Key info

22 December 2021

Neil Ferguson1, Azra Ghani, Wes Hinsley and Erik Volz on behalf of the Imperial College COVID-19 response team 

Prof Neil Ferguson

Download the full PDF for Report 50 See all reports

WHO Collaborating Centre for Infectious Disease Modelling, MRC Centre for Global Infectious Disease Analysis, Jameel Institute, Imperial College London.


To assess differences in the risk of hospitalisation between the Omicron variant of concern (1) and the Delta variant, we analysed data from all PCR-confirmed SARS-CoV-2 cases in England with last test specimen dates between 1st and 14th December inclusive. Variant was defined using a combination of S-gene Target Failure (SGTF) and genetic data. Case data were linked by National Health service (NHS) number to the National Immunisation Management System (NIMS) database, the NHS Emergency Care (ECDS) and Secondary Use Services (SUS) hospital episode datasets. Hospital attendance was defined as any record of attendance at a hospital by a case in the 14 days following their last positive PCR test, up to and including the day of attendance. A secondary analysis examined the subset of attendances with a length of stay of one or more days. We used stratified conditional Poisson regression to predict hospitalisation status, with demographic strata defined by age, sex, ethnicity, region, specimen date, index of multiple deprivation and in some analyses, vaccination status. Predictor variables were variant (Omicron or Delta), reinfection status and vaccination status.

Overall, we find evidence of a reduction in the risk of hospitalisation for Omicron relative to Delta infections, averaging over all cases in the study period. The extent of reduction is sensitive to the inclusion criteria used for cases and hospitalisation, being in the range 20-25% when using any attendance at hospital as the endpoint, and 40-45% when using hospitalisation lasting 1 day or longer or hospitalisations with the ECDS discharge field recorded as “admitted” as the endpoint (Table 1). These reductions must be balanced against the larger risk of infection with Omicron, due to the reduction in protection provided by both vaccination and natural infection. A previous infection reduces the risk of any hospitalisation by approximately 50% (Table 2) and the risk of a hospital stay of 1+ days by 61% (95%CI:55-65%) (before adjustments for under ascertainment of reinfections).

High historical infection attack rates and observed reinfection rates with Omicron mean it is necessary to correct hazard ratio estimates to accurately quantify intrinsic differences in severity between Omicron and Delta and to assess the protection afforded by past infection. The resulting adjustments are moderate (typically less than an increase of 0.2 in the hazard ratio for Omicron vs Delta and a reduction of approximately 0.1 in the hazard ratio for reinfections vs primary infections) but significant for evaluating severity overall. Using a hospital stay of 1+ days as the endpoint, the adjusted estimate of the relative risk of reinfections versus primary cases is 0.31, a 69% reduction in hospitalisation risk (Table 2).

Stratifying hospitalisation risk by vaccination state reveals a more complex overall picture, albeit consistent with the unstratified analysis. This showed an apparent difference between those who received AstraZenca (AZ) vaccine versus Pfizer or Moderna (PF/MD) for their primary series (doses 1 and 2). Hazard ratios for hospital attendance with Omicron for PF/MD are similar to those seen for Delta in those vaccination categories, while Omicron hazard ratios are generally lower than for Delta for the AZ vaccination categories. Given the limited samples sizes to date, we caution about over-interpreting these trends, but they are compatible with previous findings that while protection afforded against mild infection from AZ was substantially reduced with the emergency of Delta, protection against more severe outcomes was sustained (2,3). We emphasise that these are estimates which condition upon infection; net vaccine effectiveness against hospital attendance may not vary between the vaccines, given that PF/MD maintain higher effectiveness against symptomatic infection with Omicron than AZ (4).

Our estimates will assist in refining mathematical models of potential healthcare demand associated with the unfolding European Omicron wave. The hazard ratios provided in Table 3 can be translated into estimates of vaccine effectiveness (VE) against hospitalisation, given estimates of VE against infection (4). In broad terms, our estimates suggest that individuals who have received at least 2 vaccine doses remain substantially protected against hospitalisation, even if protection against infection has been largely lost against the Omicron variant (4,5).


中文 - Mandarin

Translation will become available shortly.

日本語 - Japanese

Translation will become available shortly.

Español - Spanish

Translation will become available shortly.

Français - French

Translation will become available shortly.

Italiano - Italian

Translation will become available shortly.

Arabic - العربية

Translation will be made available shortly