Focus Should be on COVID-19 Active Cases, Not Confirmed Cases

Even though the media has put a spotlight on the rising number of confirmed COVID-19 cases in the United States, that is the wrong statistic to look at when trying to assess the risk of spread. Rather than the number of confirmed cases, which stands at 852,000 nationwide and 252,000 in the epicenter of New York State, the statistic that matters is the number of active cases, which is the number of people who currently have COVID-19 and are contagious. This is the information needed when considering relaxing the social restrictions. Although the United States does not have good data on active cases, we can make judgments based on trends from outside the country, particularly Germany. The resulting estimate of US active cases is much lower than confirmed cases. Importantly, the trajectory flattened out in mid-April, with New York starting to decline.

The focus on confirmed cases is misleading for two reasons. First, the confirmed cases statistic, a tally of all COVID-19 cases since the beginning, is a cumulative number that must rise by definition. Solely focusing on this statistic trains the eye on an endless stream of bad news as we reach new record highs, regardless of whether the situation is getting better or not. Second, the vast majority of people do recover. People generally become symptom free within 14 days of testing positive. Since this disease has been in the United States for two months now, recoveries are well underway. Yet we continue to count them in the confirmed cases tally and keep pointing to that ever-rising statistic as a benchmark for how poorly we are doing in the fight against the disease.

In contrast, the number of active cases gives an indication of the current pervasiveness of the disease. Active cases can be calculated by subtracting recoveries and deaths from confirmed cases. Unlike recoveries, deaths, and confirmed cases, which are all cumulative, active cases can rise or fall. The trajectory of active cases is key to recognizing any progress containing the disease and ultimately reducing its social and economic impact.

Unfortunately, we do not have good estimates of active cases. Like most countries, the United States has done a poor job of collecting data that would determine the number of active cases. Specifically, the government is not following up with confirmed cases in a timely manner to produce realistic estimates of recoveries. The official total of US recoveries is just 77,000, which is 9 percent of confirmed cases and obviously too low. Most other countries' recovery statistics also lag far behind their tallies of confirmed cases. The severe undercount of recoveries implies that the US is grossly overestimating the trajectory of active cases.

Using trends from Germany, where authorities have been at the forefront of testing and have realistic recoveries data, we can make reasonable estimates of the prevalence of the disease in the United States. The German data show that about 90 percent of people who test positive for COVID-19 recover within two weeks. Once the virus is contracted, there doesn't seem to be any reason US recoveries should take longer or shorter than German recoveries. It is therefore my belief that the United States can apply the German recovery rate – 90 percent of confirmed cases recover two weeks after testing positive – to make a rough estimate of the number of active cases in the United States.

Based on these calculations, the number of active US COVID-19 cases is much lower than the number of confirmed cases. More importantly, the trajectory has improved notably. The number of active cases in the United States is 421,000 and has held at that level since April 15. In New York, the story is even more encouraging. The number of active cases peaked at 127,000 on April 13 and actually began a steady decline. By April 21, the number of active cases in New York had fallen to 118,000. In other words, the number of people recovering is rising faster than the number of new cases.

The number of confirmed cases, which continues to rise steadily each day, will keep doing so by definition. Moreover, the number of people who have contracted COVID-19 is probably much larger than the official number of confirmed cases, as the mildest cases and asymptomatic cases are generally not being tested. This, in turn, means that active cases are likely higher than these estimates. However, the mild and asymptomatic cases recover as well, probably sooner on average than other cases. Even at higher levels, there is no reason to doubt the flattening and decline in the trajectory of active cases. This is being confirmed in New York hospitals, which are seeing more discharges than admittances.

Viewpoint commentaries are the opinions of the author and do not reflect the views of Haver Analytics.