first published : 22.03.2021
Did the SARS-CoV-2 variants had an impact on Hospitalizations in Luxembourg?
background: few studies already indicated that the SARS-CoV-2 varaints B.1.1.7 (UK) and B.1.351 (South Africa) had increased the hospitalization rates and mortality rates as well, for example one recent study estimated up to 61% higher hazard of death associated with B.1.1.7 varaint. This has prompted to have a closer look at the Luxembourg data and see if we can have such correlation, especially since the week 22-28 Feb 2021 , the B.1.1.7 and B.1.351 together constitute for over 80% of varaints circulating in Luxembourg!
methodology: Luxembourg has normally a single digit mortality in terms of daily data and hence even small changes can be dramatic. Therefore we looked at the daily hospitalization admission (not the number of hospitalized as patients can be hospitalized for different length of time). Again the daily Hospital admissions vary a lot and therefore a 7-day average is used (Hosav7). Next, to normalize data over time, the 7-day average of daily new Positive cases detected is also calculated (Posav7).
Since hospitalization occurs normally 5-10 days following infection, we normalized the hospital admissions (Hosav7) by 7-day daily average from 7 days back (Posav7day-7), and calculated a 7-day average hospital admissions of per thousand Positive cases (Δ).
Now that the first case of B.1.1.7 was sampled in Luxembourg on 19 Dec 2020, here we look at the data since 19.11.20, so that 19.11.20 – 18.12.20 can act as preceeding baseline and representative of pre-varaint data. To create an ‘pseudo-baseline’ (so that the data can be shown in percentage-change), the calculated daily Δ was divided by the average of Δs from 19.11.20 – 18.12.20 (equivalent to 100%).
results: below is the percentage-change in Δ since 19.11.2020, which shows an apparent 50-100% increase in hospital admission rate since Jan 2021 (the B.1.1.7 varaint was 7.2 % of circulating strains in Luxembourg in week 1 of 2021). It has been 3 months since the first case of B.1.1.7 was detected, and for last 2.5 months the hospital admission rates remain higher than that of pre-varaints rate.
Please note that the above data shows a ‘casual relationship’ between the increase in hospital admission rate with increase in B.1.1.7 and B.1.351 variants in Luxembourg, and the author acknowledges that the data is neither peer-reviewed nor claims scientific accuracy, and welcomes scientific community for comments and suggestions.
Do we know the SARS-CoV-2 variants in daily new cases reported in Luxembourg?
Techically, all the PCR positive samples are not sequenced but only 10-20% of them (weekly). So, the graph below is a ‘best-estimation’ and does not represent accurate numbers. Also, the sequencing and data analysis takes time to be reported, so we see only the data for 1st Jan 2021 to 14 Mar 2021 below. The data (in red) show the share of B.1.1.7 and B.1.351 together in the daily total cases reported (in blue).