Saturday, May 23, 2020

The Czech COVID-19 prevalence studies

SARS-CoV-2-CZ-Preval

Czech republic closed down its borders early and was fairly successful limiting the COVID-19 pandemic. While demographic and risk factors (over 50K skiing vacations in Italy and Austria) were comparable to neighboring Bavaria and Austria the per million the death count is only about half that of Austria and almost 8 times lower than in Bavaria, not to mention Belgium, Sweden or Italy.

The first results of the Czech COVID-19 prevalence testing results were published on May 6th. Out of 27,011 tests 26549 passed the quality control and were evaluated. 8 cohorts were tested, 2 fully randomized representative of the general population, 1 representative of chronically ill patients and 5 for the known hotspot regions of the Czech republic which were non-representative.

The study was carried  out on behalf of the Czech ministry of health by multiple institutions under coordination of the  "Institute of Health Information and Statistics of the Czech Republic" : https://www.uzis.cz/index.php?pg=aktuality&aid=8398

Overal 107 tested positive for antibodies (IgM & IgG) out of 26549. From the representative cohorts an overall prevalence (past and acute cases) in the general population of 90.5/100,000 inhabitants was calculated for the day of 16. April 2020. This is approximately 2.25 times  of the known cases to that date - meaning that regular symptom based testing and contact tracing have detected approximately 44.5% of total cases.

Southern Moravia had the least cases with 38/100,000 while the known hotspot Litovel had 786/100,000 cases.

Out of the 107 cases detected 51% reported no noticeable symptoms, the other 49% have (retrospectively) reported this symptoms(multiple possible):
* cough: 65%
* temp >37.5: 54%
* pain in the throat: 40%
* loss of smell or taste:  35% ( https://smelltracker.org/ )
* breathing difficulties: 21%
* other - 42%

Summary: only a very small part of the Czech population (probably in the order of 0.1%) was ever infected by SARS-CoV-2. The elderly were advised to protect themselves and test results suggest that this worked - at least outside of institutional care..

Comment: Testing and contact tracing was successful at detecting cases - only about half of the total cases went undetected which seems comparable or only moderately worse than in "best performers" like Taiwan or South Korea.

The prevalence was much lower than anticipated by the authors of the study resulting in statistically somewhat underpowered data, especially the size of the fully randomized sample.

Mortality (IFR) calculated from that data seems to be around 2% which appears rather high.

Strong points:

* large sample

* (fairly) homogenous population without no-go areas

* good validation

More reading:

https://www.radio.cz/en/section/curraffrs/czech-study-shows-extremely-low-level-of-collective-immunity-to-covid-19-virus

https://en.m.wikipedia.org/wiki/COVID-19_pandemic_in_the_Czech_Republic



No comments:

Post a Comment