The map shows two basic indicators of the epidemic: the incidence of cases and the rate at which they are growing.
The first tells us how much virus there is in a region, in daily cases per million inhabitants.
The second informs us of the trend - are more cases detected today than yesterday? -, measured with the reproductive number R. When R is greater than one, the virus spreads.
National evolution.
The graph below shows the updated evolution of cases, hospital admissions and deaths.
So we can see the spread of the virus in detail, but also measure its consequences.
The three figures are shown by date of notification (they are the weekly increases that the ministry communicates, divided by seven days).
Hospitals notice second wave and deaths nearly double
11 of September
The basics
.
In Spain, cases, hospitalizations and deaths continue to grow.
It is true that the epidemic does not have the aggressiveness of February or March, but hospitals are beginning to notice the stress.
The number of deaths continues to accelerate: September started with about 30 deaths a day and in the last week they have been double, about 60 every day.
The death toll in the "last seven days", which are late and are only a part of those later known, have exceeded 200 throughout this week for the first time since the end of the state of alarm.
Hospitals are filling up
.
Covid-19 patients occupy 17% of the beds available in Madrid (+8 points in three weeks), 13% in the Balearic Islands (+8), 10% in La Rioja (+8), and 10% in C-La Mancha (+7).
In these four communities the occupation is going faster, but it is growing almost everywhere.
Hospitals are not collapsed as in March, and they receive more mild patients, but the trend is worrying.
Also, revenue at UCI is growing at the same rate.
An estimate by EL PAÍS based on official data indicates that La Rioja, Madrid, the Balearic Islands and Aragón have about 30% of their beds for seriously ill patients occupied with coronavirus positives.
In Murcia it is feared that they will fill up in 15 days.
More deaths, far from March, but hundreds
.
This Friday the ministry reported in its daily report 322 deaths from the previous week (from August 28 to September 3), while six weeks before it only reported 10 or 15 (from July 17 to 23).
It's an easy way to see the trend, although those numbers leave deaths that are reported days or weeks later unaccounted for.
If we look at the total notifications, September started with about 30 deaths a day, which this week have been double, 60 each day.
These figures are far from the worst week of the first wave, when there were 5,000 deaths, but they are not irrelevant: they are 10 times the deaths from traffic accidents.
As deaths follow infections, and these are hardly showing signs of slowing down, we can fear that deaths will continue to rise for at least two or three weeks.
What happens in the provinces?
The data by province are published by date of onset of symptoms.
The situation on September 2-3 (the most recent, to allow time for diagnosis) points to an increase in positives in most provinces, but especially in those of Castilla-La Mancha, Castilla y León, the Basque Country, Madrid and Murcia.
In Galicia, Asturias and Cantabria the incidence is lower, but the trend is also negative.
A positive note?
The curve of new infections no longer grows in the communities that controlled the first outbreaks in early summer, Aragon and Catalonia.
And the same happens with the occupation of its hospitals.
Spain continues to be the country with the most cases in the EU.
195 cases per million inhabitants are detected every day.
Those infected have also grown strongly in France (110), but other countries such as Italy (24), Germany (15) or Norway (20) have lower figures.
Another bad data is that we have a very high positivity.
👉 Be careful when comparing the case figures with those of March.
Detection today is much better
and there are many mild cases that were then ignored.
Partly because of this, there will now be fewer hospitalized and fewer deaths for every known case.
However, controlling the epidemic requires keeping an eye out for early spikes
👉 Remember that the death data is late.
The normal rhythm of the disease causes cases to move earlier.
For example, when there is a rebound, it will be appreciated first in the numbers of positives, then in hospitals and, finally, in deaths.
Typically, deaths today are proportional to cases two or three weeks earlier.
👉 The hospitalized are a confirmation.
A spike in cases can sometimes be due to better detection, so hospital numbers are also useful to confirm that the spike is real.
Another similar sign is the positivity numbers.
👉 How reliable is this data?
The figures of the national evolution come out of the daily report of the ministry.
The positives come out of the SiViES system which was launched in May and do not appear to have any delays.
Regarding income, the report offers two sources, that of SiViES and that reported by hospitals, but we take the second because the ministry has recognized that some communities are not updating the first one promptly.
Those same delays surely occur with deaths, but in that case we have no alternative.
In summary, the case and admissions data appear well up-to-date, but deaths are likely underestimated.
👉 What do we know about lethality?
The last week of August, Spain registered a low fatality rate: there were 160 deaths per 20,000 cases two weeks earlier, so that the fatality per case (CFR) would be around 0.78%.
Calculating the fatality per infected (IFR) involves adjusting this figure with two unknown variables: 1) reducing it with cases that are not detected, and 2) raising it with deaths that week that are not yet known.
It is possible that both figures cancel out, and that the IFR lethality is around 0.5% or 1% for Spain.
Evolution province by province
To analyze each province we use the two basic indicators: incidence and reproductive number.
The cases that are detected every day tell us what is the expansion of the virus in the region, its presence, while the rate of infections reports its trend, to know if the outbreak is advancing or retreating.
The dates are a week or 10 days ago because they show the cases at the time they began to have symptoms, but they are cases that were known recently, perhaps yesterday or the day before.
The data is updated every Friday.
The indicators, in detail
Reproductive number
(R).
It is an estimate of the number of people that each infected person infects.
In the absence of containment measures, that figure is around 3 and the virus grows exponentially.
To avoid this, the R must be less than one.
To calculate the reproductive number from the new daily cases we use the EpiEstim package of R. We assume a mean time between infections of 4.7 days with a standard deviation of 2.9, similar to that of
this study
by
London School of Hygiene & Tropical Medicine
.
Cases per million.
We take the daily data of new positive cases (almost always by PCR) from each province and express them per million inhabitants.
National data source.
The data at the national level come from the daily reports of the Center for the Coordination of Health Alerts and Emergencies (CCAES), dependent on the Ministry of Health.
We use the reported totals — cases, admissions, and deaths — to calculate the weekly increase and divide by seven to get a daily average.
The reports give two figures for hospital admissions, the one from SiViES and the one reported by the hospitals, but we use the latter because it seems better up-to-date.
Source of series and map by provinces.
The data comes from the covid-19 situation panel maintained by the Carlos III Health Institute, based on the declaration of cases through the SiViES computer platform managed by the National Epidemiology Center.
They are data by date of symptoms: “From the start of the pandemic until May 10, the date of onset of symptoms is used or, failing that, the date of diagnosis minus six days.
As of May 11, for symptomatic cases, the date of onset of symptoms or, failing that, the date of diagnosis minus three days is used;
in asymptomatic cases, the date of diagnosis is used ”.
Updating the data.
All the information is updated normally on Thursday afternoons, which is when the covid-19 situation panel of the Carlos III Health Institute is published.
Information about the coronavirus
- Here you can follow the last hour on the evolution of the pandemic
- This is how the coronavirus curve evolves in Spain and in each autonomy
- Search engine: The new normal by municipalities
- Questions and answers about the coronavirus
- Guide to action against the disease