Oregon football was forced to shut down operations for a few days as they had five student athletes test positive on a COVID-19 antigen test. Following protocol, over the next couple days, the five athletes were further tested twice more with a Polymerase Chain Reaction (PCR) test, which returned a negative result. The combined results of the antigen and PCR testing indicated that all five cases were false positives via antigen.

As we venture through this world of playing football during a pandemic, I thought it would be useful to highlight the tools in use, the logic behind using different testing, and why false positives are normal.

As mentioned above, there are two different types of testing used by the Pac-12 in trying to detect COVID-19 cases. Each test has its strengths and weaknesses, and each plays an important role in bringing football back as safely as possible.

Antigen testing – An antigen test detects the presence of specific proteins on the surface of a virus. A nasal swab (or saliva in some cases) is used to collect the sample and a solution frees the viral protein. The protein then travels down a strip to a line coated in antibodies. If those antibodies are triggered, the line turns a color (positive).

So what are the benefits to antigen testing? They are extremely fast, cheap (can be as low as $5 per test), can provide detection before symptom onset, and don’t require a lab to process. What are the limitations? They aren’t as accurate as the PCR, the “gold-standard” of testing.

PCR testing – A PCR test detects genetic materials (RNA) using a more complex testing system. A nasal swab is used to collect the sample, after which it is then placed in a protective liquid and kept at a consistent temperature for movement to a lab. Once the lab receives the sample, the lab technician adds primers and probes (chemicals) and subjects the sample to cooling and heating cycles which convert the RNA to DNA then multiply it (to make it detectable). DNA is then found with specific probes that cause the test to “light up” if positive.

The upside of PCR testing? Extremely accurate. The downside? Takes a lot longer to process (read: longer time to isolate players), and much more expensive ($50-100 per test). Daily testing of 100+ athletes and staff would quickly become a $5,000-10,000 bill per day, 6-7 times a week, with slow turnaround (relative to Antigen, which is as little as 1/20th the cost).

If you haven’t put two-and-two together at this point, using both types of testing is beneficial to a safe return to play. Antigen testing allows for mass screening, and PCR provides confirmation of testing results. As mentioned above, the quick response can capture a possible positive and allow isolation in an extremely fast nature. Even if you could afford to run daily PCR, it would still be a minimum of hours vs. minutes to get results.

Knowing the issues with the accuracy of antigen testing, you can’t have an antigen testing only system as false positives (and for that matter, false negatives) can occur.

Both false positive (Type I error) and false negatives (Type II error) are issues in all testing. Fortunately, the accuracy rate on PCR is extremely close to 100% (nothing is 100% accurate), so you rarely see any issues with these errors. When you have an antigen test, which might have an accuracy of 98% (seems pretty good, right?) and an actual occurrence in the population that is rather low (say 1% of people actually have COVID-19), you can end up with more false positives than actual positive results.

If we took a sample of 10,000 people to test, and 1% had COVID-19, that would mean there are only 100 true positive individuals in the group. This means 9,900 are negative for COVID-19. If our antigen test is 98% accurate, about 9,702 of those 9,900 cases will be properly found negative, 198 of those 9,900 cases will show positive (2% error). It would be extremely rare in this case to have any false negatives, but they could be possible (about 2 out of 10,000 tested).

Now if we look at these results, we tested 10,000 people and had 296 positive cases, but only 1/3 of those were truly positive! What a terrible test, right? Not really, but because of this issue, the secondary test becomes much more necessary to confirm any positive results. We now take those 296 positive cases, submit them for a PCR test and essentially eliminate the 198 false positives from the group.

Our cost is now around $50,000 for the initial round, and somewhere between $14,800-29,600 to confirm the cases via PCR. 10,000 people properly tested for $64,800-79,600 vs. $500,000-$1,000,000 to get it “right” the first time (plus confirmation testing on the initial positives, there is still Type I error).

Both tests are needed and create a superior return to play plan vs. conferences that decided to make an earlier return using PCR every few days. Using the multiple testing method isn’t fool-proof, as shown in the Ducks being taken out of commission for multiple days, but it provides the most protective and cost-efficient method that can currently be achieved. In an actual game week situation, the Ducks would have been able to move forward with five players missing, but that is a small price to pay to safely return.