David Putrino, a rehabilitation and long Covid researcher at the Mount Sinai medical school in New York City, has watched mounting evidence that suggests between 4 and 36 percent of Americans infected with Covid have symptoms lasting at least six months.

It makes him nervous, and not just for the patients dealing with the long-term symptoms. He’s nervous because as the pandemic outbreak continues, many more people will be put at risk for long-term disability as the virus continues to spread.

Meanwhile, policymakers don’t seem to be accounting for the risk of long Covid in public health recommendations. The condition did not figure into the CDC’s guidance on removing or reinstating preventive measures — those were based on deaths and intensive care unit hospitalizations.

“I don’t think that policy has been based on fact,” Putrino said, regarding the lifting of preventive measures nationwide. “I think that policy is being based on hope.”

The potential stakes of long Covid are high. But there’s a key tension here: Much about the condition remains veiled in mystery. Basic questions — like what long Covid even is — remain unanswered. And those mysteries make it easier for policymakers to turn a blind eye to it.

“Even in the most basic ways, we can’t agree on a definition of this,” said Harlan Krumholz, a cardiologist and scientist at Yale University who collaborates with Survivor Corps, a long Covid research group. There is no consensus among scientists on what exactly long Covid is, which makes it difficult to count the people affected by it and to understand its risk factors.

Even among those making good-faith efforts to create balanced public health policies, that knowledge vacuum means many don’t take the threat of long Covid very seriously — a grave error, Putrino said.

If projections are correct, the long-term outcomes of Covid will leave hundreds of millions of people worldwide with chronic symptoms and many people with disabling conditions. And in the US, public health policy will have done very little to reduce long Covid’s impact. Despite that threat, the administrative burdens, scientific risks, and generally tricky nature of studying new diseases mean it will be some time before the most basic questions about long Covid are answered.

But they must be answered to help treat current long Covid patients, prevent new cases from forming, and guide policy makers.

I talked to several researchers about the most urgent long Covid questions, and why we don’t yet have their answers.

Exactly who has long Covid?

According to some estimates, long Covid affects between 5 and 25 million Americans. Worldwide, 30 to 50 percent who’ve been infected with SARS-CoV-2 have long Covid — roughly 200 million people in total.

That’s a pretty wide-ranging estimate. And these estimates vary so much because long Covid is so hard to define.

In October 2021, the World Health Organization released a clinical case definition of long Covid, an official set of criteria that public health professionals use to determine whether a person has a certain condition. According to the WHO, long Covid generally occurs three months from the onset of probable or confirmed Covid-19, lasting at least two months.

Friso Gentsch/picture alliance via Getty Images

As defined, long Covid can manifest in a lot of different ways. For some people, long Covid just looks like one protracted bout of illness following their initial infection, while other people’s symptoms start after what seems to be a recovery, and might fluctuate or relapse over time.

The WHO’s definition also allowed for great variability with regard to symptoms, which “include fatigue, shortness of breath, and cognitive dysfunction, but also others which generally have an impact on everyday functioning.”

As a consequence of this broad definition, people with persistent fatigue and brain fog are grouped with those suffering from persistent vomiting and diarrhea and those with rarer neurologic syndromes leading to strokes or paralysis.

It’s not necessarily a bad thing for a newly identified medical condition to be defined broadly. Early on in the process of scientific discovery, researchers would rather incorrectly include people than incorrectly exclude them. For example, before there was a test to identify the virus that caused the 2003 SARS outbreak, anyone with fever, cough, and travel to any of 11 countries would have met its case definition.

However, overly inclusive case definitions run the risk of conflating conditions that are actually different, making it hard to determine the true number of affected people and find effective treatment. To remain useful, case definitions must eventually narrow.

Keeping the case definition excessively broad also makes long Covid harder to treat. “Starting to differentiate subtypes of long Covid is a crucial next step, so that we don’t create blanket treatment recommendations or misunderstand people’s symptoms,” said Putrino.

Having a better idea of the true number of people with long Covid is critical for determining the amount of resources they need. Around a quarter of people with long Covid are unable to return to work, said Akiko Iwasaki, a Yale University immunologist who partners with Krumholz on long Covid research. Marshaling the resources to get these patients financial, medical, and social support requires first knowing how many of them there are.

So, what is long Covid? It’s likely that it’s not just one thing, said Krumholz. The wide range of symptoms and disability that now falls under the long Covid umbrella suggests multiple mechanisms are involved.

Continuing to talk about long Covid as one entity makes it harder to study its causes, said Iwasaki. “One of the things we have to do is to determine how many subsets of diseases are under this umbrella,” she said. “That’s important for designing clinical trials for treatment that’s appropriate for each driver of these diseases.”

What causes long Covid?

What are the biological explanations for long Covid? Although scientists have yet to determine what causes long Covid symptoms, researchers have some clues. Prior research on post-infectious chronic conditions — like myalgic encephalomyelitis (otherwise known as chronic fatigue syndrome) and post-Ebola syndrome — has helped identify several good leads.

Sebastian Gollnow/picture alliance via Getty Images

“We don’t really think that there’s all that much mystery about the concrete topics we should be studying,” said Amy Proal, a microbiologist who coordinates research at PolyBio, a nonprofit research consortium focused on complex chronic inflammatory conditions. Scientists in her lab and others are focused on exploring several leading hypotheses to explain persistent symptoms after initial Covid infection.

The first is that the virus hides in the body’s tissues long after the initial bout of Covid-19 has cleared, allowing persistent viral replication that triggers chronic inflammation. (A similar phenomenon has been identified in people with long-term symptoms after surviving Ebola infections.) Another possibility is that viruses other than SARS-CoV-2 are reactivated during the immune system upheaval that can accompany Covid infection, and are themselves wreaking havoc.

Next, there’s dysbiosis — an imbalance in the usual complement of microbes that inhabit human respiratory and digestive tracts — which could also be the basis of persistent symptoms. Alternately, autoimmunity, a disease state where the body creates antibodies that attack its own tissues, could be playing a role.

And finally, researchers are studying the link between long Covid and microclots, tiny, sticky blobs of blood cells that might play a role in preventing the body’s oxygen supply from meeting organs’ demands.

What’s the best test for long Covid?

Scientists would be better able to estimate the prevalence of long Covid if they could more reliably test for it. But such a test does not yet exist. Although a variety of biological features have been identified in groups of people with persistent symptoms — including microclots, MRI changes, viral reservoirs, and various immune system changes — there’s no one test that rules in or rules out the cause.

The lack of a diagnostic test illustrates the lack of answers to these other important questions. Finding a diagnostic test requires clarity on what the disease really is. Without knowing what they’re seeking, scientists can’t tell how good any tools are at finding it. What’s more, they might need more than one tool to find different types of long Covid, if it really is more than one thing.

They’re trying, anyway. As part of her research, Iwasaki tests a battery of immune system molecules in an attempt to identify patterns associated with certain symptoms in long Covid patients. She expects to publish the findings of these efforts soon.

Others conduct complex MRI studies or invasive tests of oxygen use during exercise in an effort to find tools for identifying disease and eventually, measuring its response to time and treatment.

While Proal thinks it’s unlikely researchers will ever identify a single molecule whose presence in the blood reliably predicts long Covid, it may eventually be possible to identify certain immune signatures or the presence of microclots among people with long Covid, using relatively simple blood tests.

What improves or prevents long Covid?

All this muddiness around what long Covid is makes it very challenging to treat, said Krumholz. “People are showing up in doctor’s offices, doctors are feeling inadequate because what people are presenting with doesn’t fit any pattern that they’ve ever seen before,” he said. “There’s no real knowledge out there — no textbook chapters.”

Both researchers and patients are deeply interested in finding out how to treat and prevent long Covid, and hypotheses about the condition’s causes point to potential solutions.

For example, a small, uncontrolled trial has suggested some promise for anti-microclot therapy in easing fatigue and other common symptoms.

Paxlovid, an antiviral drug, has been reported to help some individual patients. And small studies have reported that hyperbaric oxygen therapy and novel drugs have improved people’s symptoms.

On the prevention side, vaccines appear to reduce the risk of the most common long Covid symptoms among people who later get infected, by somewhere between one-half and two-thirds.

However, most of the published estimates of vaccine protection come from the pre-Omicron era, and it’s unclear how much that protection varies with different Covid variants and levels of vaccination (e.g. how many boosters a person has received). Ultimately, determining whether any preventive or therapeutic effects can be replicated in larger groups of patients will require a big investment of time and money.

What’s standing in the way of answering these questions? A mix of risk, cost, and complexity.

These questions remain unanswered in part because studying any new syndrome — especially one that emerges during a worldwide pandemic — is challenging. In the case of long Covid, it took months to even discern that the condition was emerging on a large scale.

Now, long Covid is better recognized. However, research roadblocks still stand in the way.

Ronny Hartmann/AFP via Getty Images

While recognition of long Covid has grown, high-quality data is still catching up. The early period of the pandemic yielded datasets describing both acute and longer-term Covid symptoms. But few contained enough detail about the characteristics and experiences of people with long Covid symptoms to permit much research, said Krumholz. That leaves researchers today without adequate data to help them answer the most pressing questions about the condition.

Research funding is still trickling in. It takes time and money to gather large groups of patients in order to systematically collect data over time, he said. Until recently, there wasn’t much of the latter. In late 2020, Congress agreed to provide the NIH with $1.15 billion to fund long Covid research, and in September 2021, the agency’s director announced $470 million of that would go to set up a national network of large-scale studies called RECOVER. However, institutions appear to have only been able to apply for funding from that network since early May.

Ongoing research is siloed, and prior research is underappreciated. While researchers worldwide are making important strides in understanding long Covid, even experts have little visibility into what other scientists are researching. Plenty of side conversations are taking place, but no big group chat, Iwasaki said. “There isn’t a single website where you can find out about different studies,” although there should be, she said.

And while research teams have been studying the chronic consequences of infection for years, they haven’t been convened in a way that would capitalize on their collective expertise, said Proal. Although it would make sense to have events that bring experts in these fields together, “there’s been an attitude of ‘let’s just start from scratch’ with long Covid,” she said.

A panic-neglect cycle has developed. On one hand, many people are suffering, and providers have few therapies to offer them. On the other, attracting researchers to a subject where the payoff is not guaranteed is a challenge, and one not unique to long Covid. (Researchers who study other post-infectious conditions like myalgic encephalomyelitis/chronic fatigue syndrome have routinely hit treatment roadblocks and funding problems, said Krumholz. Long Covid researchers may be taking note.)

It’s a risk to devote valuable resources and time to an area of uncertainty, especially when it has a fragile stream of funding and interest — from both scientists and the public, he said.

There are few doctors who specialize in long Covid. Doctors who see long Covid patients are in such demand that patients might wait six months for an appointment with one, said Krumholz. Almost nobody wants to become a long COVID specialist without treatments or diagnostics to offer people who are suffering, he said. “It’s a road to nowhere.”

The virtuous cycle of answering questions

Answering any number of these outstanding questions can make progress on another one. For example, if we can learn what causes long Covid, that would make it easier to design and fund treatment trials directed at those causes.

And there has been sufficient progress to start clinical trials that could help patients. “There are enough studies suggesting pretty clearly that at least some people don’t really clear this virus” to start treatment trials of antiviral agents for long Covid, said Proal.

But those studies are not yet happening. (Though in early May, a Pfizer spokesperson told Reuters the company was “monitoring data from ongoing clinical studies and real-world evidence, and may explore the issue further.” )

There’s another good reason to move more quickly on trials, Krumholz said; their results can reverse-engineer our understanding of long Covid’s causes. “If they seem to work, then that might help even accelerate the lab work even more because it’s pointing us in a certain direction.”

Although it may be professionally risky to study long Covid, some are earnestly joining the search for answers. To the question of “should I study this?” David Putrino’s answer has been “yes.” Prior to the pandemic, his lab focused on technology innovations to maximize physical rehabilitation and performance — now, he focuses mostly on long Covid research.

“This is not an area that I wanted to become an expert in,” he said, “but yeah, we are doing it because we see the need.”