key: cord-1047755-g2dg812i
authors: Unnikrishnan, C.
title: Globally Coherent Weekly Periodicity in the Covid-19 Pandemic
date: 2020-06-04
journal: nan
DOI: 10.1101/2020.06.02.20120238
sha: 375341b829b87e9bd5335df49d05f8663b2e8d77
doc_id: 1047755
cord_uid: g2dg812i

After the major exponential phase, the Covid-19 pandemic is on a milder rate of increase globally, and past the peak in many countries. In this phase a marked systematic weekly periodicity in the Covid-19 attack has become clear, the study of which could be useful for the long term strategy to deal with the virus and the pandemic. The most important aspect of this strong weekly modulation is its global nature, phase-coherent over most of globe, independent of the geographical location, with the possible, but unconfirmed, exception of Russia and many countries in Asia and Africa, where the scale of the deaths from the pandemic relative to the population has been small so far. There could be a phase-difference of about a day between the modulations in the daily number of deaths and the daily reported cases.

The Covid-19 virus attack is the …rst global pandemic in a century. It has changed drastically the means and styles of life. Several studies on the pathological aspects of the virus, and on the possible cures and prevention are rapidly progressing. However, at this juncture, there is much uncertainty about the time required to return to near normalcy of life. There is a large amount of data accumulated by now, due to the global nature of the pandemic. It is clear that any stable systematic feature observed in the global 1 Figure 1 : The number of daily Codid-19 deaths, plotted as 3-day running average, for the USA, Europe, South America (Brazil), Asia and Africa, along with the world aggregate. The strong weekly modulation with remarkable coherence all over the globe is evident. (The hump around April 15th in the data from Asia is an artifact of an upate from China). data is potentially very useful in handling the pandemic. It is with this view that I point out a marked weekly modulation, easily noticeable in the daily data over the past several weeks, in the daily reported cases (DRC) as well as in the daily deaths (DD) due to Covid-19. The depth of modulation in the global data is as large as 30%, with enormous statistical signi…cance. Whether this has to do with any characteristic of the virus or in the virushost (human) interaction can only be answered by those who conduct studies on the biological and medical aspects. It is also possible that the modulation is some extraneous sociological factor, though this possibility is of a smaller chance, as suggested by the strength and coherence of the modulation. While a signi…cant correlation between the daily reporting of cases and the number of daily deaths could be expected, the root cause of the stable and globally universal modulation in the number of cases reported daily needs attention.

The data collected from various national sources of the countries are tabulated meticulously by many sources, and it is available as open source information [1, 2, 3] . As an indicative, yet impressive pointer, I show the 3-day running average of the daily deaths reported in the USA, Europe, South 2 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 4, 2020. . Figure 2 : The number of daily deaths in the countries Italy, France, Germany, Spain, and the UK, along with the European aggregate. 3-day running avarage is shown. The coherence of the weekly modulation is strongly evidenced.

America (Brazil), Asia, and Africa in a single chart, along with the world aggregate. The data and the plot is from the site "Our World in Data" [2] . What is unmistakable is the weekly modulation that is coherent over 3 separate continents (North America, South America and Europe). The small, but statistically signi…cant modulations in Asia follow the general trend. In any case, the stable and well de…ned weekly periodicity in the global data is indicative of a signi…cant clue about the Covid-19 pandemic.

The second plot shows the daily deaths in Italy, France, Germany and Spain, along with the data from the UK. While the modulations in the individual traces are noticeable, the coherence in the modulations across the countries makes it easily stand up, with very high statistical signi…cance, when the data is stacked to form a European aggregate. This is the strong evidence for some underlying and unifying reason, which might help in dealing with the pandemic, after detailed study.

The depth of modulation varies in the European countries, but is systematics are similar, with high coherence. Even in those countries where the population and the actual number of people a¤ected are smaller, comparatively, the same features are seen, The large depth of modulation (> 70%) 3 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 4, 2020. . in the data from Sweden is particularly noteworthy. Sweden had allowed a lockdown-less social interaction and the pandemic is still active there, linearly rising. On the other hand, Netherlands has the situation under control, well after peak, and yet, the same weekly oscillations persist strongly.

The same coherence is visible in the number of daily reported cases from everywhere in the world, where the pandemic had a strong presence so far. I show the daily reported cases globally, from Europe, the USA, Brazil, Asia, and Africa, along with the world aggregate, plotted as a 3-day running average.

This global coherence is surprising, given the fact that the factors involved in the start and the spreading of the disease vary a lot between these continents. Of course, the same trend is prominently present in the data for counties in Europe as well (…gure 4).

The same periodicity in the data on the daily deaths and the data on the daily reported cases is unmistakable. Even in Asia and Africa, where the Covid-19 cases and deaths have been lower, relative to the population, the periodic weekly modulation is visible with good statistical signi…cance. Of course, there is a small phase-shift between the peaks of the Covid deaths and the peaks of the daily reported cases, but the periodicity is the same. As mentioned earlier, the depth of the modulation varies from 20% to 40%.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 4, 2020. . is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 4, 2020. . https://doi.org/10.1101/2020.06.02.20120238 doi: medRxiv preprint

The weekly modulation that is con…rmed in both the number of daily deaths and the number of daily reported cases could be due to some temporal modulation factor in the virus itself, or more likely in the interaction of the host (humans) with the virus. Both these aspects need focussed study on the observed strong modulation with the well de…ned periodicity of about a week. Though the modulations remind one of the population modulations in the species interactions in ecological models [4, 5] , there is no further clue available for even a speculation on that front. I refrain from commenting more on the pathological aspects that might give rise to such a strong modulation. However, one needs to rule out other trivial systematics arising from sociological aspects, like patterns in reporting the disease, hospital admissions etc. I tentatively rule out such causes primarily because of the global nature of the periodic modulations, evident everywhere, irrespective of the geographical location and the economic and social environment. Also, the 'phase-di¤erence', though small, between the modulation in DRC and DD data disfavour such causes. The next steps are to be taken by the researchers in medicine as well as by the policy makers in the governments.

Covid data is compiled at the European Centre for Disease Prevention and Control

The grapher tool and the data at the site ourworldindata.org is used for the graphs in this paper

Secondary checks on the data was done with data and plots at the site www

Deterministic Mathematical Models in Population Ecology