EPH169 Persons Diagnosed with COVID-19 in Linked Clinical Practice Research Datalink (CPRD) – Hospital Episode Statistics (HES) Data: A Cohort Description

Abstract

Clinical Practice Research Datalink (CPRD) Aurum captures primary care electronic healthcare records for ∼28% of the population in England. From August 2020‒;March 2022, all SARS-CoV-2 polymerase chain reaction (PCR) tests performed were reported back to the patient’s general practitioner (GP), making the CPRD a closed system uniquely positioned to answer COVID research questions. We defined persons with COVID as those recorded in primary care with a positive PCR test from August 1, 2020-March 31, 2021. We required continuous registration with their GP practice for ≥365 days prior to diagnosis to establish comorbid conditions, and eligibility for linkage to Hospital Episode Statistics (HES) Admitted Patient Care data. Hospitalizations for COVID were defined as persons admitted with a primary diagnosis of COVID (ICD-10-CM U07.1) within 12 weeks of the initial primary care diagnosis record. Our cohort included 535,453 persons diagnosed in primary care with COVID, with 2% later hospitalized. The hospitalized group was 57% male, 42% current/former smokers, 35% obese46% with a Charlson Comorbidity Index >1 and 98% had never received any COVID vaccine. Hospitalizations increased with age; <0.1% of patients aged 1-17, 1% aged 18-49, 4% aged 50-64, 9% aged 65-74, 13% aged 74-84, and 11% of COVID cases aged ≥85 were hospitalized. Persons living in socially disadvantaged areas were overrepresented in the hospitalized cohort (25% in the Index of Multiple Deprivation’s most deprived quintile). Consistent with other studies, hospitalized COVID patients were disproportionately those with male sex, smoking history, high body mass index, comorbidity and unvaccinated status. Hospitalizations were more common with age, and for individuals living in socially and economically deprived communities. Understanding the demographic and clinical characteristics of this cohort can help contextualize future work describing healthcare resource utilization and costs, as well as the impact of vaccines, associated with COVID in England.