Friday, January 22, 2021

Denmark: SSI Clarifies Assessment Of Transmissibility Of COVID Variant B.1.1.7

 

#15,736

Overnight there have been a number of media stories (see Daily Mail Report) suggesting the B.1.1.7 COVID variant, first detected in the UK, may be less transmissible than previously thought.  While an eye-catching headline, there may be less to this story than these stories would suggest.  

The story is based on comments made during a radio interview by Dr. Tyra Grove Krause, of Denmark's Statens Serum Institute (SSI), where she estimated the infectivity of the UK variant to be 1.36 times higher than past variants. 

Dr. Krause quickly added the estimate has a high degree of uncertainty and is based on early findings.  The radio interview quotes Dr. Krause as saying, `It sounds very, very accurate with the 36 percent, but it is not. It can be between 20 and 50 percent more contagious.'

Despite these caveats, the media reports springing from this interview are clearly suggesting that the B.1.1.7 threat may be less than previously believed.  Which may explain why the following statement was released by the Serums Statens Institute in the past hour, clarifying matters. 
In a new note, the Statens Serum Institut describes the goals and concepts used to describe the infectivity of the new virus variants. The estimates continue to show that cluster B.1.1.7 is more contagious and thus there are no changes in the current risk assessment.
Last edited January 22, 2021
The Statens Serum Institut (SSI) has issued a new note . Here, SSI describes the goals and concepts used to describe the infectivity of the new virus variants.
 
The memorandum also describes why concepts such as the number of contacts cannot always be compared directly between different countries.
 
There have been misunderstandings about how contagious B.1.1.7 is, and therefore SSI has clarified concepts and calculation methods for assessing contagion.
 
The specification has no bearing on the calculated estimates of the infectivity of cluster B.1.1.7. Similarly, it does not lead to changes in the current risk assessment for cluster B.1.1.7

In addition, the memorandum clarifies the use of the term infection rate, which is included in two notes from 2 and 4 January 2021, where the term growth rate should have been used instead.


I've translated excerpts from the SSI's new note below, which explains in greater detail why Denmark may have come up with a lower number than the UK.  Different methods, and differing amounts of data, can yield different results. 

Assessment of infectivity of new varieties

In this note, different measures of infectivity of new virus variants are reviewed, as well as given an overview of the terms that have been used so far in the description of the infectivity of cluster B.1.1.7.

Summary and conclusion

The reason why there are different figures on infectivity of cluster B.1.1.7 in Denmark and others countries, may be due in part to the use of different concepts and models for calculating infectiousness. The clarification of the concepts of the infectivity of cluster B.1.1.7 in this note has no bearing on the available calculations that estimate the contagion of cluster B.1.1.7 in Denmark. It also does not involve changes in the conclusion of the risk assessment of 9 January 2021, which states:
"Although the epidemic curve has broken, the latest sequence results show that in Denmark is a growing social infection with cluster B.1.1.7. This virus is expected to stay dominant in mid-February. To prevent exponential growth of infection with covid-19 in february, it is crucial to bring both infection rates and contact numbers down during January month.
There has therefore been a need to increase the risk level to 5 and introduce new ones restrictions to ensure epidemic control ”. 
  • On 18 December 2020, the growth rate of cluster B.1.1.7 was reported to be 70% of the English authorities. With reference to this growth rate, it has been broad communicated that cluster B.1.1.7 is 70% more contagious. The weekly percentage growth rate in the share of cluster B.1.1.7, is also reported to be 72% in Denmark in note of January 2nd.
  • The relative contact number, ie the contact number for cluster B.1.1.7 in relation to other variants, is by Imperial College in England, estimated to be 1.4 to 1.8 for cluster B.1.1.7. Ie. that it should be 40-80% more contagious. In Denmark, we have estimated the relative contact numbers for cluster B.1.1.7 to 1.36 (95% CI [1.19; 1.53]) on 14 January 2021, as an expression that B.1.1.7 is 36% more contagious than the other variants. There is in the English studio used a different model and some other assumptions for the generation time, which means that these figures cannot be translated directly into a Danish context. 
    • Adapted to the estimate from the English data for a shorter generation time, this corresponds to cluster B.1.1.7 being 34-54% more contagious, which is on a par with the range for the Danish calculations of 19-53%.
  • The estimate of the relative contact number for cluster B.1.1.7 in Denmark is still encumbered with significant uncertainties due to a relatively small data base for cluster B.1.1.7, as well as random variations in local distribution (stochastic fluctuations) of cluster B.1.1.7. the safety intervals may be too narrow precisely because of these stochastic variations and the geographical unevenness.
  • In future, the expert group will continuously update and communicate the growth rate for clusters B.1.1.7 as a measure of the growth of cluster B.1.1.7. This can be measured directly and is comparable to other countries. It is because the growth rate is independent of it method used to calculate the contact number, including choice of generation time.