The Institute for Health Metrics and Evaluation estimates that only 7% of positive Covid-19 cases in the US are being detected, meaning case rates are actually 14.5 times higher than officially reported. The last time the infection detection rate was this low was at the outset of the pandemic, in March 2020.
As the use of at-home Covid-19 tests rises, so does the concern that most of those test results go unreported, leading to an undercount of the true number of Covid-19 cases across the country.
Another reason detection is so low, Mokdad said, is that the majority of people infected with the Omicron coronavirus variant don’t show symptoms, so they don’t even know to take a test.
Why tests are important
A lot of diseases have low detection rates, Mokdad said, like flu.
“The difference is, this is a pandemic,” he said.
When people hear that there are a lot of Covid-19 cases in their area, some may become more cautious. That can keep them from getting sick and from spreading the virus.
“We have to always stay ahead of this and not just hope it’s gone away,” said Mara Aspinall, testing expert and a professor of practice in the College of Health Solutions at Arizona State University.
Testing is important at an individual level, too. It tells people if they need to isolate so they don’t get other people sick. It also tells them if they need to seek Covid-19 treatments, which work best when the illness is caught early.
Changing the model
State health departments in Pennsylvania, Ohio and New York say percent positivity rates are no longer a reliable metric. Nevada has removed case counts from its data dashboard altogether.
State officials recognize that the ubiquitous availability of home tests has led to an underreporting of cases, particularly of mild to moderate disease, and they say it’s difficult to quantify the impact of this missing data.
But the New Jersey Department of Health says the unreliability of case data probably doesn’t impede its ability to characterize the severity of the pandemic. There are other metrics that can offer a sense of how much disease is in the community.
Cases are a good indicator of surges, but they aren’t everything, said Spencer Fox of the University of Texas’ Covid-19 Modeling Consortium. Its model is based off hospital admissions checked against antibody levels in blood samples.
Fox says it is the “gold standard” of estimation early on, but it’s “becoming increasingly complex because the landscape of immunity in the country is changing.”
Omicron is highly transmissible, and reinfections are becoming more common.
In general, though, he thinks the understanding of the case data in the pandemic is not necessarily worse than it’s been before — “but it’s possible things are changing.”
Efforts to streamline delivery of at-home test results
States like Tennessee encourage people who use at-home tests to submit their results to the manufacturers so they can be tracked. It’s unclear how many people actually do that, but there are several efforts underway to track home tests better.
This approach still relies on people who use at-home tests to report their results to manufacturers, but Association of Public Health Laboratories CEO Scott Becker hopes the AIMS system eases the process of manufacturers reporting to public health authorities.
The initiative “is in recognition of the fact that this gap exists in reporting,” Becker said.
“From a public health perspective, that’s sort of a missing data point, if you will, in the universe of testing,” Becker said of at-home test results. “That’s kind of a blind spot.”
The Association of Public Health Laboratories announced in March that it was awarded the NIH contract, limited to $8.8 million over three years, to work on improving the reporting of over-the-counter at-home Covid-19 test results to public health officials. The contract employs AIMS to collect data from at-home test manufacturers and then distribute the data to states and jurisdictions.
“An at-home test has a digital tool that goes with it, and we’re trying to build out, essentially, a spec that allows the data to flow through the hub to the places where it needs to get for state reporting,” Becker said. “There are states that are interested in getting that data and that’s really for situational awareness.”
“The FDA had asked for post-market digital solutions to report, but beyond that, the guidance was limited, and what that meant was when digital solutions were being built and these tests were being reported, the way the reporting was happening was inconsistent in different formats, and that just made the data collection poor. It made the data analysis very difficult,” said Dr. Krishna Juluru, a presidential innovation fellow at the National Institute of Biomedical Imaging and Bioengineering who is advising digital health solutions for the RADx initiative.
Juluru and his colleagues have worked on “setting up standards and common pathways for communications of these results.”
On the state level, AIMS connects to all US states and territories and has long been used to route laboratory test results to government health systems, according to the NIH, and ReportStream connects to the majority of states.
Every lab-based Covid-19 testing site is required to report diagnostic and screening test results to state or local public health authorities, but individuals who give themselves self-tests are not required to report the results. This can lead to many at-home test results going unreported, resulting in gaps in data.
“Although it is still dependent on individuals to choose to report, we still hope to address that blind spot,” Juluru said.
‘We have to be more strategic’
There are estimates of how many at-home tests might go unreported.
Of those who used the app, though, more than 75% reported their results to their state public health authorities. People were less likely to report if they had a positive test result.
“The voluntary reporting overall was low, but what we also found was that when people used a digital solution — when they actually accessed their digital tool to help guide their test, to help give them instructions on how to do it — we found that the rate of reporting was very high, in the order of more than 75%,” Juluru said. “That’s actually very encouraging.”
The overall low usage of the app may be due to a lack of education about the importance of reporting test results, the researchers said.
When it comes to Covid-19, the number of cases reported in each state will continue to ebb and flow, and it will be important to track that data and monitor the spread of disease, Becker said.
Jennifer Nuzzo, an epidemiologist and professor at Brown University, thinks more needs to be done to monitor the pandemic to make up for the lack of data.
“We’re going to see a signal. It’s just going to be harder to interpret and know what to do about it,” said Nuzzo, who is director of the Center for Pandemic Preparedness and Response at Brown.
Population-targeted surveys would help.
“We have to be more strategic. You just can’t make more home tests available,” Nuzzo said. “If you go out and intentionally sample in these neighborhoods, we would have a better understanding.”
Nuzzo’s main concern is that the country is now at a real disadvantage in understanding how the virus is transmitted. “If people are getting sick in the grocery store, then you need to know that so you encourage people to wear masks. If it is 5-year-olds, you can take measures there,” she said.
“We have so many more tools, and we are not using them in a strategic and meaningful way.”
Becker said that anyone who has Covid-19 symptoms — even just a sniffle — should get tested.
“People have this ability to test themselves with a really good degree of confidence,” he said. “The best approach is, if you test yourself and you’re positive, that’s when you reach out to a clinical setting, to your physician or a local health agency, to get a confirmatory test.”
Correction: A previous version of this story misspelled Andrew Weitz’s surname.
Rise in at-home testing means we could be undercounting Covid-19 cases even more than before