Coronavirus Trends: Towards Recovery

For the first ten days after the intensive closures began on March 15, the daily increase in the number of confirmed cases was relatively stable, usually in the 25-30% range.  Since late March the rate of increase has been falling quite steadily (with some fluctuations, common in epidemics, as new susceptible networks are suddenly exposed).  For the last 5 days it has been below 1%, and edging toward 0%. As a results the total number of confirmed cases has been flattening out.

We can see this in a different way by looking at the number of new confirmed cases. This has now (May 3) fallen to less than 100 per day, a level last seen in mid-March.

The initial increase in number of confirmed cases in the second half of March was made possible by a rise in the number of tests (the green line). Until 26.3, that increase in testing was associated with a rising percentage of positive tests (the blue line). Together, these parallel increases suggest that testing was lagging behind the number of people who had qualifying clinical conditions.  Since late March, however, the percentage of positive tests has fallen. It is now between 0.5-1.5% of all those tested.  We can also see another sign of progress here: the number of tests being administered is falling because there are insufficient numbers of people who even qualify for testing. That is, they do not have symptoms that would warrant testing under current protocols.

Here we see some overall results of these trends.  The number of active coronavirus cases peaked around 9,800 in mid-April and is now around 6,200. The number of confirmed recoveries—that is, people known to have been infected but then tested negative for the virus–now exceeds those active cases. And even though the number of deaths continues to increase, that too is flattening out. In fact, the ratio of confirmed recoveries to deaths now exceeds 40:1, which is significantly higher than equivalent trends in most other OECD countries.

Another way to look at the trajectory of the epidemic is to estimate the ratio of new confirmed cases to new confirmed recoveries. During the 20s of March there was a rapid expansion of cases, and the epidemic was still young, so not many confirmed cases were officially reaching a stage of recovery.  That led to ratios of more than 50 new cases for every recovery.  Since April 18, the ratio has fallen below 1. It is now (May 3) 0.2. In other words, there are now many more recoveries than new infections. The closer that ratio gets to zero, the better.

Another reason for optimism can be seen in trends in new critical cases. Over the 10 days, these have averaged 2 per day, reaching zero on 4 days.  Likewise, the number of patients on a respirator, which was more than 130 for most of mid-April, has now (May 3) fallen to 76.

A final reason for optimism can be seen when we look at both trends and outcomes of critical cases. The thick line is the cumulative number of critical cases. We can divide all those beneath the line into three groups: those who have recovered, those who have died, and those still in critical condition.

Over time, there appears to be more clinical success in keeping even these critical cases alive for longer. In early April, for example, there was a 9-day lag between the number of critical cases, and the number who had either died or recovered. During that period, the ratio of deaths: recoveries was around 2.6.  By late April into May, the lag was 21 days, and the ratio had fallen slightly to 2.3.  Hopefully this ratio can continue to fall as medical personnel develop (and share) tips about how best to treat these patients.