Now that the US is opening up again, is the covid epidemic waxing or waning? First we need an accurate basis for estimating how many cases there are, then we can look at trends over time.
At midnight GMT this morning, the official US case count based on test-positive diagnoses was 2.1 million. It’s widely acknowledged that the official count is but a fraction of the total number of people who’ve been infected. Most who catch the virus never get tested, usually because their symptoms never get severe enough. Besides, testing rates vary widely across different states and regions of the country.
As a proxy, I’ve been using mortality data to back into an estimated infection rate. So far 116,825 people have died from corona in this country. Based on well-designed antibody studies, the US covid mortality rate, adjusted for the median age of the population, is around 0.6%. How many infections does it take to cause that number of deaths? 1/.006 = 167: on average, one of every 167 infected people dies of the virus. So, multiply the total deaths by 167 to estimate the total number of Americans who have been infected: 116.8K x 167 = 19.5 million infections. I.e., the actual number of infections is more than 9 times the official number, with around 6 percent of the total US population having been infected with covid.
Are infections increasing? Here are the infection rates for the four most recent 10-day intervals. First, based on official diagnostic counts:
- June 2 – June 12: 22.0K new cases per day
- May 23 – June 2: 21.8K new cases per day
- May 13 – May 23: 24.2K new cases per day
- May 3 – May 13: 24.5K new cases per day
We’re looking here not at the absolute numbers, which are gross underestimates, but at the trends. Based on the official diagnostic counts, the infection rate has remained quite stable over the past 40 days. Two possible interpretations of these findings:
- Opening up the country a month ago has not spiked the virus into a second wave. That’s a reasonable preliminary conclusion.
- The flattened case rate curve doesn’t take into account increased testing rates and decreased test-positive percentages, so the actual infection rate is on a downward trend. That’s plausible: based on national data, covid diagnostic testing rates have nearly doubled since early May while the percentage of positive tests has dropped in half. Still, criteria by which people are tested varies widely across the country, so it’s difficult to generalize.
Now, here are the estimated infection rates based on mortality data:
- June 2 – June 12: 852.6/.006 = 142K new cases per day
- May 23 – June 2: 939.5/.006 = 157K new cases per day
- May 13 – May 23: 1316.0/.006 = 219K new cases per day
- May 3 – May 13: 1700.4/.006 = 283K new cases per day
Based on this estimation method, the rate of new infections now is half what it was in early May. This estimate supports explanation 2 of the official test-positive diagnostic counts.
Are there hot spots? Yes. But even Arizona, the hottest state right now, is averaging 20 deaths per day. North Carolina, where I live, is also hot, with 20 deaths per day. The population of those two states combined is about equal to that of New York. Compare the combined AZ and NC body count of 40 per day with New York at its mid-April peak of 800 daily deaths. I.e., the current spikes are small compared with the early onslaught of the pandemic. (It’s worth noting that New York, which had great success in curbing the spread of contagion, is still averaging around 50 covid deaths per day, so they’re not out of the woods.)
Deaths are a lagging indicator of infection: someone who dies today likely became infected two weeks ago. There are reports that hospitalizations are on the rise, portending a new surge in deaths. We’ll have to wait another week or two to see how that shakes out.