The "Andalusian anomaly": while Andalusia contains the curve of new infections, researchers try to find out why
With more than 33,000 cases and almost 2,200 deaths across the country, according to the Center for Coordination of Health Alerts and Emergencies, in recent days the number of new infections seems to be showing signs of deceleration and, in the best scenario, we will reach the peak of the epidemic this week. Even so it is early: the outbreak continues and the situation changes a lot from one community to another, from one province to another.
It is true. Each autonomous community is experiencing a different epidemic and, among all of them, there is one whose epidemiological evolution is particularly striking: Andalusia.
The most populous autonomous community in the country and the first in which an "autochthonous" infection was detected, has been registering decreasing percentages of new infections for three days in a row. Today's figure, the 236 positives, represents 12% growth. Less than 13.8% yesterday, 17% the day before yesterday and 27% on Friday. If we only serve the population, Andalusia should have many more than the 1,961 cases registered so far. What is happening in southern Spain? Is there really an Andalusian anomaly?
Coronavirus and wealth
With the data we have so far, we could say yes. Andalusia shows considerable containment of the case curve with respect to other autonomous communities. And there are several working hypotheses that try to explain this phenomenon. One of them, perhaps the best known internationally, is the relationship between wealth and disease.
Economist Max Roser noted that the data seems to show that there is a strong correlation between GDP per capita and confirmed cases of COVID per million people (what we call 'density' in our analyzes these days).
Our World in Data
This correlation, when we think about it internationally, shows some problems. It is clear that confirmed cases are only a part of the total number of cases and that, as can be seen from different indicators, in countries with higher GDP per capita, better diagnostic and health systems are available that allow better detection. After all, as Roser himself acknowledges, there is also a strong correlation between the number of tests and GDP per capita.
Coronavirus and per capita income. pic.twitter.com/FgDr6Olrru- 🐺 (@huyelobo) March 21, 2020
Hence, we wanted to better focus the data and look at whether COVID cases in the interior of the countries also followed a distribution closely related to the level of income. Above all, because in recent days, maps are being moved on this same phenomenon in Italy in which, in the absence of a more detailed analysis, everything seems to indicate that it does.
In the Spanish case, the truth is that we do not have official data with such geographical detail. Although there are provincial counts and some municipal approximations, not having an official statistic makes it difficult for us to accurately draw maps on which to make comparisons. When representing GDP per capita and density of the epidemic by communities, what we find is this.
As we can see, the correlation is quite strong also in our country. With small distortions such as La Rioja (due to the strength of the initial outbreak), Castilla - La Mancha (due to the fact that northern Toledo and eastern Guadalajara are de facto part of the Madrid metropolitan area) or the Balearic Islands (for its insularity and for some more factors that we will talk about later), the relationship between a higher GDP per capita and a higher density of the disease seems clear.
Hares and turtles: the composition of travelers (and the economy)
However, there are more theories that would explain this correlation. Some analysts, such as Kiko Llaneras, have suggested that this phenomenon may be related to the fact that rich countries are more connected through airports. In this sense, what we see is rather a scene of "hares and turtles" in which, simply, some countries (or regions, in this case) go 'before' others.
It is an interesting idea. After all, the busiest airport in Spain is the Adolfo Suárez de Madrid-Barajas. Behind him, we find Josep Tarradellas from Barcelona-El Prat in another of the most affected regions. However, the following airports on the list are in Palma, Malaga, Alicante, Gran Canaria and Tenerife.
In other words, this confirms the intuition that large airports may possibly accelerate the arrival of the virus. The Canary Islands and the Balearic Islands were the first communities to have cases and Malaga is the most affected province in Andalusia. What happens is that, while that may be the wick, it seems that other things are needed for the outbreak to end up exploding.
A few days ago, when we were talking about the differences between the number of deaths between Germany and Spain, an interesting concept emerged: the composition of households. Moritz Kuhn and Christian Bayer, two economics professors at the University of Bonn, explained that not only does it matter how the virus reaches a certain society, but also the social networks that already existed in those societies and that the disease could use to spread.
In this specific context, we could talk about the composition of travelers. Although we would need more detailed analysis, if we take into account the composition of the economy of these communities, it is not unreasonable to think that the Balearic Islands, the Costa del Sol, the Canary Islands or Alicante receive (percentage) more tourist travelers and Madrid receives more professional and business travelers. . Precisely the vector that Kuhn and Bayer identified as essential in the importation of the virus because they have a greater connection with the destination society than tourists.
Some more explanations
Mohammad Sajadi and collaborators
There are more theories on the table, ranging from the socioeconomic structure of the community (which, despite the increasing weight of Seville and Malaga, continues to be articulated in rather isolated economic hubs) to the global geographic situation. A few weeks ago, the Institute of Human Virology at the University of Maryland released an analysis that found a connection between a certain level of temperature and relative humidity and the expansion of the coronavirus.
The researchers realized that the main outbreaks of the virus had taken place in a very narrow strip of the world (the one we can see in the image above). Based on these data, it could be argued that Andalusia, Murcia, the southern Plateau, Ceuta, Melilla, the Canary Islands and the Balearic Islands have better conditions to prevent an explosion of cases of the virus.
Be that as it may, and as long as the evolution of the epidemic continues as before, Andalusia appears as a very interesting case study to learn about the weak points of the coronavirus and understand how we can implement better measures that help us contain the curve.
Image | Akshay Nanavati