Is the coronavirus epidemic (already) in its “second wave” in the United States? On social networks, the graph of the number of daily cases is massively shared, and for good reason: the curve, which has been decreasing since the end of April, and which then seemed to stabilize, has risen quite spectacularly since mid-April, approaching more and more the peaks of April, with more than 36,000 cases per day, and thus approaching the figures of Brazil, where the epidemic is raging.
–
Number of daily cases, smoothed * over a period of 7 days
–
The importance of weighing…
While these figures must definitely be taken seriously, it is also important to recontextualize them, especially given the size of a country like the United States, with a population of over 300 million. And precisely, if we bring back these figures of daily cases per million inhabitants, the comparison with Brazil is much less significant. Another important element, even during the flattening of its curve, the number of cases remained very high, illustrating the fact that the United States never really “beat” the epidemic, but just knocked out for a few weeks.
–
… And to take into account regional differences
But this weighting to the number of inhabitants of a country, if it allows a better comparison between countries, does not make it possible to understand what is really going on within a country. A national statistic will indeed erase regional and local differences, especially for a country as large as the United States, where it is in addition mainly to the level of state power that measures are taken to stem the epidemic.
Thanks to a datavisualisation developed by the Coronavirus Resource Center of Johns Hopkins University, we can compare the evolution of the curves of each state, and the situation is then obvious: the epidemic, in its “first wave”, was especially virulent in the northeast of the United States, especially in the States of New York (NY), Massachussets (MA), New Jersey (NJ) and Connecticut (CT). A wave since controlled, via various measures of social distancing, confinement, and quarantine.
The infographic clearly shows that the epidemic has moved to the south of the country, with significant increases in Arizona (AZ), Texas (TX), which has paused its deconfinement, Mississippi (MS), South Carolina (SC), or Florida (FL), where especially among the youngest, a little too happy to find their party places. In California too, the number of cases is rebounding, pushing the governor to impose the wearing of masks in public spaces.
Curves of the number of cases per 100,000 inhabitants, by state
The redder the background, the more the trend is towards an increase in the number of cases.
The greener the background, the more the trend is towards a decrease in cases.
–
–
Politicization of the epidemic?
Behind this shift of the epidemic to the south, some observers see the consequences of a certain politicization of the epidemic, triggered by a Donald Trump not hesitating to enter the electoral campaign when his country is going through a major crisis. The southern states are mainly ruled by Republicans, more inclined to follow an American president who continues to play down the scale of the epidemic. During the start of the epidemic, and the first containment measures, many anti-containment demonstrations erupted everywhere, sometimes with armed demonstrators, hooded. “Slingers” encouraged by Donald Trump, even calling to “save” the second amendment to the American constitution, which guarantees every citizen the right to carry a weapon for “the security of a free state”.
Geography, politics, population behavior: multiple factors are involved in the evolution of the coronavirus epidemic in this country as large as 300 times Belgium, calling for great caution in the analysis of statistics. Talking about the second wave is far too hasty, especially since the concept of the second wave is very vague and has no definite scientific definition.
–
* Data smoothing is a statistical technique allowing to attenuate “noise” and thus to obtain a more readable graph, more easily interpretable. Below, two graphs, one with the “raw” data, the other with the data smoothed over 7 days, to understand the value of smoothing.
–
–
–
