General practice service use towards end-of-life among patients who died before and during the COVID-19 pandemic: a nationwide cohort study using the Clinical Practice Research DataLink

RESEARCH QUESTIONS What were the characteristics of the patients who died before and during the pandemic? What were the reasons for end-of-life patients accessing general practice services before and during the pandemic?

What were the patterns of general practice service use among end-of-life patients who died before and during the pandemic?

What were the factors associated with general practice service use in end-of-life patients before and during the pandemic?

BACKGROUND The pandemic of COVID-19 has been causing widespread disruption across all care settings including primary care since March 2020 in the United Kingdom.

General practices (GP), a large part of primary care, play an important role in managing, coordinating, and gatekeeping patient access to other care such as secondary/specialist care. There is little data on how the COVID-19 pandemic has influenced GP service use among end-of-life patients.

AIM To understand the impact of the COVID-19 pandemic on general practice service use in patients who are approaching end-of-life, to inform national policy development in helping recovery from the COVID-19 crisis, mitigating its consequences, and better preparing for future public health crises.

OBJECTIVES To analyse the characteristics of patients who died before and during the pandemic; To examine the reasons for end-of-life patients accessing general practice services before and during the pandemic; To compare the patterns of general practice service use among end-of-life patients before and during the pandemic; To evaluate the factors associated with general practice service use among end-of-life patients before and during the pandemic.

STUDY DESIGN A nationwide comparative cohort study.

METHODS The data source is the Clinical Practice Research DataLink (CPRD) linked to ONS mortality, practice-level deprivation, and rural/urban classification data.

All data for patients who died before and during the COVID-19 pandemic from 01/01/2019 to the latest available will be included.

The outcomes: GP consultation (number - primary, type, location), GP prescription (number), GP referral (number, type), and Tests (number).

Independent variables: patient socio-demographic and clinical characteristics, palliative care needs recognised or not, and practice-level characteristics.

The outcome, independent variables, and reasons for using the services will be described using descriptive statistics by the period of a patient s death (before & during).

The factors associated with GP service use will be identified using a generalised linear mixed model, accounting for the hierarchical structure of the data. We will also explore how various factors interact with each other to affect GP service use. The robustness of the findings will be assessed with sensitivity analysis.

Bonferroni correction will be used to control for multiple testing.

TIMELINES FOR DELIVERY Month 1: establish project steering committee, recruiting key stakeholder and the pre-doctoral researcher, workshop 1; Month 2-5: clean & harmonise data; descriptive analysis, interim results; Month 6-9: multiple regression analysis, final results; Month 10: written-up, dissemination & workshop 2.

IMPLICATIONS The data generated in this study are key to understand care needs and service gaps, identify and monitor care inequality in primary care, maintain the quality of end-of-life care in current and future public health emergencies.