Hospital workforce retention and patient outcomes research project privacy notice

This research project will investigate the economics of hospital workforce retention. This under-researched topic is highly policy-relevant, not only due to the actual English NHS workforce retention crisis, but also because workforce retention may be a key factor to increase efficiencies in the provision of healthcare, especially in a publicly funded healthcare system with regulated salaries like the English NHS.

The project will investigate two main research questions:

1. Understanding the drivers of hospital workers’ retention in the English NHS
2. Understanding the impacts of staff retention on the patient care and if there is any association between the two.

It may be that increased staff retention results in worse patient outcomes, but also that staff prefer to work in places where general conditions (e.g. local funding, equipment, protocols) are better and thus their patients have better outcomes (e.g. lower mortality). To separate these causes and understand how they impact patient care, the research team will look at the shifters (e.g. Brexit referendum) that impact staff retention only but have no direct effect on patients’ outcomes, if not through an indirect effect due to the impact on staff retention. Essentially, the study will look at changes in clinical outcomes following changes in retention of staff over several years.

The investigation will make use of high-quality large administrative datasets with records on the hospital workforce and patients’ outcomes (e.g. mortality, readmissions, length of stay) for Acute and Mental Health English hospitals only, from 2009/10 to 2021/22. The study will include all NHS staff and patients as this will allow answering the research questions both at the national and the regional level.

The analysis will last for four years, but it will take at least six years from the start of the project to publish most of the research outcomes.

To enable us to carry out our research, we access and process secondary data collected by the Department of Health and Social Care (DHSC), NHS Digital (NHSD), and the Office for National Statistics (ONS). For the statistical analyses of our project we use the information technology and analysis capability of the Clinical Informatics team based at the University of Surrey.

Data is provided annually or biannually, depending on the data provider, and all personal data are pseudonymised by the data providers. The data is accessed and processed through a secure portal to the University’s data storage facility. The portal is accessed through a dual-factor identification system which is available only to the research project team members. The secure IT system is designed to preserve the confidentiality of NHS patients’ and staff records, and to prevent the leakage of administrative data at individual staff or patient level outside the secure system itself.

The data for the project will be processed within the UK. Data processing is permitted outside the UK during temporary working visits abroad of project team members to Italy, Australia and the United States of America, subject to the following conditions: the data is accessed only remotely via a secure connection protocol (virtual desktop solution for remote access or equivalent) from Surrey servers; during the potential stop-offs en route, any device being used to access the data if being transported with the individual must be shutdown, services and virtual desktop solution for remote access logged out and all sessions terminated prior to travel between countries; any incident that arises during the travel and whilst abroad, is reported to NHSE with immediate effect. Additionally, for travels towards Australia and the US, the laptop will not be turned on during the journey to that country.

The research will process only patients’ pseudonymised information, which is needed to investigate patterns in hospital performance and patients’ health outcomes due to variations in hospital workforce retention. It will also process pseudonymised information on English hospital workers in order to investigate the supply and demand of healthcare factors having a direct influence on variations in hospital workforce retention over time and across hospitals.

The processing of identifiable NHS staff records may happen on an ad hoc basis and conditional on the provision of data from Department of Health and Social Care. This will happen during the second year of the project and it will be necessary in order to: a) follow doctors employment spells through time, within and across NHS organizations, so that we can produce accurate models for the hospital doctors retention patterns based on individual characteristics; b) link doctors employment to health and process outcomes for patients treated in NHS hospitals, in order to  investigate the effect leavers doctors on patients’ outcomes.

The patient information that we are processing include clinical records of diseases, including records of date of admission and discharge from hospital, reasons for admission to hospital, comorbidities, hospital treatments received, and basic demographic and geographical characteristics such as age, sex, NHS Region, LSOA of residence, ethnicity, deprivation, urban/rural living, civil registration date of death.

The hospital worker information that we are processing include the NHS Trust’s Electronic Staff Records, including categories of NHS workers (e.g. consultants, nurses, ambulance workers),  their salary or wages, the NHS Trust organisation where they work, their full-time equivalent, the period of employment at their respective NHS organisations, and basic demographic and geographical characteristics such as age, sex, NHS Region of the NHS Trust, ethnicity.

We will use this information to produce statistical analyses which will feed into working papers, presentations to academic and non-academic audiences, policy briefs, and peer-reviewed publications. The project will not involve any ‘automated decision making rights’ processing (i.e. all analysis will always need human involvement) and we will not use profiling of any sort with respect to any patient nor member of staff.

Our lawful basis for processing personal data is Article 6(1)(e) of the GDPR, in that processing is necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller. Our lawful basis for processing special category data is article (9)(2)(j) of the GDPR, in that processing is necessary for scientific research purposes in accordance with article 89(1) of the GDPR.

The project will be compliant with any other relevant legislation, e.g. any NHS statutory requirements, including the ‘Health and Social Care Act 2012’ and the core NHS Values and the patient’s right to data confidentiality as stated in the ‘NHS Constitution’.

Only pseudonymised data will be retained. In general, research data are archived in our secure and private server for 10 years after the completion of the project in accordance with the Open Research Policy (PDF) of the University. For this project, source data on both NHS patients and staff will be erased as soon as the last publication arising from the study has taken place.

No personally identifiable data of patients will be used in our project. Any other personally identifiable data, where used in specific projects with appropriate approvals, are used only for a brief duration within our secure network for the specific purpose of de-identification and data linkage. All interim files created during the de-identification process containing strong identifiers will be irreversibly deleted immediately after the de-identification process.

Patient level and staff level pseudonymised data are accessible to the immediate study team who are authorised to work on the data from secure workstations within the secure network. The pseudonymised data held in our secure servers will not be shared with other organisations.

The findings of our research may be published in economics, social science or healthcare journals, and in conference presentations, working papers or policy briefs and reports to the funder of the study. The published research will be available to the wider clinical and academic communities, and it will become open research conditional on the availability of the funder to cover open access publications fees.

Individual patients or staff will never be identified in any report or publication; NHS organisations will not be identified in any report or publication without specific written consent.

Information will be archived in our secure and private server for 10 year after the completion of the study in accordance with the Open Research Policy (PDF) of the University. Unless there are prior patient consent or Section 251 of the NHS Act 2006 approval, only pseudonymised data will be retained.

Personally identifiable data, where used in specific projects with appropriate approvals, are used only for a brief duration within our secure network for the specific purpose of de-identification and data linkage. All interim files created during the de-identification process containing strong identifiers will be irreversibly deleted immediately after the de-identification process.

Details on university-wide measures surrounding IT security can be found in Our Data Policy Statement (PDF) (incorporating Information Security Policy).

The patient data that we use is collected by NHS Digital and it is subject to the national data opt-out policy. As we do not hold any identifiable patient data for our project, you must follow the procedures publicised on the NHS Digital website if you do not want your pseudonymised NHS data to be used in this study.

As an individual whose data we process (a data subject), you have certain rights in relation to the processing. Find detailed information about your rights as a data subject.

You have the right to:

• Withdraw your consent for us to process your personal data where we have relied on that consent as our basis for processing your data.
• Ask us to confirm that your personal data is being processed and to access (i.e. have a copy) of that data as well as to be provided with supplemental information about the processing.
• Ask us to stop processing data for a period if data is inaccurate or there is a dispute about whether or not your interests override the University’s legitimate grounds for processing data
• Request that we rectify any inaccuracies where the data we hold on you is inaccurate or incomplete.
• Have your data erased by us, although in certain circumstances we may not be able to do this. The circumstances where this applies can be found in the guide to data subject rights information.
• Restrict the processing of your personal data in certain ways.
• Obtain your personal data for reuse.
• Object to certain processing of your personal data.

If you would like to exercise any of your rights please visit our make a privacy request section.

Make a complaint

If you have any concerns about the way that we have handled your personal data please email Dr Giuseppe Moscelli as we would like to have the opportunity to resolve your concerns.

If you’re still unhappy, you have the right to complain to the Information Commissioner’s Office (an independent body set up to advise on information rights for the UK) about the way in which we process your personal data.