University of Leicester

Clare GilliesSharmin ShabnamTom YatesFrancesco ZaccardiKamlesh KhuntiKaren TingayNazrul IslamDr Sharmin ShabnamDr Yogini ChudasamaDr Francesco ZaccardiThomas YatesDr Nazrul IslamDaniel AyoubkhaniVahe NafilyanIvan TyukinProfessor Kamlesh KhuntiDr Cameron Razieh

HDR-UKUniversity of Leicester

Yes

618C-F540-EC17-4B04-B813-9694

The indirect effects of the COVID-19 pandemic are likely to have a big impact on people with long-term health conditions such as those with high blood pressure, diabetes, chronic kidney disease, depression, and cardiovascular disease. For this project, we plan to use data that is routinely collected in health care settings to investigate how lockdown has affected these individuals. We will see how their health has changed over lockdown, look at serious health problems that develop (including hospitalization and death), and work out the major risk factors for these events. This will help us identify which people should be given priority for face-to-face appointments as health care settings open up. In addition, we want to look at which type of people with long term conditions are more likely to be readmitted to hospital after COVID-19, and finally we will look at different methods that can be used for answering these clinical questions. We are collaborating with the Office for National Statistics (ONS) and will use data collected in hospitals, at doctor surgeries, national Census data, and mortality data. Our results will help identify which patients, including those with long term conditions and those discharged from hospital, should be prioritised for review and management by their general practices. We will work to increase the benefit to the public by sharing our results widely with others via social media, academic publications and conferences, and patient groups and forums, using culturally appropriate messaging.

Our work will provide important information for health professionals and the public on the impact of the pandemic, and subsequent reduced access to care, on the health of individuals with long-term conditions. The results will inform which patients should be given priority for accessing care, as primary care and outpatients clinics open up to more face-to-face care. We also plan to assess which individuals are at risk of adverse events (re-admittance to hospital, and death), after discharge following COVID-19. As well as addressing important clinical questions, we will utilise our datasets to assess and evaluate the use of predictive AI techniques (such as random forests and neural networks) for risk prediction analyses, and for addressing identification bias. By doing so, we seek to assess and make recommendations on the novel use of established AI methodologies going forwards. As well as disseminating our work with a methodological paper exploring AI methods for epidemiological research, we will share all our analysis codes on user-friendly platforms such as GitHub. This will include all work carried out to improve the coding of variables (e.g. ethnicity) within the accessed datasets.

23/11/2022

(e) 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;

(i) processing is necessary for reasons of public interest in the area of public health, such as protecting against serious cross-border threats to health or ensuring high standards of quality and safety of health care and of medicinal products or medical devices, on the basis of Union or Member State law which provides for suitable and specific measures to safeguard the rights and freedoms of the data subject, in particular professional secrecy;

Not applicable

Recurring

TRE

ONS guidance will be followed to ensure individuals cannot be identified in aggregated reported results and tables.