Background The first wave of the COVID-19 pandemic in England resulted in an approximate 50% increase in all-cause mortality during mid-March to mid-May 2020. Previous analyses have studied COVID-related deaths, but these were under-recorded early in the pandemic due to paucity of virological testing. We propose a novel method to identify excess mortality during the pandemic by accounting for usual all-cause mortality in pre-pandemic years using a large electronic primary care database. We estimated the impact of risk factors (RFs) on excess mortality during the first wave and compared these with the impact of the same RFs on total mortality in non-pandemic times.

Methods An average of 4.8 million patients aged 30–104 years active in 770 CPRD Aurum practices between 18th March and 19th May over a 6-year period (2015–2020) were included. Concurrent medical history, ethnicity, area-based deprivation and fatalities were extracted for each year. Poisson regression models fitted total mortality adjusting for age and sex, with interactions between each RF and the pandemic and reference periods. Total mortality during the pandemic was partitioned into ‘usual’ and ‘excess’ components, assuming 2015–19 rates represented ‘usual’ mortality. The association of each RF with the 2020 ‘excess’ component was derived as the excess mortality ratio (EMR), to compare with the usual mortality ratio (UMR).

Results RFs where excess mortality was greatest and notably higher than usual were age >80, black ethnicity, BMI>40, dementia, learning disability, London practices and people in care-homes. For example, people of black ethnicity vs. white had an EMR=2.50 (95%CI 1.97–3.18) compared to a UMR=0.92 (95%CI 0.85–1.00). Excess mortality was more comparable to usual mortality for sex and area deprivation. The EMR for men (1.46, 95%CI 1.32–1.60) was not significantly different than the UMR=1.34 (95%CI 1.32–1.37). The EMR=2.05 (95%CI 1.76–2.38) in the most deprived quintile (vs. least) was only slightly higher than the UMR=1.70 (95%CI 1.65–1.75). Although some RFs produced EMRs significantly lower than their UMRs (Cancer, COPD), the EMRs were still >1. However, current smoking was inversely associated with excess mortality: while current smokers were 64% more likely to die than non-smokers in 2020, when the UMR=2.12 (95%CI 2.07–2.18) was accounted for, the EMR=0.80 (95%CI 0.65–0.98).

Conclusion Utilising large electronic patient databases to study trends in excess mortality during the pandemic confirmed some reported findings (e.g. ethnicity, obesity and care-homes), but also highlighted important differences not apparent from studying cause specific mortality during the pandemic (e.g. smoking, sex and area deprivation).