In 1928, Alexander Fleming famously returned from vacation to discover mold growing on a Petri dish. At the time, his peers largely disregarded his discovery of what would become the world’s first antibiotic, penicillin. 

More than a decade would pass before scientists would purify it, stabilize it for human use and secure proof of its efficacy. Ultimately, the timing was right and penicillin found its use case during World War II, when the global crisis created a need to scale and the British government funded drug production. 

Looking back over the first year of the coronavirus pandemic, it’s easy to point out everything that has gone wrong, and the magnitude of our loss is astounding. But the enormous scale of this problem — coupled with an infusion of capital and a global sense of urgency — has also created the conditions for existing innovations to find their penicillin moment, potentially defining healthcare in the decades to come.

Here are four areas where existing research and inventions found their use case over the past year. 

The power of public health data

During the earliest days of the pandemic, one thing was clear: We did not have the requisite data to make sound decisions. Without sufficient testing, we could not effectively quantify where the virus was spreading, hindering individual and community decisions that make or break public health outcomes. And once data started to flow, there was no central, trusted repository. What we had were spreadsheets. 

In early 2020, Robinson Meyer and Alexis Madrigal launched the COVID Tracking Project as part of their investigative work for The Atlantic. They quickly joined forces with Jeff Hammerbacher, a data scientist, who had started a similar spreadsheet. Until the COVID Tracking Project published its final update on March 7, 2021, it was the preferred COVID-19 testing and patient outcome data source for media outlets, research projects and public health dashboards in the U.S.

“Data sometimes seems so simple: Just collect it on where the pandemic is moving and spreading,” says Ryan Panchadsaram of Kleiner Perkins. “But to do it right you have to make it as local as possible.” 

This was the premise behind COVID Exit Strategy, a dashboard he founded in April 2020 to help state and local leaders, as well as individuals, make decisions. “The virus doesn’t care about borders and boundaries; it’s like the weather,” says Panchadsaram. “You should know what category hurricane you are in.” 

Having previously served as U.S. deputy chief technology officer in the Obama administration, Panchadsaram quickly recruited fellow public health and crisis experts to fill a gap in government information. Key collaborators and sources included the COVID Tracking Project as well as COVID Act Now. By December 2020, COVID Exit Strategy had merged with COVID Act Now, and Panchadsaram now serves on the board of COVID Act Now and is an adviser to the COVID Tracking Project. 

“COVID Act Now is the best way to get a local view of risk,” notes Panchadsaram. 

His hopes for a federal response were answered in December 2020 when the Department of Health and Human Services, Centers for Disease Control and Prevention, Office of the Assistant Secretary for Preparedness and Response and the U.S. Digital Service launched the COVID-19 Community Profile Report, a daily release of the federal government’s key COVID-19 indicators, as well as the HHS Protect Public Data Set, which augments HHS data with “datasets from academia, nonprofit organizations, industry, hospitals and facilities reporting.”

Dr. Kristen Honey, chief data scientist of the Department of Health and Human Services, was a driving force behind the public release of this data. The effort was guided by four principles: transparency, sharing, privacy and security. 

“If you want to make informed decisions, they have to be science-based and data-driven,” says Honey. “Without good data to know where the hotspots are, where PPE is, where medical supplies are needed and where vaccines are most needed, you cannot allocate your resources and optimize your response.” 

A longtime advocate of open data, Honey notes that our current crisis required a whole government response. She drew a parallel with the post-9/11 era, when numerous agencies came together to share information, transforming how national security offices collaborate today. 

Likewise, the pandemic was a catalyst for cross-agency sharing: “It’s like taking 10 years of data sharing and data standards and condensing it into 10 months,” explains Honey. Going forward, she anticipates that this investment will also pay dividends during future pandemics, or even during “peace time,” as we optimize for economic growth. 

For Honey, data is important for decision-making, but it has also proven that health is the foundation of the society. “Without wellness, we don’t have an economy. We don’t have education and school. This pandemic has shown us how interconnected everything is.

Translating data into action

While public health data was in short supply, scientific research was plentiful. Researchers have extensive knowledge of coronaviruses, and the COVID-19 genome was sequenced and published quickly, which helped us rapidly develop an understanding of its origin, spread and mutations. 

“The speed at which we’ve developed an epidemiological understanding COVID-19 has thus far been unprecedented — in no small part due to the advances we’ve made in data availability between SARS ‘Classic’ in 2002 and H1N1 in 2009,” says Dr. Maimuna Majumder, computational health informatics instructor at Boston Children’s Hospital and Harvard Medical School. Majumder’s work focuses on the epidemic dynamics of infectious diseases; for example, understanding how transmissible they are and what interventions work to curb transmission.

The widespread nature of this disease, while devastating, has contributed to our understanding of it. “The sheer pace at which we’ve learned about COVID-19 and its causative agent, SARS-CoV-2, has only been possible because of the pandemic — and more specifically, how universal the pandemic has been,” she says. More data points produce better models, which accelerate research. 

The virus doesn’t care about borders and boundaries; it’s like the weather. You should know what category hurricane you’re in.Ryan Panchadsaram, Kleiner Perkins

“Because this particular pandemic has far-reaching consequences around most of the globe, researchers have produced more scientific knowledge in a single year about this novel pathogen than we have during any previous epidemic,” adds Majumder. By summer 2020, her team noted that 20,000 research papers had been written about SARS-CoV-2 or COVID-19.

In addition to understanding spread, researchers are using data to inform medical decisions at scale. Trained as an ER doctor, Ziad Obermeyer, Blue Cross of California distinguished associate professor at University of California, Berkeley School of Public Health, came to know that humans often make mistakes. ER doctors often reflect on when they gave the wrong diagnosis or the wrong treatment, or wished they knew which patents would deteriorate quickly.

Today, Obermeyer’s work focuses on algorithms that help doctors make decisions better, as well as noting where algorithms can go wrong due to biases. This is particularly challenging in the context of an entirely new disease. 

“This is such a hard decision with COVID,” he says. “There are people you send home who show up two days later in respiratory distress needing a breathing tube because they suddenly got worse. And there are a lot of people you put in the hospital, and it turns out they didn’t need the bed, and after a few days they go home. That triage decision is incredibly hard.”

To make the job easier, Obermeyer and his colleagues are training an algorithm based on 500,000 X-rays from a large health system to answer the question: “Is this patient stable enough to send home, or should they stay in the hospital?” The prediction is based on subtle cues that humans wouldn’t catch. 

“That is not something a human radiologist does,” says Obermeyer. “A human radiologist says there is pneumonia or there is no pneumonia; it looks like COVID or it doesn’t look like COVID. The radiologist is not going to make a prediction about what is going to happen in five days.”

Obermeyer is quick to note that not all algorithms get it right. Stanford recently faced public backlash when its vaccine prioritization algorithm ranked older people safely working from home over younger residents who see patients daily. The problem, from Obermeyer’s perspective, wasn’t the algorithm. We were asking the algorithm the wrong questions.

In the Stanford case, vaccination priority was based on who was more likely to die from COVID-19. As a result, older people were vaccinated, even if they had a low risk of exposure. Meanwhile, younger people, including residents with high exposure, were deprioritized. In this situation, the algorithm should have balanced the risk of death with the preventative measures. 

“If you [are at risk of getting] infected, you need the vaccine, but not everyone has the same chances of getting infected,” explains Obermeyer.

This same thinking must also be applied to algorithms that determine how we allocate ICU beds and ventilators. Older patients placed on ventilators don’t do as well as younger patients. But younger people may have better outcomes, regardless, and older patients may have a greater need for intubation. How is outcome data weighted when developing algorithms for ventilator allocation? 

“The question is not how well you do on a vent, but does a vent make you better? An older person might get more mileage, whereas the younger person will be fine,” explains Obermeyer. He also found “enormous racial bias hiding in plain sight” when another algorithm was trained to identify “expensive” patients. Conflating costs with needs resulted in racial bias.

Despite these concerns, Obermeyer is optimistic. He envisions a future that embraces the power of data-driven decision-making without sacrificing humanity and ethics.

Remote everything

Remote care isn’t new, but the pandemic has dramatically accelerated adoption. Widespread shutdowns earlier this year prompted a sharp drop in in-person elective and preventative visits — and the decline has been partially offset by a significant increase in telehealth. This was due in part to regulatory shifts that fostered both awareness and adoption. 

According to Sukanya Lahiri Soderland, SVP, chief strategy officer at Blue Cross Blue Shield of Massachusetts (BCBSMA), her organization had a history of offering telehealth, “but the usage rates were very low until COVID hit.”

BCBSMA’s decision to pay for phone or virtual telehealth at parity with in-person visits and at no cost to members ushered in a new way to do business. “Between March and December of 2020, BCBSMA processed approximately 6 million new telehealth claims,” says Soderland. “During the height of the pandemic, we were receiving telehealth claims at a rate of nearly 40,000 per day.”

Determining whether in-person visits matter has now become part of a physician’s decision-making matrix. And rather than seeing it as a trade-off, it represents a new era of healthcare delivery. 

“While there will likely always be a need for traditional models of care, remote care introduces the ability to bring more care into the home and on the go, to use advanced technologies to monitor conditions, to help predict and prevent acute events before they happen, to expand access to care teams and expertise outside of one’s geography and ultimately, to substantially expand care delivery access points and supply,” says Soderland.

Of course, remote care is more than just telehealth. The same technology has applications for studies and clinical trials, according to Craig Lipset, founder of Clinical Innovation Partners and co-chair of The Decentralized Trials & Research Alliance (DTRA). Spring 2020 shut-
downs created enormous challenges for clinical trials; some stopped enrollment and others paused new starts. 

“The theme was trial continuity,” says Lipset, noting that all attention shifted to those already enrolled in clinical trials. “Can I get them their study drug? Can I monitor them for safety? Can I collect data around the efficacy of my medicine?”

What struck him was just how much technology already existed — it just wasn’t widely adopted prior to the pandemic. In addition to remote monitoring, studies and clinical trials could already be tapping existing solutions to safely deliver drugs to participants or clinical-trial specific visiting nurse capabilities. “This is development, not research,” he adds.

Another bright spot for Lipset: Remote clinical trial solutions also meet participants’ needs for flexibility, in a pandemic and beyond. “Survey after survey told us that people want this,” says Lipset. “My goal is to create options for people.”

“The theme was trial continuity. Can I get them their study drug? Can I monitor them for safety? Can I collect data around the efficacy of my medicine?”Craig Lipset, Clinical Innovation Partners

A new era for vaccines

Most experts agree: Messenger RNA (mRNA) is the most likely penicillin of our age. Like Fleming, researchers had devoted decades to research, but there was no proof that mRNA could be used for the benefit of human health — until now. 

“A pandemic sped that up,” says Dr. Michael Rosenblatt, chief medical officer of Flagship Pioneering, the investment firm behind Moderna.

In a now widely told story, Hungarian biochemist Katalin Karikó’s earliest work exploring how mRNA could combat viruses was met with rejection — including a demotion and loss of research funding. She spent a decade proving mRNA could be safely injected, which shaped the future of the two biotech startups behind the earliest COVID-19 vaccines, BioNTech and Moderna, and her subsequent team-up with fellow RNA researcher Drew Weissman paved the way further. 

Unlike traditional vaccines, which inject a weaker or inactive version of the virus into the body, mRNA vaccines teach the body to produce a protein that triggers an immune response. In recent years, both BioNTech and Moderna had started working on mRNA vaccines, but it was the pandemic that created a unique set of conditions for success. 

Moderna, which had previously received government funding to develop other mRNA vaccines, entered a new realm: In April 2020, it was awarded $483 million from the Biomedical Advanced Research and Development Authority (BARDA) to accelerate the development of a COVID-19 vaccine. By August, it landed a $1.25 billion government contract for 100 million doses of its vaccine. By January 2021, Moderna disclosed $11.7 billion in advance purchases for 2021. 

Rosenblatt anticipates this funding will bring an end to this pandemic while also paving the way for an entirely new class of therapeutics. This may include new vaccines, as well as non-vaccine applications that use an array of different proteins, antibodies, hormones and secreted factors.  

But don’t call this “drug discovery,” says Rosenblatt, who previously served as chief medical officer of Merck. Like all new novel therapies and therapeutic modalities, it is not a discovery — it’s an “invention.” 

He stresses the importance of investing in foundational science and the patience required to follow the winding path of invention. “Epidemics create a side effect of amnesia,” he notes, and people are eager to forget the struggle. But he hopes this crisis prompts the U.S. to finally harness invention, invest in vaccines and build its national stockpile.

Strengthening vaccine stockpiles will require rethinking not just the vaccines themselves, but also form factors and supply chains. Prior to the pandemic, Marc Koska, founder and head of product of Apiject, saw the glass vial shortage coming; we simply can’t produce them fast enough. 

His firm had been in discussions to evaluate how Apiject, a novel single-use plastic injector, could address this problem. Then the pandemic hit. 

“Although the door on the Strategic National Stockpile was opening, it was the pandemic that became the catalyst for … creating capacity that could react to the need for the national distribution and then into global, allied distribution,” says Koska. Apiject received a $590 million loan from the U.S. International Development Finance Corporation to build a pharmaceutical packaging factory.

Solving the stockpile problem was not initially on Koska’s list of priorities. His career had previously focused on making vaccination safer, starting with the invention of the self-locking K1 syringe to prevent needle reuse. Years later, he watched a honeybee and wondered, “Why don’t we give injections like insects?” A self-contained and pre-measured blister form factor could solve numerous barriers to mass vaccination. 

“It’s faster, it’s cheaper, it’s easier to build the factories, easier to transport, easier to administer, easier for disposal, and the energy consumption is about 90% less than making a glass syringe,” Koska explains. And unlike glass vials, the plastic injector supply chain can scale up quickly. 

The solution is so elegant and simple that it seems obvious. “Someone had worked it out before, and that someone happened to be nature,” adds Koska. 

Today, as glass vials are diverted to COVID-19 vaccine production, other vaccines may face supply chain setbacks. Apiject is preparing to address the logistics challenge for COVID-19 or other vaccines.