All GP practices have a contractual obligation to provide data on their staff to cover a 12 month period.
This data is currently collected by a Microsoft PowerApp tool, managed by NHS National Services Scotland. Guidance on how practices should complete the survey is provided
The 2018 Scottish General Medical Services Contract is a joint agreement between the Scottish Government and the BMA setting out the need for robust and reliable data about general practice.
The data supplied through the survey has missing values, especially in the areas of Age, Sex and the number of sessions or hours, which we address using data imputation methods.
Missing data on contracted hours, contracted GP sessions, and staff age have been imputed using a donor-based imputation that relies only on this year's survey sample data for replacement values, and preserves existing relationships and distribution present in the sample data. These methods help to provide a more complete data set for analysis.
For Doctors, age is imputed using Health Board, Main Staff Role, Sex and Contracted Sessions. Contracted Sessions are imputed using Health Board, Main Staff Role, Sex and Age.
For other staff who work in a GP Practice, age is imputed using Health Board, Main Staff Role and Contracted hours. Contracted Hours are impute using Health Board, Main Staff Role and Age. Sex is not used in imputing data for these staff groups as the vast majority are female.
As the response rates to the survey is never 100%, data collected in the survey is weighted to obtain an estimate of the total WTE.
The response rate are than 100% so appropriate weighting are applied to provide representative estimates at NHS Board and national levels.
The table below describes the different weightings used for each measure in this report.
| Staff Group | Measure | Weighting Used |
| General Practitioners | WTE | GP headcounts* |
| Contracted sessions | GP headcounts* | |
| Vacant sessions | GP headcounts* | |
| Locum sessions | GP headcounts* | |
| Absent sessions | GP headcounts* | |
| Practices using locums | Practice response rates per board | |
| Practices with vacancies | Practice response rates per board | |
| All Other Staff | WTE | Practice list sizes per board |
| Contracted hours | Practice list sizes per board | |
| Vacant hours | Practice list sizes per board | |
| Absent hours | Practice list sizes per board | |
| Practices with vacancies | Practice response rates per board |
GP headcounts are sourced from the NPCCD. These are matched to the totals of headcounts reported from the survey in combinations of NHS Board, designation, age group and sex. Weights of less than one are rounded up to one.
GP work is measured in sessions. The BMA standard for a session is 4hours 10 minutes, but each GP practice can define their session length. A full time GP is defined as 8 sessions per week.
Full time for other occupations is defined as 37.5hours. This has not changed along side the reduction in the working week for Agenda for Change staff within the NHS.
The Vacancy Rate is calculated as:
Vacant GP sessions / (Sessions of GPs in post + Vacant GP sessions)
where
Vacant GP sessions in the year is the sum of the weekly sessions of all vacant GP posts at 31 March of each year
Sessions of GPs in post is the sum of the weekly sessions of all GPs in post on 31st March of each year
For nursing and other staff groups vacancy rates we have used WTE instead of sessions.
The definition of vacancies reported here differs from that in the NHS Scotland Workforce publication. GP Practices are asked to include vacancies that are not advertised, and to provide a reason for why this has not been advertised.
Absences are collected by the number of sessions or hours lost to absence in any of the following 4 categories:
Absence rates are calculated as the number of sessions or hours lost to absence over the year divided by the total number of contracted sessions or hours over the year.
Sessions Lost to Sickness / Total Contracted Sessions over the year
where the total sessions is the number of contracted sessions for staff employed at 31 March multiplied by the number of weeks in the year (52.179), an estimate of the total contracted sessions worked over the year.
This survey reports on the whole workforce within Primary Care, categorised into 4 occupational groups.
The above occupational groups are sub-divided into occupational roles. A full list of these can be found in the GP Workforce Survey.
Headcount, WTE and Mean WTE are reported. Data on how many sessions per week doctors work is also published.
Vacancies are as at the 31st of March each year.