The US Centers for Disease Control and Prevention (CDC) and several state health departments have reported COVID-19 diagnostic tests and antibody tests as one big result, rather than keeping their results separate, The Atlantic reported.
Reporting these numbers as a lump sum, rather than two separate data points, presents several major problems.
Namely, the combination of the numbers could make American diagnostic capabilities and rates higher than they actually are, according to The Atlantic. And since tests serve profoundly different purposes, the “positive” results of one test or the other cannot be interpreted in the same way. Reporting all of the positive results together, in a single number, could distort our understanding of the number of new cases of COVID-19 that emerge over time – a crucial measure to help control epidemics like states begin to reopen.
Related: 11 (sometimes) fatal diseases that have crossed species
When told how the CDC chose to combine the results of the two tests, Harvard Global Health Institute director Ashish Jha told The Atlantic: “You have to make fun of me … How could the CDC make this mistake? a mess. “
Diagnostic tests for COVID-19 work by detecting viral genetic material in nasal swabs, throat swabs and saliva samples taken from patients, Live Science previously reported. The tests, although not 100% accurate, may reveal whether a given patient has an active COVID-19 infection now, at the time of the tests. New, saliva tests appear to be more reliable than swab tests, in terms of minimizing false results.
Antibody tests cannot be used to diagnose an active COVID-19 infection; tests instead scan patients’ blood samples to find evidence of previous infections. During infection, the body produces an immune response against the virus by building specific antibodies – molecules that cling to the virus and help neutralize it, Live Science previously reported. A positive antibody test result would indicate that the person has already been exposed to the virus and probably already recovered from an infection. If a person has a negative antibody test result, they may still have an active COVID-19 infection because their body has not yet accumulated enough specific antibodies to be detected by the tests.
A positive antibody test result may also indicate that the individual is now somewhat immune to the pathogen; however, more reliable antibody tests and more clinical data are needed to find out if a positive antibody test actually involves immunity and how long this immunity could last.
“The viral [diagnostic] the tests are to figure out how many people are infected, while the antibody tests are like looking in the rear view mirror, “Jha told The Atlantic. The two tests are completely different signals. “
The combination of these two signals makes the data difficult to interpret and could mislead the public, as the combined number does not reflect the rate of new infections (and the number of infectious people circulating) in their area, William Hanage, professor of epidemiology at the Harvard TH Chan School of Public Health, told The Atlantic. “Combining a test that is designed to detect a current infection with a test that detects an infection at some point in the past is really confusing and cloudy,” he said.
In addition, people are more likely to get diagnosed if they are suspected of being infected, which means they have symptoms of the disease or have been in contact with an infected person. Antibody tests can be given to anyone, and because relatively few people are infected with COVID-19 in the United States, many antibody tests can be negative. When combined with diagnostic test results, antibody test results could artificially lower the percentage of positive tests out of the total.
The percentage of positive diagnostic tests gives an idea of the amount of virus currently circulating in a community and, if biased, this skewed statistic could wrongly influence public health policy decisions.
In addition to the CDC, Pennsylvania, Texas, Georgia and Vermont also report their antibody test and diagnostic results as a grand total, according to The Atlantic. Virginia and Maine have also mixed their data before, but they started reporting the numbers separately in the past week.
According to The Atlantic’s COVID monitoring project, COVID-19 testing capacity in the United States has more than doubled in the past month, from approximately 147,000 tests per day in mid-April to more than 413,000 tests per day on May 20. At the same time, the proportion of positive tests recorded each week increased from 10% to 6% of the total of tests carried out. The two trends look encouraging at first glance, but since diagnostic and antibody tests are sometimes pooled, it’s impossible to know if the data reflects reality, The Atlantic reported.
Asked about the mess, CDC spokeswoman Kristen Nordlund said the agency “hoped” to separate the data from their COVID Data Tracker in the coming weeks.
Originally posted on Live Science.