key: cord-0939221-z2c4pp93
authors: Lindstrom, J. C.; Engebretsen, S.; Kristoffersen, A. B.; Ro, G. O. I.; Palomares, A. D.-L.; Engo-Monsen, K.; Madslien, E. H.; Forland, F.; Nygard, K. M.; Hagen, F.; Wiklund, O.; Gantzel, G.; Frigessi, A.; de Blasio, B. F.
title: Increased transmissibility of the B.1.1.7 SARS-CoV-2 variant: Evidence from contact tracing data in Oslo, January to February 2021
date: 2021-03-30
journal: nan
DOI: 10.1101/2021.03.29.21254122
sha: 8495878c0507c612fe8238f23dab890d063b933e
doc_id: 939221
cord_uid: z2c4pp93

We use data from contact tracing in Oslo, Norway, to estimate the relative transmissibility of the new SARS-CoV-2 lineage B.1.1.7. Within households, we find an increase in the secondary attack rate by 60% (20% 114%) compared to other variants. In general, we find a significant increase in the estimated reproduction number of 24% (95% CI 0% - 52%), or an absolute increase of 0.19 compared to other variants.

New, more transmissible variants of the SARS-CoV-2 pose a significant challenge to the response against the ongoing Covid-19 pandemic. The B.1.1.7 lineage that emerged in the United Kingdom (UK) in September 2020 [1] and subsequently spread to other countries and is reported to be more transmissible than the previously circulating lineages [2, 3] . There is also evidence of higher risk of hospitalisation for the B.1.1.7 variant [4, 5] . Though our study is not the first to estimate the relative transmissibility of B.1.1.7, local estimates are imperative due to differences in infection control measures, population behaviour, and other circulating variants. Contact tracing data contain important information that can be used to estimate secondary attack rates (SAR) and reproduction numbers during an emerging epidemic [6] .

Using data from the contact tracing database PasInfo in Oslo Municipality, Norway, we compared the number of secondary cases among close contacts of persons infected with B.1.1.7 with those infected with other circulating lineages to estimate the increased transmissibility of B.1.1.7 in Oslo.

The data contain information on primary cases that tested positive in early 2021, from January 4 until February 28. Primary cases were eligible for inclusion if the virus lineage test result was based on Whole-Genome Sequencing or Sanger Sequencing, and they were recorded as infected outside their household. For each primary case, the record contains age, number of close contacts, and number of SARS-CoV-2 positive close contacts within and outside the household. Close contacts are defined as persons who had direct physical or close contact (≥ 15 minutes and < 2 meter) within 48 hours of symptom onset of the case or time of test for asymptomatic cases.

During the study period, testing on arrival at the Norwegian border was obligatory, and all direct flights from the UK were cancelled for several weeks. In Oslo, social activities in homes were discouraged, cultural events were banned, teaching at universities was digital, teleworking was encouraged, restaurants could not serve alcohol, schools were in part digital except for children below 13 years. When the first non-imported cases of B.1.1.7 were detected on January 23, all non-. CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 30, 2021. ; https://doi.org/10.1101/2021.03.29.21254122 doi: medRxiv preprint essential shops closed, and teleworking was obligatory when possible. Some of these restrictions were eased on February 16 [7] .

After exclusion of 45 cases with incomplete information, the contact tracing data included 415 index cases, and their 2718 reported close contacts, of which 368 tested positive. In total, 5818 positive cases were registered in Oslo in the same period. Table 1 The total number of close contacts of index cases in the age group < 19 years is higher likely due to cohorting and school class testing. The number of contacts is overdispersed, in accordance with previous findings from Ireland [8] .

We used Poisson regression, with the number of secondary cases as the response to estimate the SAR and the relative risk and for obtaining age-adjusted estimates of these. To obtain the lineagespecific SAR, we fitted separate intercept-only models without covariates to the B.1.1.7 data and the other lineages data. The relative risk was estimated using an indicator variable for the cases infected is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 30, 2021. ; https://doi.org/10.1101/2021.03.29.21254122 doi: medRxiv preprint with the B.1.1.7 lineage. We included an offset to account for the number of close contacts in all models; hence the SAR is the probability of secondary infection per close contact. Age group was added to the models using sum-to-zero constraints to preserve the interpretation of the intercept and of the relative risk parameters. Table 2 We used the same Poisson regression approach to estimate reproduction numbers but without adjusting for the total number of contacts. The results are shown in Table 3 . The reproduction number for the B.1.1.7 lineage is estimated to be 24% greater than for the other lineages, or 0.19 difference in absolute numbers.

We estimate a 60% higher transmissibility of lineage B.1.1.7 within households. To our knowledge, this is the first estimate of the relative transmissibility of the B.1.1.7-variant in household settings.

Our estimate is compatible with results from the United Kingdom that report higher transmissibility in the 43%-82% range [2] and 40%-80% [9]. Our findings are also in accordance with results from Denmark, which report 36% higher transmissibility for the B.1.1.7 variant than the other circulating Danish variants [3] . Note that our 95% confidence interval is wide (20% -114%). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint When we also consider contacts outside households, we find lower relative transmissibility. There are two potential explanations for this. There could be bias in the data when considering all close contacts. By restricting our analysis to contacts within the household, we control for potential selection bias of contacts introduced by the contact tracing, as contacts within households are more likely to be complete in any case and thus comparable across variant. Another possibility is that there is very little difference in the transmissibility for contacts outside households because of stringent and efficient distancing interventions. Transmission in the home setting has been shown to be more important during strong interventions [10] . The fact that the estimates are not affected when we control for age means that the susceptibility by age group is similar for B.1.1.7 and for the other circulating lineages. This finding is in line with studies from other countries, e.g. from UK showing that B.1.1.7 is associated with increased transmissibility in all age groups [11] .

We estimated a household attack rate of 27% for non-B.1.1.7 lineages, which is lower than 45%, reported in a Norwegian prospective household study enrolling cases between February and April 2020 [12] . As testing was not easily available during the early phase of the pandemic, delay in testing may have increased the exposure time for household members, thereby contributing to the high transmission observed in this study. Still, both estimates are high compared to 17%, found in a systematic review of studies of the spread of SARSCoV-2 within households; however, documenting high heterogeneity between studies [13] .

The data are subject to several limitations. We cannot distinguish between the case infecting the contact or vice versa. One contact could be the contact of several cases. There can be undetected cases: close contacts who are not registered, infected contacts that escape detection, and infected contacts who have not yet developed the infection (censored). Some non-positive contacts could be immune due to a previous infection. However, we do not expect any of these problems to occur often for some lineages than others, and hence the relative estimates should be less affected by these sources of bias. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 30, 2021. ; . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 30, 2021. ; is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 30, 2021. ; https://doi.org/10.1101/2021.03.29.21254122 doi: medRxiv preprint

Preliminary genomic characterisation of an emergent sars-cov-2 lineage in the uk defined by a novel set of spike mutations

Estimated transmissibility and severity of novel sars-cov-2 variant of concern 202012/01 in England

Vurdering af smitsomhed af nye varianter

Increased Risk of Hospitalisation Associated with Infection with SARS-CoV-2 Lineage B.1.1.7 in Denmark. Preprints with The Lancet

Risk of mortality in patients infected with sarscov-2 variant of concern 202012/1: matched cohort study

Realtime estimation of epidemiologic parameters from contact tracing data during an emerging infectious disease outbreak

Tidslinje: myndighetenes håndtering av koronasituasjonen

Numbers of close contacts of individuals infected with sars-cov-2 and their association with government intervention strategies. medRxiv

Investigation of novel sars-cov-2 variant, variant of concern 202012/01 technical briefing 3

Synnøve Ygre Hauge, and Therese Bredholt Onyango. High attack rates of sars-cov-2 infection through household-transmission: a prospective study

Household Transmission of SARS-CoV-2: A Systematic Review and Metaanalysis