Here’s what the COVID experts get wrong with an armchair

However, social media posts routinely compare COVID-19 numbers to other cause-of-death numbers that show:

Even when researchers talk about exponential growth, they can still be misled.

An analysis by an Israeli professor widely shared claimed that the exponential growth of COVID-19 “fades away after eight weeks.” Well, he was clearly wrong. but why?

His model assumed that COVID-19 cases grow exponentially over a number of days, rather than a series of transmissions, each of which could take several days. This led him to chart the irregular growth of the early stage of the outbreak.

Better visualizations cut out those first instances of irregularity, for example by starting from the 100th state. Or, they use estimates of how many days it takes for the number of cases to double (about six to seven days).

3. Not all infections are cases

Then there is the confusion about COVID-19 infection versus cases. In epidemiological terms, a “case” is a person diagnosed with COVID-19, often with a positive test result.

But there are many more infections than cases. Some infections do not show symptoms, and some symptoms are so slight that people think it is just a cold, and testing is not always available to everyone who needs it, and the test does not capture all infections.

The infection “causes” cases, and testing detects cases. US President Donald Trump was close to the truth When he said The number of cases in the United States has been high due to the high rate of testing. But it is And others I still got it wrong.

No more tests calendar In more cases, an extension is allowed More accurate estimate From the real number of cases.

The best strategy, epidemiologically, is not to test less, but to test as widely as possible, and to reduce the discrepancy between cases and infections in general.

4. We cannot compare deaths with cases from the same date

Estimates vary, but the period between injury and death could be up to a month. The variance on recovery time is greater. Some people get really sick and take a long time to recover, and some people don’t have symptoms.

So the deaths recorded on a given date reflect deaths from cases recorded several weeks ago, when the number of cases is less than half the number of current cases.

Rapid condition doubling time and prolonged recovery time also cause significant variance between numbers Active and Restored Cases. We’ll only know the real numbers later.

5. Yes, the data is messy, incomplete and may change

Some social media users Get angry When the stats are tweaked, they fuel conspiracy theories.

Few, however, realize how huge, messy, and complicated the task is to keep track of the statistics of a disease like this.

Countries and even states may count cases and deaths differently. Data collection also takes time, which means retrospective adjustments.

We’ll only find out the true numbers for this epidemic at a later time. Likewise, the early models weren’t necessarily wrong because the modelers were disingenuous, but because they had insufficient data to work through.

Welcome to the world of data management, data cleaning, and data modeling, something that many armchair statisticians don’t always appreciate. Until now.