Overview
Data for all staff are sourced from the Scottish Workforce Information Standard System (SWISS). Workforce information was first captured through SWISS in 2007 for the data as at 30th September 2007. SWISS holds live individual level information for all staff and includes data items on the employee, their current post(s), absences and earnings. It is fed directly from NHS Boards' local payroll and HR systems.
The original source of absence data is Scottish Standard Time System.
NES extracts employment data on the second Tuesday of the month for staff in post on the last day of the previous month. For example, an extract of all staff at 31 March 2021 was taken on 13 April 2021. The lag between census date and extraction date is to allow time for the source systems to be updated.
Additional paybill data are extracted annually from SWISS. These extracts contain details of staff earnings paid by NHS organisations, and are taken at the end of each financial year. The extract takes monthly and weekly records for each employee and contains employee details, employment details, and employment earning details. These data, unlike the staff in post extract, include details of bank contract staff.
What data are published?
Staff in post data are reported for 31 March, 30 June, 30 September, and 31 December census as part of the NHSScotland Workforce Official Statistics publication. The data show the number of staff employed by NHSScotland by location (NHS Board), occupation (job family, sub job family / medical specialty, AfC band / medical grade, contract type) and demographic (sex, age group, median age) indicators. These terms are defined in the glossary.
In addition to these data, SWISS data are used to report the following
Background
The following lists the conditions for records in the SWISS data to be included in the NHSScotland Workforce publication.
A person must be in an active post and not working on a Bank contract type:
Furthermore, not all staff recorded on local HR and payroll systems are directly employed by NHSScotland; we therefore run a series of exclusions in agreement with NHS Boards. These exclusions are:
Medical doctors in training data are sourced from NES' Turas People system which allows accurate reporting of trainee board of placement and specialty. Turas People is an application hosted on the Turas platform that supports implementation of the lead employer model which came into force on 1 August 2018. For reporting purposes we link data from SWISS and Turas People to maximise the value of data held in each system. This linking process affects reporting from the 30 September 2018 census date.
Quality assurance
A number of steps are in place to ensure that these data are reported accurately.
1. Workforce information is sourced from NHS Board HR and payroll systems. These are dynamic operational systems in which the data can change over time due to their live nature and potential additional updates made by NHS Boards. Accuracy of coding is crucial to the quality of the data. The responsibility for data accuracy lies with the Boards providing the data and Board-level quality assurance practices are described below. However, NES seeks to further minimise data inaccuracies by maintaining lists of data providers and providing them with a detailed data quality report based on the extract taken prior to the publication census extract. This allows Boards to check that staff records are accurate at a global level and in regard to specific staff groups, and to amend any potentially incorrect data at source ahead of the publication extract. The QA report provided by NES includes:
2. Census data are then extracted from SWISS and added to NES' workforce data warehouse where further processing occurs, such as allocating occupational codes, and joining data for doctors and dentists in training data with Turas People. NES carries out a second phase of quality assurance at this point which involves: (i) checking for any obvious errors or missing values in the data, and (ii) comparing figures with previous extracts for any unusual trends/outliers. When necessary, NES raises any large inconsistencies with data providers.
3. Reporting software, PowerBI and Excel tables, are then refreshed to include the new data. The outputs are checked to ensure the values are shown accurately and any additional points of clarity are noted in notes pages or interactive tooltips.
4. These data are part of the two-week early release quality assurance process where outputs are made available to NHS Boards to confirm as accurate. Communication with NHS Boards over this period allows us to confirm or update any of these data ahead of release to the public.
The above quality assurance processes result in trustworthy information being released as part of the NHSScotland Workforce publication and reduce the likelihood of any issues arising after publication.
Board-level data quality assurance
SWISS sources its HR data from eESS, a single national approach to HR data management for NHS Scotland. The national eESS team supports Boards with the operation of eESS and SWISS through the publication of Standard Operating Procedures, best practice guidance, and reporting functions aimed at ensuring data completeness and robustness.
Best practice guidance for Board HR teams outlines daily, weekly, and monthly data management activities.
One such weekly activity is to action a data quality dashboard to ensure data is complete and robust for local and national reporting. The reports available through this dashboard have been partly developed with regard to the workforce statistics published by NES (formerly by ISD). Example checks include
(Data quality dashboard SOP approved by reports working group 2019. Last updated 5 February 2021)
Limitations
Sometimes there may be occasions where these data cannot be verified before they are extracted, for example due to turnover or absences of staff updating the systems. In these occasions, NES will work with providers to highlight any important quality assurance or intelligence to our end users.
Detailed information on occupations are not always available in the source data. This has led to the creation of a number manual surveys to identify staff roles in more detail (for example the managed pharmacy workforce survey).
There will likely be small discrepancies in the numbers reported locally (by NHS Boards) and nationally (by NES). This may be because the data are extracted from the live system at different times, or that the methods of extraction and processing are not the same (for example, inclusion of bank staff).
This system allows NES to report the number of staff employed in NHSScotland. This may be different from the number of staff actually working at any given time (for example due to rostering arrangements or absences).