Population Revisited - Is Population A Factor In Early Growth Rates?

After I published yesterday's article on Case Density vs Case Counts, I got some good feedback from people who looked at it. Paul R pointed out that if I was going to factor in population, that starting everyone's timeline at 100 cases was no longer valid because 100 cases looks very different in different populations.  The better approach would be to build a timeline where Day 1 was at x% of the population for each country.  I may do this work later, but I agree that it explains why the US and China looked so flat on this graph.  We're flatter because we're looking to early on our curve relative to others if we're weighting based on population.

Whether you follow and agree with that or not, it does get at the question: When is population a factor when looking at COVID trends?

Later today, we'll be looking at testing in different countries. When criticizing the US on testing amounts, we are often compared as tests per 1M people, like this one from Business Insider:


The US Tests counts have gotten MUCH better since this chart, but do we really need to be measuring on tests per capita instead of absolute tests?  At an intuitive level, it makes a certain sense because there's so many more of us, but when most countries have infections in the low thousands do large populations really change how fast the infections spread?  To address this question, I went back to my common timeline graph.

Looking at the populations of different countries, there doesn't seem to be any correlation in the population of the country and the number of cases at this stage.  Many may wonder if that is due to low testing in the US in the early stages.  I'll deal with this question in the next blog, but there's a second piece of data we can graph to see if population is connected to growth at this point.  Here's a graph of reported fatalities on the same timeline.

Some of the lines are broken where the data isn't in my tables, but again there appears to be no connection between the population of the country and the rate of fatalities.  Even if the US was undertesting the population, it seems reasonable they would have an accurate account of fatalities from the disease.   From these two charts, I think its at least plausible that the fact we have a population several times that of many other countries being tracked, that it doesn't mean we needed several times the number of tests at this point.

So when does population figure in?  As the virus spreads, we have the potential for MANY more cases than other countries.  There is a long term threat of much higher fatality and hospitalization numbers if we don't take steps to contain the epidemic now; but in the early stages, we seem to growing at a similar pace with other countries regardless of population.

Have Questions? Think I got it wrong? Let me know in comments.