The NBA is investigating pathways to a return to action and monitoring developments in COVID-19 research that could ease that return. Commissioner Adam Silver said in April that he hopes the league can be part of the movement to restart the economy in a way that doesn’t compromise safety. But the league must first create a safe environment. And one of the major keys in doing so is coming up with a plan for extensive testing that will identify those who carry COVID-19 and limit its spread.
The league has researched various ways to bring basketball back safely, sources say, including the use of a sampling procedure called “group testing,” which aims to examine a large number of people with just a few tests. The league is also contributing to a nationwide antibody study at Mayo Clinic that involves an innovative new fingerstick test kit.
Based on my conversations with sources at the league office, team executives, and medical professionals, here’s what the NBA is working on now, and what the testing process might look like if games were to resume.
Searching for Antibodies
“Hi, this is Dr. Sampathkumar. I’m so sorry I’m late,” Dr. Priya Sampathkumar said as soon as she got on a conference call with me Tuesday morning. She didn’t need to apologize. After all, she’s busy fighting a global pandemic. Sampathkumar is a consultant in the division of infectious diseases at Mayo Clinic in Rochester, Minnesota. There, she’s leading a research team working with the NBA to determine how many players and staff members have antibodies to COVID-19, which could potentially provide immunity to reinfection. The results of her research could help jumpstart the league’s reopening while also aiding the global battle against COVID-19. “With coronavirus being so new, we’re still not sure whether these antibodies will last more than a few months or if it truly means you are immune,” Sampathkumar said. “But with most viruses, when you have antibodies, it means you have immunity to the virus for several months or several years.”
Mayo Clinic has developed a fingerstick test that it believes is just as accurate as a blood draw through a vein puncture at identifying people who have been exposed to coronavirus and have since developed antibodies. They are seeking to validate the fingerstick test’s accuracy, which, if successful, could be a simpler and safer way of detecting antibodies. “It would be hugely helpful because the fingerstick blood draw could be done at home and the sample could be mailed in for testing at a lab,” Sampathkumar said. “Vast numbers of people could be tested without any need to go to the hospital or a clinic to receive a blood draw.”
The NBA has been looking for ways to support research of the pandemic since March—including recommending players who have successfully recovered from coronavirus to donate blood to the National COVID-19 Convalescent Plasma Project, which is also run by Mayo Clinic. So far, close to 400 people from about half of the league’s teams have voluntarily participated in Mayo Clinic’s antibodies study. Portland, Minnesota, Cleveland, and Boston are among those teams, and more teams may join when their facilities open, league sources say. According to Sampathkumar, over 1,000 total people have contributed to the Mayo Clinic’s study. Participants from the NBA receive both a vein puncture and a fingerstick blood draw at their respective team practice facilities using supplies that the Mayo Clinic shipped to team doctors. In addition to helping the research of antibodies, the tests help the league get a read on COVID-19’s spread amongst the NBA population.
NBPA executive director Michele Roberts told The Ringer that players were fully on board with providing samples. “Our players have embraced the opportunity to contribute to this important public health study that will help researchers better understand the prevalence of COVID-19, potentially improve care for patients, and promote long-term efforts to develop a vaccine and treatment for the virus,” Roberts said.
The NBA’s primary objective is to help a study that could be integral to the recovery efforts of the entire U.S.—including every person involved with the NBA, from concession workers to security at arenas. Finding out who has and hasn’t had the coronavirus will also be useful in determining who should and shouldn’t attend games during the pandemic, especially among older staffers.
Major League Baseball recently participated in a similar study at Stanford University, which found only 0.7 percent of over 5,000 employees tested positive for COVID-19 antibodies. It didn’t include players, but the results did show a low number of people infected with COVID-19.
Single-digit positive tests have been found across the country. Stanford also studied Santa Clara county, which found antibodies in 2.8 percent of participating residents. USC estimates 4.7 percent of Los Angeles county has antibodies. Massachusetts General Hospital found antibodies in 9.9 percent of a test group of Boston-area residents. These numbers are merely estimates based on a sample of people, but New York City does appear to be an outlier: 25 percent of its residents are estimated to have antibodies, according to a New York state study. The results of the Mayo Clinic’s study haven’t been revealed yet, but Sampathkumar said, “Based on those numbers, we still have a long way to go before we can say as a nation that we have herd immunity and are able to return to business as usual.”
The good news, both for the country and the NBA’s reopening efforts, is social-distancing measures have slowed down the spread. The bad news is there’s still a long way to go. “The epidemic is far from over. The MLB evidence, and other places, suggests they’ve successfully slowed the spread of it,” said Dr. Jay Bhattacharya, a professor at Stanford who helped conduct the MLB study. “But the serological evidence, even in the MLB study, suggests that the epidemic is too widespread to eradicate. It spreads via asymptomatic contact, people who don’t have many symptoms. People who even have mild cold symptoms can spread it, and they’re not going to show up with symptoms or go to a doctor so you won’t know they’re spreading the disease. The L.A. county numbers alone suggest 400,000-plus people have had the disease. It’s not possible to eradicate it with a lockdown, so I think we have to come to terms with that.”
Repeated Testing During Quarantine
Knowing who had coronavirus is valuable, but knowing who has the virus is most important. The latter group is infectious and can further spread the disease. Sampathkumar, Bhattacharya, and many other doctors and epidemiologists think asymptomatic carriers are largely responsible for COVID-19’s spread, which presents a major issue for the NBA’s hopes of playing games in the near future.
As I reported two weeks ago, the NBA prefers to have teams play at one or multiple neutral sites; Disney World in Orlando and MGM Grand in Las Vegas are the most likely possibilities. Other locations are also under consideration, including Houston, multiple sources say. In downtown Houston, Toyota Center, the Rockets’ home arena, neighbors the George R. Brown Convention Center; combined, they have the facilities necessary to serve as a neutral site to host games. It remains possible that teams could play games in their own arenas. On Monday, governors in three of the country’s most populous states—California, New York, and Texas—signaled they are open to having sports games without fans. MLB and the NFL plan to do just that. But playing games at a neutral site makes it easier to control variables—the more people involved, the greater the risk. With travel comes the inclusion of pilots, drivers, and hotel workers. Players and staffers would be living with family members or roommates, all of whom can’t be tracked by the league. Hosting the rest of the season at a neutral site would create less risk, though it remains to be seen what the league and players union will agree on. No matter where games are played, thousands of swabs and tests for players, coaches, and other personnel will be needed.
Sources around the league and medical professionals agree that a quarantine with each person staying by themselves for multiple days or longer would be the most effective way to reduce the chances of an outbreak. Attendees could quarantine for some amount of time and undergo multiple rounds of RT-PCR tests, which is the most common way of determining the presence of coronavirus in an individual. The test requires a swab from the back of a person’s nose and throat—which is extremely uncomfortable for the patient. It takes only a few hours to receive results, not factoring in transportation of the sample to a lab.
Those who test positive would need to quarantine off-site, since they are contagious. But negative results are tricky—there is always a risk that a person with the virus slips through the cracks even during a quarantine period with repeated testing. “It’s unlikely there are false positives. But when it’s negative, you can’t be sure a person doesn’t have coronavirus,” Sampathkumar said. “It could be a sampling error. It could be the virus levels were too low in the test. It could be that your test isn’t accurate enough.”
Research in China suggests the RT-PCR tests can have a false negative rate around 30 percent, but other studies—including a Swiss study—found that when the test is conducted properly, accuracy is over 90 percent. There is a possibility that a test doesn’t catch someone with the coronavirus, which is why repeated testing during a quarantine is crucial for the NBA’s efforts.
The odds of a false negative are significantly reduced with repeated testing because the more times a test is done, the greater the odds that the results are accurate. Let’s do some simple math to understand this a bit better.
Pretend the NBA has a population of 2,000 people that needs to get tested. Of those 2,000, let’s say the coronavirus is present in 5 percent. That would mean 100 people are currently contagious, even if they are all asymptomatic. If the NBA gave every person a single RT-PCR test that has an efficiency of 70 percent, they could expect that 30 people will be false negatives. The odds of a false negative during a second test is 30 percent of the first 30 percent, which would result in an expected nine people, and a third test would be an expected 2.7 people. If the test had 95 percent accuracy, the respective expected false negatives would be five, 0.25 and 0.01. Proper swabbing and testing leads to tremendously better results.
All it’d take is a single false negative for the coronavirus to potentially spread like wildfire in an environment with close physical contact like a basketball game. That’s why Sampathkumar warns that negative diagnoses from RT-PCR testing shouldn’t provide a false sense of security. Repeated testing is critical, as are other precautions such as CT scans and chest X-rays to check for signs of pneumonia, on-site monitoring with temperature checks, face coverings worn at meetings and in public places, regular cleaning and sanitizing, and basic social distancing practices. The NBA will likely need to deal with positive tests no matter what, but preventing a surge of cases could be enough to conclude the 2019-20 season.
A Potential Way to Limit the Tests Used
Once a quarantine period is over, basketball activities could resume. But there is still risk since the coronavirus has an incubation period with a median time of four or five days—meaning symptoms don’t arise for some time. Those at the neutral site will likely need to get swabbed often during their stay—perhaps every 48 hours.
Many tests will be needed. To limit the amount required, a testing procedure called “group testing” is under consideration by the NBA, league sources say. Group testing, also known as “pooled testing,” was created by a statistician, Robert Dorfman, during World War II to help the United States remove men from the Army who had syphilis. It’s most valuable when a disease has a low prevalence, like syphilis in the 1940s and COVID-19 today in many parts of the United States. In the past, group testing has been used to detect influenza, chlamydia, malaria, and HIV. Today, researchers in Germany, Israel, and the United States have found grouped samples can successfully detect COVID-19. The state of Nebraska has also begun utilizing group testing. It could enable the NBA to test every person it needs to and do it using far fewer tests.
Group testing is simple: Several samples from multiple individuals are taken and “grouped” together into the same lab test—this could be done randomly, or by mixing samples from members on the same team. Between five to 20 samples are usually mixed into the same test, which reduces the amount of tests needed, saving time, money, and resources. If the test is positive, those people’s samples would be separately retested to locate which individual sample(s) are responsible for the positive test in the group.
A bit more math: Using the previous theoretical example of 2,000 people, each person must have a sample drawn from a swab. Every sample could be tested individually, requiring 2,000 total lab tests. But if 2,000 lab tests are done multiple times per week, the amount of lab tests needed becomes exorbitant pretty quickly. With group testing, let’s say all 2,000 samples were put into 100 groups of 20; that means only 100 lab tests would be needed. If 10 of the 100 groups tested positive, only those groups of 20 individuals (so, 200 individuals) would need to be retested. In total, only about 300 lab tests would be needed for one round of group testing compared to 2,000 for one round of standard testing.
Repeated testing would remain a crucial step because of the possibility of a false negative. But if this stage is followed by a quarantined phase with more rigorous testing, the prevalence of the coronavirus should be low. And if a player or staffer does test positive, it would hopefully be caught quickly and before infecting others.
The league received backlash in March after the Utah Jazz used Oklahoma’s state tests and the Brooklyn Nets bought private tests to evaluate their players and staff. There was a perception that a big business was hogging tests for asymptomatic players making millions of dollars at a time when many symptomatic citizens were struggling to get tested at all. But testing is becoming increasingly accessible to the public in the United States, as hundreds of thousands of tests are conducted per day, which ranks highly per-capita compared to the rest of the world. Still, that number falls short of the millions of daily tests recommended by health experts to impede outbreaks from occurring as businesses open and communities return to work. More people need to choose to be tested, and more swabs and test kits need to be made available.
Team executives believe public perception is factoring into the league’s thinking. It’s not such a bad thing for the NBA to mirror what the public is facing. But if they do play basketball games again, perhaps the NBA can use its platform to help mainstreamize group testing or distribute unused tests to the public.
Until then, the NBA is keeping a watchful eye on other leagues. Thus far, KBO baseball in South Korea has successfully restarted its league by testing players and staffers daily. Bundesliga, the German soccer league, has just begun to play. The Premier League announced Tuesday that six of 748 people tested are positive for COVID-19—and they must quarantine for at least seven days. Allen Sills, the NFL’s chief medical doctor, also said this week he fully expects positive cases to arrive in football.
The NBA has already had its own share of positive cases, the first of which led the league to shut down operations. The NBA and the NBPA understand the risks in restarting the season, which is why they’re investigating all options to figure out a testing plan that can best find the next positive cases before they infect others. Without the right testing measures, the rest of the complex logistics of their reopening won’t matter.