Linda Patterson had a fever, a headache, chest pain and weakness.
She slept whenever she could, but rest was fleeting. Day and night started to blend together.
Two doctors presumed she was infected with the novel coronavirus, but she was never tested.
“They said, you know, the tests are scarce and that’s the problem,” said Patterson, a former high school principal who is running for the Arizona Senate.
Due to limited testing, the number of COVID-19 cases are widely undercounted in Arizona and across the country, infectious disease experts said. Reported cases don’t reflect the true number of infections in the community.
Many patients, like Patterson, have been left out.
“We’re probably missing 90% to 95% of cases currently with the amount of testing that’s going on,” said Michael Worobey, who is an infectious disease expert. He heads the University of Arizona’s department of ecology and evolutionary biology.
Increased testing is the only alternative to stringent social-distancing restrictions that come at a very high economic cost, he said.
It will be difficult to reopen society so long as there’s a dearth of testing, said Dr. Joe Gerald, an associate professor with the UA’s Zuckerman College of Public Health.
“I don’t think there’s any evidence that we actually have sufficient testing,” he said, adding that the number of daily tests has hovered around 2,000 for the last several weeks.
“That just doesn’t seem near enough.”
Testing blitz
Widespread testing is nowhere close to happening in Arizona. As of May 1, about 1% of Arizona’s population had been tested. An even smaller percent, 0.8%, has been tested in Pima County, a rate that lags behind Maricopa County.
Ideally, these numbers would be 50 times higher, Worobey said.
“That’s just the start. Then you have to do contact tracing and isolation,” he said.
The benefits of widespread testing come when those who test positive isolate themselves and when anyone else they’ve infected does the same. Each person who isolates after testing positive reduces the rate at which the virus is reproduced.
Last week Ducey announced a testing “blitz” to increase testing over three weekends, beginning Saturday, May 2. The effort aims to test 10,000 to 20,000 people each Saturday.
Among the testing sites are two locations in Tucson, one on the northwest side and one on the east side, although there have already been complaints about equity of the locations and the ability to get tested.
During a press conference last week, Ducey acknowledged that Arizona Department of Health Services data has ultimately not shown that the state has completely overcome the pandemic, but said that what they’ve seen is enough to begin gradually reopening parts of the state, starting with some retail outlets on Monday.
Ducey also defended the state’s efforts to handle a number of shortages should there be a surge in cases throughout the state, saying they’ve overcome a lack of ventilators, personal protective equipment and hospital beds should they be needed.
“There is not a trend. And what I’m looking for, what (health director) Dr. Cara Christ is looking for, are trends,’” Ducey said. “We’ve addressed each one of these problems and situations in turn. It has all been so that we can care for the individuals that contract COVID-19. Today we can do that. We’re not even showing you what’s possible ... in terms of increasing and surging field capacity if it were to be needed.”
This testing blitz, however, is not going to help us understand the full extent of the coronavirus in Arizona, Gerald said. It falls awkwardly between two testing strategies.
Nevertheless, the testing blitz will still produce meaningful data, said Tim Lant, a mathematical epidemiologist at Arizona State University. He added that biases, or issues, will arise from nearly any group of people tested.
“A lot of people are working really hard to get (the testing blitz) done,” he said. “And it is important.”
Those getting tested in the blitz won’t, however, represent the actual population that’s infected, Dr. Gerald said, so it’s not very helpful from a public health perspective. And since, the blitz isn’t testing a “representative sample,” it’s not ideal for the type of periodic public health surveillance that could show trends in cases over time.
We need this type of data to help answer questions like when Arizona can open back up, he said. “We’re kind of flying blind and so we’re trying to make educated guesses.”
The testing blitz is also not ideal for contact tracing because high-risk individuals are not targeted, he said.
Anyone who thinks they might have been exposed to the virus can get tested during the blitz, depending on each testing site’s criteria. Normally the criteria to get tested is much more stringent.
Gerald also emphasized the importance of quickly following up with infected patients. But it may take four days before the county and state get patients’ results, said Dr. Francisco Garcia, Pima County’s chief medical officer. Then public health workers must do the investigative legwork to track down the patient.
Garcia said he only expects about 300 extra tests to be done here locally, which wouldn’t cause too much extra leg work for contact tracing.
The positive test rate in Pima County is 9.9%.
“From my standpoint, there is no testing blitz, it will be a testing blip in the number of cases that we have to follow up on,” Garcia said. “I do not believe that it is going to be significantly disruptive.”
Pleading for tests
When Ducey called for the “testing blitz” last week, Garcia said he knew health clinics in the area wouldn’t have enough tests.
El Rio Health Clinic, for example, only had 10 to 20 tests a day.
Garcia pleaded with laboratories to send more tests. Health clinics in the area were getting “embarrassingly small amounts,” he wrote to LabCorp, one of the laboratories that supplies the tests. He also wrote to Sonora Quest.
Some federally qualified health centers serving upward of 200 people are getting fewer than 20 kits, he wrote. The same was the case for Arizona Community Physicians, a large private practice. They got 50 kits to respond to an outbreak at a facility with 250 residents and staff.
“We will be unable to respond to the governor’s directive to increase screening without these resources,” Garcia wrote in the letter, which was obtained by the Star.
As of Saturday morning, the Kaiser Family Foundation ranked Arizona last among states in number of tests with results per capita.
Test shortages are a problem across the country, Gerald said.
“I’m actually kind of surprised that our testing capacity hasn’t ramped up more quickly than it has,” he said.
It could be due to kinks in the supply chain, but Gerald said he doesn’t know where the logjam is.
Mike Geller, LabCorp spokesman, said the company is “not currently experiencing shortages of supplies to perform tests.”
“We are carefully monitoring and managing our supply chain and are in close contact with clients in Pima County and elsewhere and suppliers to help maintain adequate supplies to meet anticipated demand,” Geller said.
In a statement, Sonora Quest said it is going above and beyond to meet demand.
While LabCorp and Sonora Quest did deliver additional tests for the blitz, Garcia said he was still frustrated with the laboratories.
“I’m sure that they think that from their standpoint there’s not a supply-chain issue,” Garcia said. “It might not be their supply chain, but it certainly is our supply chain.
“When I can’t get my hands on 1,000 viral testing collection kits to conduct one of these blitzes, that means that there’s a problem.”
The county’s testing capabilities were detailed in a memo this week from County Administrator Chuck Huckelberry, which said that from March 11 through April 29, the county was able to do 8,456 tests, with the peak at 362 tests in a single day.
Clinics can access tests through contracts with labs, like Sonora Quest and LabCorp. The county also signed three additional agreements with companies and have a fourth on the way. The UA also donated 2,800 test kits.
Once those agreements are up and running, it’s estimated there could be 6,500 tests a week.
Other data for decision-making
Other metrics can help determine when to lift social-distancing restrictions. But testing remains the biggest obstacle, along with increasing case counts, Gerald said.
We can look at data on hospitalizations and deaths, but they are less-than-ideal metrics and should be a last resort, he said.
“Those lag several weeks behind changes in actual infections,” he said. “And (deaths and hospitalizations) are outcomes we want to avoid, if possible.”
Deaths, for example, are a robust metric to consider because it’s harder to miss someone who dies of coronavirus, although it still happens, Worobey said.
Just as there has not been widespread testing on the living in Pima County, the Medical Examiner’s Office hasn’t done widespread testing on the dead either, said Dr. Greg Hess, the county’s chief medical examiner. Only about 30% of deaths in the community are sent to his office.
If Hess’ office receives the body of someone staffers suspect may have been infected, they will do a postmortem coronavirus test, he said, adding that he has enough tests to do so.
He also examines people who died of causes unrelated to coronavirus, like suicide and drug overdose. And he wonders if the pandemic response contributed to their deaths. For example, what if isolation contributed to someone’s decision to end their own life?
“Yes, there’s a trade-off,” Gerald said. “There are also, potentially, health consequences for maintaining social distancing and social isolation.”
In his view so many people are susceptible to the virus that it’s more beneficial to maintain social-distancing restrictions.
What about the flu?
Dr. Bob England, the county’s interim health director, acknowledged the shortfall in data, saying he primarily pays attention to the deaths, hospitalizations and outbreaks sections of the dashboard to gauge the spread of the virus.
Asked to compare the pandemic with a bad flu season, he said that every year deaths from the flu are undercounted, with an official death toll calculated using a formula by the CDC based on a bunch of indicators.
He said he’s sure the same thing is happening with the pandemic, but to less of an extent.
“I think we’re undercounting deaths probably less often than we are with the flu,” he said. “But I don’t know what the actual amount of that undercount is going to be.”
He said about 200 people in Pima County were lost during a bad flu season a couple years ago, according to an estimate based on the CDC formula, while 89 have died from the coronavirus as of Sunday morning. While that means less than half have died than a typical bad flu season, that does factor in strong social-distancing measures that limited people from leaving their homes.
He said antibody testing can help gauge the actual outbreak, and that studies have shown in other states, such as California, that it’s between 50 and 85 times worse than what’s reported. That would likely put the death rate at about 0.2% — double a bad flu outbreak.
“Obviously if we hadn’t done all this dramatic social mitigation, that number would have been a lot higher,” he said. “And obviously if we hadn’t protected the health-care system from being overrun by a surge, that’s where you would have really had a problem.”
Data-based decisions
England admitted he was on the conservative side of implementing restrictions. Now he’s on the conservative side of relaxing them.
He understands the damage to the economy. He also understands the health impacts a recession can have.
He said he doesn’t fault any decisions as “long as they’re being thoughtfully made, trying to include as much information as possible on both sides of the equation.”
Tucson Mayor Regina Romero, who enacted one of those emergency proclamations that has closed businesses, said widespread testing is one of the criteria needed before Arizona lifts quarantine restrictions, labeling the availability of testing as “not adequate.”
Asked about a target number of tests, she said that she’s relying on the Pima County Health Department but that she is hoping to get to a point where more than 50% of the county’s population has been tested.
“I am relying heavily on the input and expertise of public health experts,” she said. “It’s not something that I’m making a decision on but basing it on what our public health experts are saying. What they’re saying right now is that it’s not adequate.”
Garcia was a lot less aggressive, saying he “would feel a hell of a lot more comfortable” with a testing coverage of 2% to 5%.
“We’re going to have to make decisions with the information we have,” he said. “Ideally we’d much rather have a higher rate of testing in this community, and that would be one of the data points.”
“Stuck up here at the top”
“I think people’s notion of what the peak means has changed over time, including mine,” Gerald said.
It’s become more of a plateau, he said. “We reached a peak, and we just kind of stayed there.”
It was unexpected, he said, adding that he and his colleagues have been talking a lot about this phenomenon.
He warns that if social-distancing restrictions are lifted, or if people stop obeying them, cases could quickly spike.
“We’re just kind of stuck up here at the top,” he said. “And so it gives us worry.”
