Introduction: A second wave of SARS-CoV-2 infection spread across the UK in 2020 linked with emergence of the more transmissible B.1.1.7 variant . The emergence of new variants, particularly during relaxation of social distancing policies and implementation of mass vaccination, highlights the need for real-time integration of detailed patient clinical data alongside pathogen genomic data . We linked clinical data with viral genome sequence data to compare patients admitted during the first and second waves of SARS-CoV-2 infection .
Methods: Clinical, laboratory and demographic data from five electronic health record (EHR) systems was collected for all cases with a positive SARS-CoV-2 RNA test between March 13th 2020 and February 17th 2021 . SARS-CoV-2 viral sequencing was performed using Oxford Nanopore Technology . Descriptive data are presented comparing cases between waves, and between cases of B.1.1.7 and non-B.1.1.7 variants .
Results: There were 5810 SARS-CoV-2 RNA positive cases comprising inpatients (n=2341), healthcare workers (n=1549), outpatients (n=874), emergency department (ED) attenders not subsequently admitted (n=532), inter-hospital transfers (n=281) and nosocomial cases (n=233). There were two dominant waves of admissions starting from around March 13th and October 20th, both with a temporally aligned rise in nosocomial cases (n=96 in wave one, n=137 in wave two). 1470 SARS-CoV-2 isolates were successfully sequenced, including 216/838 (26 %) admitted cases from wave one , 472/1503 (31 %) admitted cases in wave two and 121/233 (52 %) nosocomial cases . 400/472 (85 %) of sequenced isolates from admitted cases in wave two were the B.1.1.7 variant . The first B.1.1.7 variant was identified on 15th November 2020 and increased rapidly to comprise almost all sequenced isolates in January 2021 (n=600/617 , 97 %). Females made up a larger proportion of admitted cases in wave two (47.2% vs 41.8%, p=0.012), and in those infected with the B.1.1.7 variant compared to non-B.1.1.7 variants (48.0% vs 41.8%, p=0.042). A diagnosis of frailty was less common in wave two (11.5% v 22.8%, p <0.001) and in the group infected with B.1.1.7 (14.5% v 22.4%, p=0.001). There was no difference in severity on admission between waves, as measured by hypoxia at admission (wave one : 64.3% vs wave two : 65.6%, p=0.658). However, a higher proportion of cases infected with the B.1.1.7 variant were hypoxic on admission compared to other variants (70.0% vs 62.5%, p=0.029).
Conclusions: Automated EHR data extraction linked with SARS-CoV-2 genome sequence data provides valuable insight into the evolving characteristics of cases admitted to hospital with COVID-19 . The proportion of cases with hypoxia on admission was greater in those infected with the B.1.1.7 variant, which supports evidence the B.1.1.7 variant is associated with more severe disease . The number of nosocomial cases was similar in both waves despite introduction of many infection control interventions before wave two, an observation requiring further investigation.