Up to a THIRD of all coronavirus antibody tests may return as false negatives, research shows
- Researchers looked at 40 studies that analyzed antibody tests for coronavirus, which determine if someone has been infected before
- Sensitivity rates, meaning the rate at which tests return positive ranged from 66% to 97.8%
- This means that between 2.2% and more than a third of the tests can give false negative results
- Sensitivity rates were also lower among commercial test kits at 65% than in a laboratory at 88.2%
There is little to no evidence that demonstrates the accuracy of blood tests with coronavirus antibodies, a new report suggests.
Researchers reviewed dozens of antibody kit studies, which determine whether or not someone has previously been infected by the virus and has built up an immune response.
The specificity rate ranged from 66 percent to 97.8 percent, meaning the test can be false negative up to a third of the time.
In addition, commercial test kits more often missed cases than those in a laboratory.
The international team, led by the McGill University Health Center in Montreal, Quebec, Canada, says the findings should prompt governments to consider issuing “ immunity passports ” to get people back to work and support the economy.
In a new study, researchers found that the sensitivity rates, meaning the rate at which tests return positive, ranged from 66% to 97.8%. Pictured: myCovidMD medical staff offers free COVID-19 antibody testing at Faith Central Bible Church, Inglewood, California, June 19
This means that between 2.2% and more than a third of the tests can give false negative results. Pictured: Nurse Christina poses with a nasal swab at JFK International Airports Terminal 4, the first airport-based COVID-19 testing facility in the US in New York City, June 29
“These observations indicate significant flaws in the evidence regarding COVID-19 serological tests, especially those marketed as point-of-care tests,” the authors wrote.
For the report, published in The BMJThe team looked at 40 studies evaluating the specificity and / or sensitivity of antibody tests.
Sensitivity is how often the test correctly determines which people have been infected in the past and specificity is how the test identifies those who have never been infected.
Of the 40 studies assessed, more than 70 percent were China, the rest from Denmark, Germany, Japan, Spain, Sweden, the UK, and the US.
Half of the studies were not peer-reviewed, and the majority had a high or unclear risk of bias, meaning there are problems with the study design that could affect the findings.
Only 10 percent of the studies analyzed included outpatients and only two looked at tests taken at the point of care.
The sensitivity rate, that is, how often they gave a positive result, ranged from 66 percent to 97.8 percent.
This means that between 2.2 percent and 34 percent of patients get results that say they don’t have antibodies while they do.
In addition, the sensitivity rates among commercial test kits were lower than those in a laboratory, at 65 percent versus 88.2 percent.
Sensitivity rates were also lower in tests performed for the first or second week for symptom compared to tests performed after the second week.
The team says the “poor performance” of antibody tests “raises questions” about its use in making medical decisions.
Researchers added that the results “also need to pause” for governments considering using antibody test results, which issue so-called “immunity passports.”
“Higher quality clinical trials evaluating the diagnostic accuracy of serological tests for COVID-19 are urgently needed,” the authors wrote.
“Currently, the available evidence does not support the continued use of existing point-of-care serological tests.”