Random Testing In Indiana Shows Covid-19 Is 6 Times Deadlier Than Flu

September 1, 2020

Since the first day of the coronavirus pandemic, the United States has not had adequate testing. Faced with this shortage, medical professionals have used the tests they have on people with the most severe symptoms, or whose careers put them at high risk of infection. People who are less ill or asymptomatic are not being tested. Because of this, many infected people in the United States are not being tested, and much of the information that public health officials have about the spread and lethality of the virus does not provide a complete picture.

In addition to testing everyone in the United States, the best way to get accurate data on who and how many people are infected with the coronavirus is through random testing.

I’m a professor of health policy and management at Indiana University, and random testing is exactly what we’re doing in our state. From April 25 to May 1, our team randomly selected and tested thousands of Indiana residents, whether they were already sick or not. From this testing, we were able to get some of the first truly representative state-level data on coronavirus infection rates.

We found that 2.8 percent of the state’s population had been infected with SARS-CoV-2. We also found that minority communities – particularly Hispanic communities – were hit harder by the virus. With this representative data, we were also able to calculate just how deadly the virus is.

The process of randomized testing
The goal of our study was to find out how many Indiana residents, in total, are currently or have previously been infected with coronavirus. To do this, the population our team tested needed to accurately represent the overall population of Indiana, and we needed to test each person twice.

With the help of the Indiana Department of Health, numerous state agencies, and community leaders, we set up 70 testing sites in Indiana cities and towns. Then we randomly selected people from a list created using state tax records and invited them to be tested for free. Some groups were more likely to show up than others, and we adjusted the numbers accordingly to be representative of the state’s demographics.

Once a person shows up at one of our mobile testing sites, they receive both a PCR swab test that looks for current infections and an antibody blood test that looks for evidence of past infections.

By randomly testing and looking for current and past infections, we can extrapolate the results to the entire state of Indiana and gain information about the true prevalence of this virus.

The research team also worked with civic leaders from disadvantaged communities to also conduct public, non-randomized testing to see how the results would differ between the two testing methods.

How widespread and deadly it is
As part of the first wave of testing in the study, we tested more than 4,600 Indiana residents. This included more than 3,600 randomly selected people and more than 900 volunteers who participated in public testing.

During the last week of April, we estimated that 1.7 percent of people had an active viral infection. Another 1.1 percent had antibodies, showing evidence of previous infection. Overall, we estimate that 2.8% of the population is currently or has been infected with coronavirus, with a 95% confidence level that the actual prevalence of infection is between 2% and 3.7%.

Since our random sample is intended to be representative of the state’s population, we can almost certainly assume the same numbers for the entire state. That means that by late April, about 188,000 Indiana residents will have been infected. At that point, the official number of confirmed cases – not including deaths – would be about 17,000.

Focusing the test on severe or high-risk populations underestimates the true infection rate by a factor of 11.

Having a reliable estimate of the true number of infections also allowed us to calculate the infection mortality rate – the percentage of people who died after being infected with SARS-CoV-2. In Indiana, we calculated a rate of 0.58%. In this calculation, we divided the number of COVID-19 deaths in Indiana – then 1,099 – by the cumulative number of infections determined to be 2.8 percent – 188,000.

Early estimates suggest that 5 to 6 percent of cases in the U.S. are fatal, which is similar to the 6.3 percent obtained by dividing the confirmed cases-17,000-divided by the number of deaths-1,099 in Indiana. Thankfully, the 0.58% mortality rate from infection is much lower, but nearly six times higher than seasonal influenza, which has a mortality rate of 0.1%.

This randomized test also allowed us to make an accurate estimate of the proportion of infections that were asymptomatic. In our study, about 44% of people who tested positive for active viral infections reported no symptoms. While this has been suspected by experts, our estimate may be the most accurate to date.

Race, work and life status matter
General trends and information about the virus are important, but equally important are the ways in which human actions affect who is most affected.

We asked each person tested about their race, ethnicity, and whether they lived with someone who had been previously diagnosed with COVID-19.

Our analysis of the random sample showed that the prevalence of COVID-19 is much higher in minority communities, particularly in the Hispanic community, where about 8 percent of people are currently or have been previously infected. Although we do not know definitively why, it is possible that members of the Hispanic community in Indiana are more likely to be essential workers, to live in extended family structures that include relatives outside the nuclear family, or both.

We further found that people living with a COVID-19 positive person were about 12 times more likely to contract the virus themselves than those living in an uninfected household. Living with extended family and being more likely to be exposed to the virus due to their own work may make it easier to spread the virus in some communities.

These findings, along with a relatively low prevalence rate of 2.8%, suggest that social alienation slows the spread of the virus in the larger population. However, the most severely affected communities were those that, on average, were unable to practice social alienation as consistently as others.

What’s next?
Now that we have this information and have established a baseline, we will continue to test random samples of the state on a regular basis. Doing so will tell us the extent to which the virus has penetrated our population so that policy decisions can be made based on the situation.

This is the first statewide random sample study in the U.S., and the numbers provide both points of hope and concern.

The good news is that social distancing has worked. By slowing the spread of the virus in the community, Indiana bought some time to determine the best way forward by mitigating efforts to contain the virus to only 2.8 percent of the population. That gives researchers more time to both determine the extent to which the infection leads to immunity and to speed up the development of a vaccine.

But there’s bad news. If only 2.8 percent of people are infected with SARS-CoV-2, 97.2 percent of people who are not infected could still contract the virus. The risk of a massive outbreak that could dwarf the initial wave of people is still very real.

The demographic distribution of infections, while troubling, provides important information that can help public health officials guide testing, education and contact tracing resources that are linguistically and culturally sensitive. The research team and state health departments are working with leaders in these communities to figure out how best to control the spread of the virus in the most affected areas.

As businesses slowly reopen, we need to be vigilant about any and all safety precautions so that we don’t lose the land we gained by hibernating. Hopefully the numbers will go down, but no matter what happens in the future, we now have a better understanding of the enemy we are fighting.