practice location to be recorded on the general practice’s appointment book and would therefore unlikely to be counted in these statistics. These types of appointments would usually be scheduled through different appointment systems. 

If a patient had a consultation with a GP and during the consultation is instructed to see a nurse (for a blood test, for example) this would only be recorded as two appointments if the nurse appointment was also entered into the appointment book. 

Patient calls to make appointments and calls for test results should not be included in the appointment numbers. 

As part of the activity data quality improvement project of the Quality Assurance and Improvement Framework (QAIF) for 2021-22, practices were requested to categorise appointments by mode, type of practitioner and reason for consultation using the PCIP. It became a requirement for practices to undertake this exercise for data from July 2022 onwards.

Face-to-face appointments are those where a patient physically attends the practice. Remote appointments are those conducted over the telephone or through video call. Home visit appointments will include appointments where practice staff visit the patient’s residence which includes communal living establishments such as nursing homes. 

GP and independent prescriber appointments should capture appointments with all GP types including GP registrars, as well as appointments with staff such as advanced nurse practitioners, pharmacists and physiotherapists who are qualified to prescribe as defined by NHS Wales Shared Services Partnership. 

Nurse, allied health professional, or other clinical role appointments should capture appointments with clinical staff who are registered with a professional body (for example, nurses registered with the Nursing and Midwifery Council) but who are not qualified to prescribe. It should not count interactions with administrative or non-clinical support staff.

Appointments that are classified by their reason as chronic, planned, or non-acute conditions are typically those for pre-existing conditions which have been ongoing for a longer period of time and are undergoing chronic disease management; while urgent or acute conditions are typically those which have occurred over the short-term. A patient’s reason for an appointment may be complex and staff recording this information should use their best judgement when categorising the reason for appointment. Typically this would be done by an administrative member of staff for clinical reasons and be subject to change, following advice from the health professional providing the appointment or practice manager. 

Telephone calls

General practices in Wales do not operate on the same telephony system, and telephone call information is submitted by practice managers through a pro-forma, rather than data being automatically extracted from telephone systems. Information on calls should be reported on a broadly consistent basis collected across each practice. 

Calls to the practice should count all calls received to the main practice phone number, which typically go through a switchboard process. While the large majority of these calls would be from patients, any other calls to the main practice phone number from others such as health boards, district nurse teams, and suppliers will also be counted. Any calls directly to staff which bypass the main switchboard, for example urgent calls from district nurses to GPs directly, or calls made to a mobile phone would not be counted in these statistics. 

For the measure ‘calls answered within two minutes’, the clock starts at the end of the automated message practices have when their number is dialled, and ends when a member of the practice answers the call. 

Calls not answered refers to all calls where a member of the practice did not answer the call before the call ended. This will include when calls were made outside of opening hours and when the information in the initial automated message was sufficient to meet the caller’s need without having to speak to a member of the practice.  

Administrative communications issued

This counts the number of letters and emails sent by the practice to patients. These might include responses to digital requests and information dissemination. 

Text messages sent and received

This counts the number of SMS text messages sent and received by the practice. These would largely be to and from patients but may include texts to and from other parties. 

Examples of practices sending text messages included automated appointment reminders from practices, options for cancelling no longer needed appointments, and information about conditions.   

Fit notes issued

A fit note is a statement of fitness for work (Department for Work and Pensions), (also known as a ‘med 3’), that provides medical evidence to enable an individual to access health-related benefits or evidence eligibility for Statutory Sick Pay. Fit notes are typically required is the individual is unable to work for more than 7 days due to medical reasons. 

The statistics count fit notes issued by GPs and other qualified staff at general practices and should include fit note extensions. 

Referrals made

In most cases, a referral is required to access specialist health services through the NHS. Referrals are most commonly provided by primary care health practitioners following a patient consultation. 

The number of referrals includes referrals made by any member of practice staff, to another healthcare provider (most commonly to secondary care hospitals, but also to other primary care or community health providers) for specialist treatments or testing.

Note that referrals data presented here is a measure of activity performed in general practices, therefore it is different to the number of GP referrals for first outpatient appointments published on StatsWales which is a measure of new patient pathways opened in NHS secondary care. 

Total digital requests submitted to practices

This counts instances where a patient submits a request to their practice via digital platforms (such as eConsult, Mysurgery app, MyHealth), or through live chats. 

Examples of digital requests include requesting appointments, repeat prescriptions and test results. 

Prescription items issued

The number of items prescribed in general practices is also collected on the Primary Care Information Portal but is not used in this statistical release as prescriptions data is already published as part of an established ‘primary care prescriptions’ statistical release.

Primary Care Information Portal (PCIP)

The Primary Care Information Portal (PCIP) is a tool that holds general practice activity data and presents data tables and visualisations which are available to the NHS in Wales on a restricted basis. 

Appointments data is extracted from each general practice in Wales using the Audit + tool and the portal is overseen by the General Medical Services (GMS) Data Quality System governance group, chaired by the General Practitioners Committee (GPC) Wales.

The main purpose of the PCIP is to improve the information available to general practices so that it can be used to inform service decisions. 

The PCIP is a relatively new tool and is under regular development.

General Medical Services (GMS)

The contract by which all general practices in Wales provide services on behalf of the local health board.

Statistical unit

The statistical unit is the type of activity undertaken by the general practice. 

The primary unit is the number of appointments, with supplementary statistics published for referrals made, fit notes issued, telephone calls, administrative communications issued (letters or emails), text messages sent and received and, total digital requests submitted to practices.   

Statistical population

The statistics refer to all scheduled appointments and selected additional activities in all GMS general practices in contract with health boards in Wales, during the reference period. 

No data from non-GMS or non-NHS practices are included. 

Reference area

Statistics are published at Wales, local health board and cluster level, based on the geographic location of the practice.

Time coverage

Appointments data was first collected and published for October 2021. 

Supplementary activity data was first collected and published for July 2022. 

New monthly data is published on a quarterly basis.

Source data

All data is sourced from various general practice systems and then uploaded onto the PCIP. 

General practice appointment book systems

There are currently two providers of appointment book systems in Wales, EMIS and VISION. Practices can currently choose to work from either system; however, all practices will be transitioning to EMIS between 2024 and 2027. 

Appointment book information from both systems is extracted daily by Audit +, a GP data quality, audit and reporting software tool hosted by DHCW and available to all general practices. Data files are generated from the Audit+ data and data for the previous month are automatically extracted by DHCW every month. The data is extracted by DHCW on the 10^th of each month and uploaded on the restricted PCIP by DHCW shortly after. Data is published on the PCIP with a two-month lag.

EMIS and VISION systems are configured differently which limits the categorisation that a practice undertake on appointments. 

In practices which use EMIS, the level at which activity is extracted by Audit + is the ‘session type’ description. This means that some of the appointment types may not fit exactly into the defined categories, as there may be multiple users or types of appointment contained within each ‘session’. The practice staff recording the information uses their best judgement as to the best fit in this scenario. For example: a ‘morning surgery’ session for a GP may be created containing 10 telephone slots, 4 face-to-face contact slots and 5 e-advice slots. This mixed session can be easily categorised by professional heading, but not easily categorised by mode of consultation. As a result, the practice would make a ‘best fit’ decision whether this session is best initially categorised as ‘Remote’ or ‘Face-to-Face’ based on the session mix at the initial categorisation stage. 

In practices which use VISION, the level at which activity is extracted by Audit + is the appointment ‘slot type’. This means that some of the appointment types may not fit into defined categories. For example, it may not be possible to categorise 'telephone consultation' to a professional group, in which case practices can choose 'unable to map to professional heading'. Practice staff may also be unable to categorise these slot types to the reason of consultation and are expected to make a pragmatic assessment of which is most appropriate by volume. 

For both systems there may be entries into the appointment book where categorisation to the various activity descriptions is not possible or appropriate, such as lunch breaks. These appointments should not be counted in the appointment statistics. 

General practice telephony systems

There are various telephony systems used in general practices. Each month practice managers view their call logs and manually submit data for the relevant telephony metrics to the PCIP using a pro forma. 

As described in the ‘telephone calls’ section, only calls received to the main practice phone number should be counted.

Selected additional activities

Digital requests data are collected from various sources including email, eConsult and other digital platforms.

Similarly to telephone call data collection processes, practice managers undertake searches on their systems every month for activity in the previous month and manually upload data to the PCIP using a pro forma. 

Data validation

Broadly, the appointments and selected additional activities data extracted from general practices are published on the restricted PCIP with limited validation carried out by DHCW. 

DHCW monitor data from practices to ensure that data are included for all practices in the reference month. One further data validation step is performed by assessing if a patient has an arrival time but has an appointment status of ‘did not attend’. In these cases, the ‘did not attend’ status is removed.

The PCIP also has an inbuilt validation step which ensures that practices can only submit data for additional selected activities with a numeric data type. 

Validations by Welsh Government are limited by the granularity of data supplied. However, the following data checks are performed:

• Ensuring that geographic aggregations sum to published totals on the PCIP. For example, clusters should add up to the relevant health board number and health boards should add up to the Wales number
• Ensuring that appointment categories sum to published totals
• Checking that numbers are proportionate with what might be expected based on previous time periods
• Ensuring that the total number of calls is not less than either calls answered within 2 minutes or calls not answered

Where any suspected data quality issues are identified, information is relayed to DHCW for further investigation. 

Adjustment 

No routine adjustments are made to the data.

Quality assurance

Quality assurance is undertaken in line with the quality strategy and the Quality pillar of the Code of Practice for Statistics (UK Statistical Authority).

Quality assessment

These statistics are currently classified as ‘official statistics in development’ as the data, definitions and processes are being developed and understood by all organisations involved. 

User needs

The aim of these statistics is to inform users about activity in general practice. The key users of the statistics are: 

• ministers and the Members Research Service in the Welsh Parliament
• local health boards
• local authorities
• GP collaboratives and primary care clusters
• general practices
• the department for Health and Social Services in the Welsh Government
• other areas of the Welsh Government
• National Health Service
• Public Health Wales
• the research community including students, academics and universities
• GPC Wales and other representative bodies
• individual citizens, third sector organisations and private companies
• other UK government departments

The statistics are used in various ways. Examples include:

• advice to ministers
• to inform debate in the Welsh Parliament and beyond
• to provide publicly available data on general practice activity in Wales

Completeness

A summary table showing the level of coverage from practices in the appointments data will be added after this information becomes available.

Overall accuracy

These statistics are based on administrative data and are therefore not subject to any survey sampling error. 

As detailed in the ‘appointments section’ only appointments recorded in the appointment book system will be counted and any ad-hoc clinical activity will be missing from these statistics. 

Practice staff also need to use their best judgement when categorising appointments that do not fit into high-level categories. This process may result in similar activities being recorded differently in different practices.

Practices can retrospectively re-categorise an appointment that has been published on the PCIP. While this would be unlikely for the very large majority of cases, published data can be subject to change.

There are manual processing steps involved in submitting data for additional activities and therefore these data are subject to human processing errors.

As a result of these data accuracy concerns, the statistics are presented as estimates. 

Data revision

The data submitted to the PCIP is not routinely revised, but practices are able to retrospectively re-categorise appointment data. There is no restriction on how far back practices can revise data and any revised data would be included onto the PCIP the following day. 

It is planned that Welsh Government statisticians will process the previous 12 months data with every quarterly statistical publication. This means that data published for the previous 12 months may be revised. Any large revisions will be highlighted in the statistical release and data tables.

Published data, from time periods longer than 12 months before the reference period will not be revised unless an error is discovered. If this does occur the revision will also be highlighted in the statistical release and data tables.

Timeliness

Monthly statistics are planned to be published every quarter on a March, June, September, December cycle. The latest release will contain new statistics for the reference period ending three months before the publication date (for example, the June publication contains new statistics for January, February and March). 

Punctuality

All outputs adhere to the Code of Practice by pre-announcing the date of publication through the upcoming calendar. Furthermore, publication dates are announced well in advance and any delays are communicated via notices on our website. Any revisions or postponements to outputs follow the Revisions, Errors and Postponements policies published online.

Geographical comparability 

Data between different geographies in Wales is broadly comparable, subject to the data quality issues specified in this document. 

No comparability work has been done with other countries to date. Given the data quality issues within Wales and the different appointment systems used in other countries it is unlikely that GP activity data is directly comparable between countries. 

England: NHS England publish statistics on appointments in general practice every month here: Appointments in General Practice (NHS England Digital).

Scotland: Public Health Scotland publish statistics on general practice activity every quarter: Primary Care in-hours General Practice activity visualisation - as at 31 March 2024 - Primary Care in-hours General Practice activity visualisation - Publications (Public Health Scotland).

Northern Ireland: No statistics on general practice activity are currently published for practices in Northern Ireland. Related statistics are published here: General Medical Services Statistics - Business Services Organisation (BSO) Website (hscni.net).

Comparability over time

As there are seasonal variations, fair comparisons of activity levels should only be made on the same reference period in each year; for example comparing December 2022 with December 2023. 

Monthly data is most fairly compared by working day as the number of working days in each month can change from year-to-year.

Numbers for total, attended and not attended appointments were first collected and published from October 2021.

Numbers for specific categories of appointments and selected additional activities were collected on a consistent basis and published from July 2022 onwards.

Publication

The statistics are pre-announced and then published at 9:30am on the date of publication on the Welsh Government website. 

For publications relating to reference months in the quarter ending June, September and December, data tables are published on the StatsWales website with an accompanying short statistical headline. 

For publications relating to reference months in the quarter ending March, an annual report is also published in HTML format, with analysis, charts and commentary focusing on the latest financial year. Data tables for all analysis covered in the HTML report are also published on StatsWales. 

Data published on StatsWales can be viewed and restructured online or downloaded into spreadsheets. Additionally, a connection can be made to the datasets from a software package (for example, Microsoft Excel) using the OData feed of the StatsWales cube.

Online databases

Full datasets are published on the ‘general practice activity’ section of StatsWales.

Documentation on methodology

Further information on the quality improvement project which led to the development of these statistics is available here: General medical services contract, GP activity data quality improvement project advice for GP practices

The PCIP has already been established as tool used within DHCW, general practices and the NHS and therefore there is limited additional burden on general practices in producing these statistics.

Confidentiality: policy

The Welsh Government’s statistics and research statement on confidentiality and data access describes our approach to data confidentiality and conforming with the data governance principle of the Trustworthiness pillar in the Code of Practice for Statistics (OSR).

Confidentiality: data treatment

Only aggregated data is contained in the extract supplied to Welsh Government from the PCIP, with no personal identifiable information. Small numbers are not deemed to be disclosive and therefore no disclosure control methods (for example, suppression) are applied to the published statistics.

Statistician: Craig Thomas
Email: stats.healthinfo@gov.wales

Media: 0300 025 8099