‘It’s noisy’: Competing Covid-19 efforts could hamper progress, experts warn

As many of the most forward-thinking tech and biopharma behemoths — from Apple and Google to Gilead and MIT — rush in to use their savvy and expertise to help fight Covid-19, some of their independent efforts risk undermining their common goals.

For all the know-how and good intentions of these institutions, responding to a global pandemic is far different than operating in the private sector. In interviews with STAT, several researchers and technology experts said that instead of collaborating and seeding innovation, some groups are effectively duplicating each other’s work or competing for limited resources — which could stymie progress in the pandemic response, the experts warned.

“It’s noisy,” said Eric Perakslis, a data science researcher and fellow at Duke University who led technology efforts for Ebola response programs in West Africa. “While these efforts are well-meaning, they do lower all boats in a way.”

There are multiple ongoing efforts to trace people infected with the virus and their contacts, clinical trials competing to enroll the same pool of patients, and even competition to collect cough samples from people with Covid-19. That dueling work is typical of how markets function in the private sector. But there’s nothing typical about responding to a pandemic.

“That approach doesn’t work as well when we’re in a situation like this — where time is of the essence, and lives are at stake, and the economy is at stake,” said Andrew Matzkin, a partner at the consulting firm Health Advances.

Matzkin said that while there’s a strong argument to be made in favor of more coordination and centralization among these efforts, there’s not the right infrastructure to make that possible.

“There is no Facebook page for tech companies that have something that might be useful for contract tracing. There’s no Slack channel where they can all agree: ‘Hey, why don’t we all work together on this,’” Matzkin said.

“That type of cross-institution, cross-industry collaboration takes time and is really challenging,” he added.

In digital contact tracing, tech giants Apple and Google have teamed up to help to log infected Covid-19 patients and their contacts; a research group at Massachusetts Institute of Technology has independently launched work on its own tracing system. Apple and Google’s work in contact tracing has been a rare example of collaboration for the normally fierce tech competitors. But it’s also, in some ways, very similar to MIT’s work.

Both tools use Bluetooth to flag when the phone of an infected person has come close to that of a healthy person, for example, and both tools involve apps. While the MIT group is encouraging people to download an app it’s developed called PrivateKit, the Apple-Google team has said it’s leaving it up to public health agencies to develop apps. If MIT also works with those agencies, and they share information with each other, then someone using one public health agency’s app could still be notified if they cross paths with an infected person who is using another agency’s app, a spokesperson from Apple said.

The Apple-Google and MIT projects are not expected to compete with each other in one key respect: that is, they won’t be divvying up users into two cordoned-off universes, in which, say, people with the PrivateKit app couldn’t be traced if they came in contact with a Covid-19 patient using an app supported by the Google-Apple system. Rather, each system is expected to be able to help trace people who are using the competitor’s technology.

Getting enough people to use contact tracing apps — and making sure those apps don’t divide users into siloed pools — is critical to making any tracking efforts effective. A recent University of Oxford study estimated that in the U.K., any successful digital contact tracing effort would need roughly 60% of the population to participate.

Experts said the MIT project also looks a lot like some existing digital contact tracing tools. One such tool, called CommCare, is currently in use by the National Institutes of Health, USAID, and the Bill and Melinda Gates Foundation. Perakslis and other researchers used the tool to do contact tracing in Sierra Leone in 2015 during the Ebola outbreak. Another, called Medic Mobile, works similarly.

“I wish some of these groups had done some due diligence and realized these tools already exist,” said Perakslis.

Studies of potential treatments could run into similar problems. Although a large number of Americans have been confirmed to have Covid-19, many clinical studies are only seeking enrollees who are hospitalized with severe disease — a much smaller pool of potential patients.

Gilead is seeking so many patients — at least 6,000 — for a late-stage study of the antiviral drug remdesivir that it threatens to complicate other clinical trials, experts said. That Gilead trial is enrolling hospitalized patients with severe Covid-19, some of whom are on ventilators. There is concern that the sheer size of Gilead’s trial might dissuade other pharmaceutical companies from conducting their own studies and that it could also usurp patients who might otherwise participate in other trials.

“I think Gilead will suck some people away from doing other studies,” said Brian Skorney, a senior biotech research analyst at investment bank Baird, adding, “Taking 6,000 patients is taking a meaningful number of patients.”

At least three projects are also clamoring for sound bites of people with Covid-19 coughing.

Cough for the Cure, Cough Against Covid, and CoughMode are all collecting audio from people who have tested positive for Covid-19, as well as from people who have not. The idea is to develop artificial intelligence tools that can identify unique features of a Covid-19 cough and distinguish them from the hacking sounds of someone with the seasonal flu or another respiratory illness.

Competing projects could be particularly detrimental because these efforts involve building machine learning models, which are more accurate and reliable when they can be trained and tested on as much data as possible. If there’s a finite number of people willing to upload their cough audio, divvying up that pool of samples could reduce the predictive power of each individual tool. (Though people could, in theory, submit their coughs with all of the initiatives.)

There can be positive aspects, too, to all the competition, some experts said.

In a crowded race to develop tech and therapeutic solutions, the best few products usually win out in the end, said Lisa Suennen, an investor who leads the firm Manatt’s venture capital fund and its digital and technology businesses.

Another advantage of competition: While a single, coordinated effort may lead to monopoly pricing, rival products can keep prices down, she said.

“But what’s the right number [of solutions]?” Suennen asked.

She, like the market, didn’t yet have an answer.