Mosialos: How do you comment on the IHME impact forecast of 7,000 deaths in Greece for January?

For the shock forecast of the IHME institute, which predicts 7,000 deaths from coronavirus in Greece until January, it is mentioned in a new post on Facebook.

According to Mosialos, “this is not the first time that this institute has made predictions about the course of the pandemic,” which, however, have fallen in earlier times.

“So far, for example, for Turkey, they have been announcing results, forecasting figures that a primary school student would predict, with a range of 10-118,000 killed last August,” he said.

“In general, future forecasts are always very interesting. In the case of the coronavirus, predictions about future cases or death rates also arouse public interest, but also give publicity to the authors. So let’s be careful, “he explains.

“During all this time, many institutes have made many predictions, and almost all have failed. They generally do not adapt to the changes that could potentially occur in a dynamic pandemic situation,” he said.

“At best, as I have said before, forecasts can be made for the next 2-3 weeks. It is almost impossible to predict the dynamics of the pandemic and how it will translate into cases or deaths in the next 3-4 months” he says, refuting the IHME’s forecast.

“I think we are facing another wrong forecast from the institute. One of the many wrong predictions it has made so far.” I think we need seriousness and not panic, “he concludes.

See the publication:

The new forecast from the IHME Institute (Institute for Health Metrics and Evaluation) is widely distributed, predicting 7,000 deaths in Greece for January.

This is not the first time that this institute has made predictions about the course of the pandemic. Sadly, they didn’t meet all of their initial predictions, predicting other times more or fewer deaths.

So far, for example, for Turkey they have been announcing results, predicting figures that would be predicted by an elementary school student, with a range of 10-118 thousand deaths as of last August.

More specifically, the model predicted, for August 4 • For Portugal, 471 deaths (they already had almost 1,400 deaths at the end of May) • For France 15,058 deaths (they already had almost 30,000 at the end of July) • For the United Kingdom, 66,314 deaths (and there is a total of almost 43,000 now)

In general, predictions for the future are always of great interest. In the case of the coronavirus, predictions about future cases or death rates also arouse public interest, but also give publicity to the authors. So let’s be careful.

During all this time, many institutes have made predictions, and almost all have failed, usually failing to adapt to the changes that could potentially occur in a dynamic pandemic situation.

Why;

• Because they are based on the ‘uniform’ behavior of the data population of the measures or directives
• Because they can’t include all measures in forecasting models
• Because they do not take into account geographical differences by region or country
• Why don’t you take the applicability of the measures into account?
• Because they may not include possible changes and / or adjustments to measures or directives
• Because it is impossible to include changes in therapeutic approaches
• Because it is impossible to predict the health and behavior of the vulnerable

At best, as I have said before, forecasts can be made for the next 2-3 weeks. It is almost impossible to predict the dynamics of the pandemic and how it will translate into cases or deaths in the next 3-4 months.

I think we are facing another wrong prognosis from the institute. One of the many wrong predictions you’ve made so far. I think we need seriousness and not panic.

The publication in detail: