Politics, Policy & Law
What it will take to get to mass asymptomatic screening for COVID-19
A look at the barriers to mass asymptomatic testing, and the path to removing them
A look at the barriers to mass asymptomatic testing, and the path to removing them.
Making widespread COVID-19 screening a priority for public health will require a broad and abundant basket of testing technologies, and a coordinated plan for how to use and pay for them.
The first barrier is having enough tests. But even as more manufacturers ramp up production of more types of tests, confusion about whether and how asymptomatic testing is recommended, regulated and paid for will limit their reach.
With CDC flip-flopping on asymptomatic testing, and questions swirling about whether other agencies have inappropriately influenced its guidelines, the baton has been taken up by the private and philanthropic sectors, and the workplaces, schools, and state and local governments following their lead.
Instead of converging on a single asymptomatic testing technology and protocol, these groups are creating and implementing strategies that factor in the target populations’ pre-test probability of infection, and the frequency and conditions under which testing is feasible in their environment.
Such multi-pronged strategies present opportunities for test developers of all stripes.
But maximizing those opportunities will require shifting away from healthcare models that have been optimized to control spending on testing and that are heavily weighted toward testing in centralized labs instead of more distributed models, including point-of-care and home-based tests (see Back to School: “Governments Must Work Alongside Industry to Fill the Cracks Exposed by COVID-19”).
The emerging picture that 30-60% of SARS-CoV-2 infections are transmitted by asymptomatic people makes it clear that testing seemingly healthy populations is not a wasteful luxury, but rather a critical need.
“We’re talking about something that isn’t just about individual care. This is about a public good,” said Margaret Hamburg, who served as FDA Commissioner from 2009-15. “It’s about public health, and also about protecting our workforce, protecting our economy, protecting our social fabric, and protecting our national and international security.”
However, with the exception of screening programs for nursing homes, the federal government has sent mixed messages on the value of asymptomatic testing, and how to reduce it to practice.
The most notorious example is the CDC, which in August made controversial changes to its guidelines that downplayed the need for asymptomatic testing, even in cases of known exposure (see “CDC Reversal on Asymptomatic Testing Prompts Backlash”).
As of Friday, those changes have been reversed. But reports suggesting they were the work of HHS officials instead of CDC scientists have further jeopardized public confidence in the agency’s independence.
In a press briefing on Aug. 26, HHS Assistant Secretary for Health Brett Giroir said the policy’s goal was “appropriate testing, not less testing,” with priority going to prospective surveillance studies instead of individuals seeking to make personal decisions such as whether to visit an elderly relative, who might have misplaced confidence in the results.
But Rahul Dhanda, co-founder, president and CEO of Sherlock Biosciences Inc., thinks that distinction is arbitrary, particularly in the absence of coordinated guidance for individual decision-making. Sherlock is developing tests based on CRISPR and synthetic biology.
“Asymptomatic testing is more valuable if it’s used more broadly, because more people who are going to make individual decisions are going to have more information than none,” he said. “Nothing will ensure everyone makes the right decisions, but will they be more informed decisions? Yes, they will."
Giroir said the change had not been prompted by testing shortages, and that the U.S. was “flush with testing.”
But capacity is unquestionably a barrier to mass asymptomatic screening. According to a report by The Rockefeller Foundation and the Duke-Margolis Center for Health Policy released Sept. 9, the U.S. would need about 200 million tests per month to routinely screen schools and nursing homes, yet is reporting fewer than 25 million tests per month.
That gap could be closed by big ramp-ups in production of rapid point-of-care COVID-19 tests, with antigen tests particularly on the rise — which are less sensitive, but cheaper, than more commonly used polymerase chain reaction (PCR)-based tests (see “Roche, Abbott Pump Up the Volume on COVID-19 Antigen Tests”).
But even if enough capacity were available, authorizing and reimbursing broad asymptomatic screening tests is a big departure from business as usual in a system set up for limited use of diagnostics in clinical management.
“The problem is we have to operate in a world where we have to align existing systems for reimbursement and delivery of services in midst of a public health crisis,” Hamburg said.
In an interview for BioCentury’s Back to School issue, former FDA Commissioner Scott Gottlieb said those systems are part of the reason the U.S. lags behind other countries in use of point-of-care tests (see “Surveillance Capacity Needed to Detect Future Pandemics”).
“The reimbursement system is somewhat biased towards lab-based testing, the way Medicare reimburses tests,” Gottlieb said. “It's also a function of the fact that to the extent that the lab-based tests provide a higher degree of sensitivity and specificity, we prefer the incremental margin of rigor offered by lab-based tests and don't necessarily value the access that's facilitated by an antigen-based test or other kinds of point-of-care diagnostics.”
New regulatory paths
FDA’s communications to industry make it clear the agency values that rigor, but is willing to hear out alternative strategies.
The agency’s default view, reflected in the standards set by its templates for asymptomatic tests, is that the risks of harm to healthy people require screening tests to meet a very high bar for performance.
But FDA has also said it is willing to consider regimens that compensate for lower sensitivity with a higher frequency of testing, or limit a test’s intended use to deeming negative results “presumptive negative” (see “Asymptomatic Testing Pushes Forward”).
“They are open for dialogue and discussion,” said Dhanda. “Their position is, in the absence of somebody telling me there's a good reason to do it differently, follow the template. If your reason is good, our job isn’t to stop testing that could be helpful, it’s to facilitate testing that could be helpful.”
The vast majority of FDA-authorized COVID-19 tests are only indicated for use in people suspected of SARS-CoV-2 symptoms. On July 24, the agency gave its first emergency use authorization (EUA) for COVID-19 testing of asymptomatic individuals to a molecular test from Laboratory Corp. of America Holdings (NYSE:LH), but only if they have a prescription from a physician.
An HHS communication in August stating laboratory-developed tests (LDTs) for COVID-19 can bypass FDA could theoretically open new avenues to asymptomatic testing (see “New Wild West on Horizon”).
But because the rapid point-of-care tests poised to support the biggest growth in asymptomatic testing capacity are in vitro diagnostics (IVDs), which do fall within FDA’s purview, the policy change is unlikely to have a big impact on screening capacity, said Mara Aspinal, professor of practice in biomedical diagnostics at ASU’s College of Health Solutions, who co-authored the Rockefeller Foundation report.
FDA offers guidelines for physicians ordering authorized COVID-19 tests off-label in asymptomatic populations, which is one of the mechanisms supporting college and workplace screening programs that have popped up around the country.
But some companies are building an evidence base to support authorization of their tests for asymptomatic screening.
For example, Quidel Corp. (NASDAQ:QDEL) is using data from studies of Pac-12 Conference athletes and University of Arizona students to make a case for how its tests perform in the real world; in a study of 25,000 students, the company said the test had between 96% and 98% positive percent agreement with a more sensitive PCR-based test, a metric of its sensitivity, and an 100% negative percent agreement, meaning it was equally specific.
“We think it’s important to get as much data as possible, so that we can add it to our package insert, and start to have discussions with payers about what the appropriateness of testing is going to be,” said Quidel CEO Doug Bryant. "Everyone who's paying for it now, it's just out of pocket. They're not getting reimbursed."
Pay to play
Even if a wide array of tests become authorized for use in asymptomatic populations, the question of who pays for screening remains a major hurdle.
Some asymptomatic testing has been available through public health programs, such as the Project Baseline initiative from Verily Life Sciences LLC, but access to these tests for individuals without symptoms or recent exposure to a confirmed case has been variable (see “Bumpy Ride for Verily’s COVID-19 Screening Platform”).
The Families First Act, as amended by the CARES Act, requires private health insurers to cover testing to detect or diagnose COVID-19 without cost-sharing.
But in June, the Departments of HHS, Treasury, and Labor jointly issued a guidance that clarified that requirement did not extend to workplace screening programs, public health initiatives, or “any other purpose not primarily intended for individualized diagnosis or treatment of COVID-19 or another health condition.”
Language on private payers’ websites about COVID-19 testing overwhelmingly says coverage is limited to cases where an individual is symptomatic or has been exposed to a known case, and frequently cites CDC guidelines as the basis for its decisions, making the agency’s wavering on whether exposed individuals should be tested an acute issue.
“In some instances, simply saying you’ve been in contact with somebody who tested positive doesn't get you a test,” Bryant said.
Asymptomatic screening also faces steep hurdles with U.S. public payers.
By law, Medicare doesn’t cover screening tests unless they pass through a stringent evaluation process by the U.S. Preventative Services Task Force and are written into statute. In addition, CMS hasn’t issued any national coverage decisions for antigen tests, which would shoulder much of the burden for asymptomatic screening.
In the case of Medicaid, which is supported by both state and federal governments, the major issue is lack of funds, particularly as the economic crisis brought on by the pandemic has caused widespread state budget shortfalls.
AdvaMedDx executive director Susan Van Meter said the medical device trade organization is working in coalition with other stakeholders including laboratory associations, labor unions and business groups to advocate for legislative action from Congress that would clarify the CARES Act language to support coverage of symptomatic, pre-symptomatic and asymptomatic COVID-19 testing via a wide range of technologies.
The trade group has advocated for the allocation of dedicated federal funding to support coverage and reimbursement for testing that supports public health surveillance and the return to work and school, with support by private payers and Medicaid. Such changes could come through the next long-anticipated COVID-19 stimulus bill, or through standalone legislation.
According to Josh Bilenker, CEO of Loxo Oncology at Lilly, making testing more broadly available will also require increasing what healthcare systems are willing to pay for it.
“If we paid hundreds of dollars per molecular test with the condition that it would have the turnaround time needed to be no more than 48 hours — promise you, promise you the infrastructure would pop up overnight. We're in effect facing a reimbursement crisis in diagnostics,” he said (see “Why Reimbursement Crisis is Creating Innovation, Access Crises”).
In the absence of federal guidelines and funding for broad asymptomatic testing, philanthropic and private players have partnered with states and employers to implement strategies informed by data-driven models.
The Rockefeller Foundation has partnered with the governors of 10 states to form a buyer’s club for antigen tests, with the goal of supporting screening in high-priority areas such as schools (see “Bipartisan Constellation of States Bet on Antigen Tests”).
In its most recent report, the foundation and its collaborators at Duke-Margolis and ASU provided granular, model-based guidelines on the types of asymptomatic testing plans that could reduce COVID-19 transmission in schools and nursing homes (see Figure).
The report also outlines how variables such as community case counts influence what type of testing technology will give the most useful results. “It’s all about having the right test for right person, in right setting, at right time,” said Aspinal.
For example, a rapid test with 85% sensitivity and 97% specificity may be a good fit with communities with high prevalence of COVID-19, whereas in communities with low prevalence, such a test would primarily turn up false positives. In contrast, pooled tests are cost-saving in low prevalence settings, but inefficient when case counts go up (see “EUA for Pooled Test Points to Effect of Prevalance on Performance”).
Health plan administrator Collective Health Inc. is also turning to modeling to guide its workplace COVID-19 screening programs.
The company’s Collective Go program optimizes testing solutions for different types of employers — including ones with a distributed network of employees, such as babysitters and delivery drivers — based on factors such as case prevalence, ability of workers to socially distance, acceptable turnaround times and feasibility of setting up testing sites versus sending self-collected samples through the mail.
Sanjay Basu, Collective Health's director of research and population health, said the company uses a “Costco model” to bulk procure tests and “load-balance the demand” across a large network of small companies and labs with different testing technologies.
He said that among the factors used to pick the right test for a given workplace is its limit of detection (LoD), a metric of sensitivity. FDA recently published LoD data for tests with EUA that it generated using a harmonized reference panel, which provides a clearer picture of different tests’ relative performance than validation data provided by individual companies using different samples and reporting units (see “Harmonizing Limits of Detection”).
Basu said making widespread asymptomatic testing a reality is “as much a logistics, execution and implementation challenge” as a technological one.
“We’ve solved the laboratory biology part to a much greater extent than the logistical obstacles,” he said. “I’ve learned as much or more from my colleagues who work in logistics of supermarket chains and have advised us on how to optimize the supply chain, as from medical colleagues.”
Going beyond individual workplaces to broader-scale testing that supports societal reopening will require a higher level of governance, said Basu. “What it will take is instituting an effectively clear, consistent and facilitated federal-level solution, so that we no longer have these obstacles to providing people with rapid, frequent, high quality testing.”
Hamburg thinks building an effective national asymptomatic testing strategy would require coordination across multiple government agencies, including the CDC, NIH, FDA, CMS, DOD and VA. Such an initiative could in principle be taken on by a group like Operation Warp Speed, she said.