Using data to improve public health: COVID-19 secondment

The COVID-19 pandemic has placed considerable pressures on health systems. During the first COVID-19 wave in the UK, a large amount of NHS activity was postponed to free up capacity to treat patients with COVID-19. Some of this postponed activity was tackled in the summer when cases were lower, but subsequent waves have had similar impacts on healthcare. While there has been greater resilience in the NHS since the first COVID-19 wave, there has still been significant disruption that continues to affect the level of care delivered. It has been estimated that there were 4 million fewer elective treatment pathways in England in 2020 than compared to 2019. Cancer screening programmes, non-essential surgeries and diagnostic procedures were postponed or cancelled. Waiting lists have continued to get longer, resulting in delayed access to care for new treatment pathways. The impacts of healthcare disruption are unlikely to have been evenly experienced across society. The COVID-19 pandemic has amplified existing social and health inequalities, with groups from low socio-economic status and from marginalised backgrounds disproportionally affected in terms of their health and social outcomes. Early evidence suggests that disruption of elective treatment pathways has been greater in deprived areas. Narrowing health inequalities represents a key government priority, and therefore tackling any inequalities resulting from healthcare disruption will be key to ensure inequalities do not widen as a result of the pandemic. The fellowship will look to answer the following overarching research question: To what extent did healthcare disruption lead to negative health and wellbeing outcomes, and for whom? To answer this research question, the project will further answer the following ancillary research questions: 1. Who was affected by healthcare disruption during COVID-19 (including during different phases of the pandemic)? 2. Where was healthcare disruption greatest and did this contribute to geographical inequalities in health outcomes? 3. Were people more likely to have a delayed diagnosis of health conditions during COVID-19 (i.e., present at healthcare later than normal), for which conditions, and for whom? To answer our research questions, we will utilise two main types of data. First, linked longitudinal records will be used to assess the impacts of healthcare disruption. We will use the core UK cohort studies, with their additional COVID-19 waves that ask individuals about their experiences of health care disruption, to examine the impact on individuals. These data have been recently linked to health and care records, allowing analyses to investigate the nature of their disruption and any impacts on health. We will follow established methods deployed by the National Core Studies teams, with individual regression-based models fit on each independent cohort dataset and then a meta-analysis to combine insights collectively. Second, we will interrogate large administrative records of health care utilisation and mortality outcomes to examine the short- and longer-term population impacts of any disruption on health systems. We will explore general measures that may be influenced by any disruption (e.g., amenable mortality), as well as investigate specific pathways through which disruption may produce negative health outcomes (e.g., cancelled cancer screenings resulting in individuals presenting at later stages of cancers). Analyses will include descriptive statistics, GIS and data visualisation, interrupted time-series and regression-based analyses.