Severity of Severe Acute Respiratory System Coronavirus 2 (SARS-CoV-2) Alpha Variant (B.1.1.7) in England

Abstract

Background: The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) alpha variant (B.1.1.7) is associated with higher transmissibility than wild-type virus, becoming the dominant variant in England by January 2021. We aimed to describe the severity of the alpha variant in terms of the pathway of disease from testing positive to hospital admission and death.

Methods: With the approval of NHS England, we linked individual-level data from primary care with SARS-CoV-2 community testing, hospital admission, and Office for National Statistics all-cause death data. We used testing data with S-gene target failure as a proxy for distinguishing alpha and wild-type cases, and stratified Cox proportional hazards regression to compare the relative severity of alpha cases with wild-type diagnosed from 16 November 2020 to 11 January 2021.

Results: Using data from 185 234 people who tested positive for SARS-CoV-2 in the community (alpha = 93 153; wild-type = 92 081), in fully adjusted analysis accounting for individual-level demographics and comorbidities as well as regional variation in infection incidence, we found alpha associated with 73% higher hazards of all-cause death (adjusted hazard ratio [aHR]: 1.73; 95% confidence interval [CI]: 1.41-2.13; P < .0001) and 62% higher hazards of hospital admission (1.62; 1.48-1.78; P < .0001) compared with wild-type virus. Among patients already admitted to the intensive care unit, the association between alpha and increased all-cause mortality was smaller and the CI included the null (aHR: 1.20; 95% CI: .74-1.95; P = .45).

Conclusions: The SARS-CoV-2 alpha variant is associated with an increased risk of both hospitalization and mortality than wild-type virus.

Keywords: SARS-CoV-2; alpha; case fatality; hospital admission.

Figure 1.

Summary population characteristics for alpha and wild-type infections. a, Regional distribution of alpha cases; b, number of alpha and wild-type cases by epidemiological week; c, number of outcomes analyzed; d, age distribution (median, IQR); e, presence of comorbidities. Abbreviations: ICU, intensive care unit; IQR, interquartile range.

Figure 2.

Relative severity of alpha compared with wild-type virus. All models include covariate adjustment for age, sex, ethnicity, smoking status, obesity status, categorical number of comorbidities, index of multiple deprivation, household size, residential rural or urban location classification, epidemiological week, and care home status, except for the Death | ICU admission model, which excludes adjustment for care home status. Cox proportional hazards regression was used; all models are stratified on region by UTLA, estimating a separate baseline hazard function for each UTLA, with model parameters estimated by maximum likelihood over the full study population. Abbreviations: aHR, adjusted hazard ratio; CI, confidence interval; Death | Hospital admission, death given hospital admission; Death | ICU admission, death given ICU admission; ICU, intensive care unit; UTLA, upper tier local authority area.

Figure 3.

Summary characteristics of deaths occurring with and without hospital admission. a, Total deaths; b, age distribution (median, IQR); c, presence of comorbidities; d, sex and care home residence proportions. Abbreviations: ICU, intensive care unit; IQR, interquartile range.