What Would It Mean for Scientists to Listen to Patients? (original) (raw)

Soon after she got sick, Amy Siniscalchi felt as though her limbs were burning. “Then it morphed into, like, electrical buzzing,” she remembers. “At times, it would feel like my blood was gurgling, like seltzer in my veins.” Siniscalchi’s story follows a now familiar arc. In March, 2020, as a healthy forty-four-year-old, she was exposed to COVID, became ill, and never fully got better. Months later, she was still suffering from an array of strange symptoms: debilitating fatigue, memory problems, trouble walking, the seltzer sensation. She sought help from her primary-care doctor but left the encounter feeling dismissed.

“It wasn’t, like, ‘I’m so sorry for your suffering, I believe you, we just don’t know anything yet,’ ” Siniscalchi recalled. “There were a couple of doctors in the beginning that really just acted like nothing was wrong.” In search of answers, Siniscalchi went online. There, in Facebook groups and Slack channels, she found something more affirming than simple fellowship: thousands of others like her were organizing to make their suffering visible while sick and isolated at home.

Siniscalchi and I met over Zoom in May. Her face was framed with bright-red ringlets, and she had an air of eager, determined intensity. Prior to getting sick, she was a nonprofit administrator who worked fifty-to-sixty-hour weeks and exercised almost every day of the week. Since becoming ill, she has had to quit her job and, on many days, hasn’t had the energy to leave the house. In place of what she’s given up, her illness has become a kind of vocation. She spends much of her time online, connecting with others suffering from Long COVID and looking for information that might help. “When I lay down and I’m supposed to be resting, I’m scrolling studies,” she said. “And I don’t understand half of it, because I don’t have a science background, but I try to get it in context.”

Siniscalchi told me that she was hesitant to describe herself as a “citizen scientist,” but people like her—sick people who became lay experts in their own illnesses—were the first to name and characterize Long COVID. In April, 2020, a grassroots group, later called the Patient-Led Research Collaborative (P.L.R.C.), started surveying several hundred who remained sick for weeks after presumed COVID infections. Their results, published in a Google Doc, became the first systematic study of patients with Long COVID. Other groups also began to collect and disseminate data; in online forums, people with unresolved illness referred to themselves as “long haulers.” The academics Felicity Callard and Elisa Perego would later make the provocative argument that “Long COVID has a strong claim to be considered the first illness to be collectively made by patients finding one another through Twitter and other social media.”

In June of that year, the journalist Ed Yong described the P.L.R.C. survey results in an article for The Atlantic that brought “long-haulers” to mainstream attention. Conventional scientists and institutional authorities began taking notice; the Centers for Disease Control and Prevention reached out to the leaders of the P.L.R.C., asking to see their data. By the end of the pandemic’s first year, Congress had appropriated more than a billion dollars to the N.I.H. to study the condition, thanks in large part to patient-advocacy efforts.

During her first two years of illness, Siniscalchi signed up for a few institutional research efforts but found most of her experiences unsatisfying; either the studies felt like impersonal, disorganized bureaucracies, or shallow efforts to extract her data without engaging with her knowledge or perspective. Then, in September, 2022, she became a participant in an unusual clinical-research effort called the LISTEN study.

LISTEN (Listen to Immune, Symptom and Treatment Experiences Now) is an ambitious effort to understand Long COVID and similar conditions led by two renowned scientists at Yale: Harlan Krumholz, a cardiologist and clinical-outcomes researcher, and Akiko Iwasaki, an immunologist who studies virus-host interactions. Krumholz is a longtime champion of “patient-centered research”—the practice of designing and conducting a study in consultation with the people it intends to help. Iwasaki became famous beyond her field at the start of the pandemic through her active use of Twitter, on which she would explain complex scientific topics to a lay audience that had suddenly taken up virology as a hobby. Both of them appreciated that the first people to describe the symptoms of Long COVID were those afflicted with it, and recognized that the emerging patient movement would be a critical resource in studying the condition. Amid the uncertainty that characterized the early pandemic, both scientists recognized an opportunity to improve upon traditional research practices.

The project is based at Yale, but people can enroll online from anywhere in the world. The study’s extensive symptom surveys were created with patient input, and participants are invited to freewrite about their illness experience in online journals. They can also virtually consent to let the LISTEN investigators access their medical and pharmacy records. A signature feature of the study is its series of “town halls,” frequent forums in which Krumholz and Iwasaki meet virtually with any study participant who wants to attend.

Usually, investigators will identify a few patient representatives to sit on an advisory group. Krumholz and Iwasaki, in contrast, envision their study as a “safe space” designed to consider participants at every step of the discovery process. Theirs is an exercise in transparency that has few precedents in clinical science. Like local politicians showing up to regular community meetings, they open themselves to the good, bad, and out-of-left-field commentary of their entire constituency. Siniscalchi is an enthusiastic town-hall attendee. “They’re never dismissive,” she told me. “They can’t change course every time a patient says boo . . . but they at least make you feel heard, and they respond to you.”

If one legacy of the past four years has been what Iwasaki calls “the pandemic after the pandemic” of Long COVID, a related phenomenon has been an acceleration toward the type of research collaboration that LISTEN exemplifies: a democratic mode of scientific inquiry, with all the aspirational ideals and practical turmoil that a shift toward democracy implies. The LISTEN model makes explicit the politics of science: how research agendas are set, how results are interpreted, what counts as evidence. Compared with the conventional way of doing things, its path to discovery is both friendlier to its subjects, and more fraught.

One of _LISTEN_’s main objectives is to create a taxonomy of chronic-illness presentations that follow likely COVID infection. This involves sorting out how patients with different symptoms—respiratory illness, brain fog, neuropathies—cluster along multiple axes. In addition to looking for patterns among patients’ symptoms, demographics, and infection time lines, the study collects blood and saliva samples from a subset of participants. Iwasaki’s team analyzes these specimens for “immune signatures” that correlate with various illness presentations: Do people who have lost their sense of smell have specific T-cell responses? Or do people with “internal tremor” sensations tend to have elevated levels of certain biomarkers?

The enduring murkiness about what Long COVID actually is reflects something about its unusual ontology. Currently, it is an inclusive diagnosis organized around self-report: If you feel you have symptoms weeks or months after a likely COVID infection, you may have Long COVID. This capaciousness has made for effective coalition-building among the sick, who have consistently advocated that no sufferer be dismissed, but it poses a challenge for researchers and clinicians, who usually require clear metrics to delineate between who does and does not have a disease.

These competing priorities between patients and scientists can escalate into frank tension. Last year, investigators from the RECOVER Initiative, a government-sponsored Long COVID research effort, proposed a carefully modelled framework that identified twelve symptoms frequently associated with the disease. They published their results in the Journal of the American Medical Association, in an article filled with caveats about how this work was only the first step toward deriving a more precise Long COVID definition. But some Long COVID patient-advocacy groups swiftly denounced the study, lamenting that the framework’s scoring cutoff would leave thousands of sick people without a diagnosis. One such group called the RECOVER criteria “regressive and blunt.” More than forty-two thousand people have sent letters to the N.I.H. demanding that it retract the study and its scoring system.

Krumholz and Iwasaki’s sincere, and canny, approach to working with a highly engaged patient population is to honor everyone’s experience, no matter how controversial. They’re also investigating the link between Long COVID and other inflammatory conditions. Shortly after the study launched, the researchers began receiving messages from people who report long-term adverse effects that they attribute to COVID vaccination. Many of these patients—who variously refer to their condition as “vaccine injury,” “post-vaccine syndrome,” or “Long Vax”—describe symptoms similar to those reported by Long COVID sufferers: brain fog, dizziness, fatigue, pain. Intrigued by these similarities, Krumholz and Iwasaki decided to expand _LISTEN_’s scope.

People with post-vaccine syndrome do not identify with anti-vaxxers, but do believe that vaccine adverse effects have been undersold. As a result, these patients live in a culturally fraught zone where they feel isolated and stigmatized. When I asked Iwasaki about her decision to embrace this community and their concerns, she joked that she was grateful to have tenure so she could follow her instincts. “It’s such a tricky topic to tackle that most people shy away from this,” she said.

Both Long COVID and post-vaccine syndrome fall into a category that Abigail Dumes, a medical anthropologist at the University of Michigan, calls contested illness—disorders that present with more symptoms, like exhaustion, than signs, like abnormal exam findings or lab results. “Contested” is not a euphemism for fake or psychosomatic; it’s a term to describe conditions that have controversy surrounding their causes, diagnostic criteria, and treatment approaches.

Patients with contested illnesses—others include post-viral conditions like myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) and chronic Lyme disease—classically report feeling dismissed and told that their illness is “all in your head.” “There’s an unfathomable amount of medical gaslighting and medical trauma that happens in particular with post-viral patients,” Hannah Davis, one of the P.L.R.C.’s founders, told me. (Instead of “contested illness,” Davis prefers the term “illnesses without a biomarker.”) Siniscalchi also told me she felt “gaslit” by her doctor when she first sought help. “Everybody just calls it anxiety and calls it depression, but you know your body—you know that it’s not supposed to buzz internally,” she said.

Every LISTEN participant I spoke with shared a story that was as much about their traumatic interactions with the health-care system as it was about their physical ailments. Eventually, some found clinicians who offered them some treatment, but no one I spoke to had tried anything that led to full remission. Siniscalchi sees one doctor who, she said, “keeps sort of throwing spaghetti against the wall. He reads all the studies. So he is always, like, ‘Oh, let’s try this, let’s try that,’ which I’m grateful for.” But the supplements, she said, are getting expensive. “Nothing’s really working.”

I’m a physician. In my practice, I am generally against flinging spaghetti. When there aren’t randomized control trials or even large-scale observational trials to guide treatment, I tend to feel I’m working in an evidence-poor zone, and proceed conservatively. Many post-viral patients, however, might dispute my view of the literature in this area and consider my approach withholding. There are, after all, thousands of studies published about novel interventions for ME/CFS, chronic Lyme, and Long COVID. The distinction between my approach and a try-anything-that-might-work approach revolves around fundamentally subjective matters: what types of evidence one considers reliable enough to act on, how one interprets “first do no harm,” and how one calibrates risks and benefits.

Krumholz and Iwasaki do not characterize LISTEN as an attempt to bridge a contested-illness divide, but that is, in fact, one aspect of what they are doing, and it’s part of what makes the project both promising and thorny. During a recent LISTEN town hall, Iwasaki briefly mentioned micro-clots—tiny circulating blood clots that have been proposed as a potential disease mechanism, in post-viral illnesses—as a promising lead. Some proponents of this theory recommend treating patients with triple-anticoagulant therapy, an intervention that comes with the risk of uncontrolled bleeding.

During the Q. & A. section, one patient asked to return to the topic. Krumholz impassively read the participant’s question out loud, careful not to betray his own thoughts: “I’ve been tested and treated for microclots. Is this something that’s going to be more readily available in the United States? I’ve been paying out of pocket to see a doctor. What do you think about microclots?”

Iwasaki paused. “Right now, I don’t want to make any claims that micro-clots are causing the disease,” she began cautiously. She walked through the challenges—developing a standardized way to detect and measure micro-clots, being able to compare the levels in sick people with the levels in controls, and then making sense of the significance of any findings—but carefully avoided dispensing specific medical advice.

When we spoke about the micro-clot exchange later, Iwasaki admitted that she finds moments like that one challenging. She’s not a physician, and prior to COVID, most of her work had been with animal models, not people. The LISTEN team reminds town-hall attendees that they cannot give medical advice. “Until we have the final data, I really hesitate to comment one way or the other,” she told me. “But,” she acknowledged, “they also want my input.”

This unusual level of intimacy gives the investigators a sense of how high the stakes are for their study participants. “I think one of the things that is most concerning is if our best results are invalidating some things that people believe,” Krumholz said. “So it’s up to us not to be too dogmatic in our research . . . but, on the other hand, we can’t be afraid to tell them what we found, and to engage constructively knowing that some people may find a given result inconvenient.” It would be most convenient if the study’s revelations affirmed the stories patients have come to rely on to make sense of their suffering. But if the results prove surprising, Krumholz and Iwasaki will have to hope that their constituents trust them enough to change their minds.

At the end of last year, Krumholz and Iwasaki’s first clinical trial—Pax LC, a Phase II randomized control trial to study the therapeutic effects of the antiviral medication Paxlovid for Long COVID sufferers—began enrolling participants. The trial is decentralized, and creating it essentially required the investigators to build new clinical-research infrastructure from scratch. Krumholz managed to obtain ethics approval from Yale for a study that was only nominally based there, avoiding the regulatory hurdles that usually delay research that’s conducted across multiple states. People can enroll online from anywhere in the contiguous United States without needing to be screened or referred by a doctor, and almost all study interventions, including blood draws and medications, are delivered to patients at home. Krumholz also took questions from potential participants before the study began enrolling, and has promised to share results with participants and hear their feedback before publishing.

The trial design represents the LISTEN investigators’ most concrete achievement to date. Both Krumholz and Iwasaki believe that patient-centered trials don’t just make for a nicer way to do business but are a necessary improvement over conventional practice. Most clinical research is based at academic medical centers situated in major cities. Sick participants are asked to lend significant time and effort, or to bear physical discomfort or economic cost, without any guarantee of benefit to themselves, and, if they cannot meet the requirements of strict protocols, they are often forced to drop out. Those who are bed-bound or live in “clinical-trial deserts” are especially vulnerable to attrition. Patients enrolled in the Pax LC trial bear no economic costs, and are repeatedly and profusely thanked for their time.

Robert Harrington, a cardiologist and clinical-trials expert who is now the dean of Weill Cornell Medicine, told me that Krumholz and Iwasaki’s model represents a potential leap forward for the American research enterprise, which often struggles to engage the same people it wants to help. The LISTEN investigators, he said, are “taking a lot of ideas that have been done in pieces, and knitting them together in a way that is very innovative.” But he did wonder whether their highly personalized tactics would be easily replicable for other scientists. Most patient communities aren’t as organized and as eager to participate in research as the Long COVID constituency is, and most researchers don’t have the LISTEN investigators’ star power: Krumholz is described by others in his field as a “visionary,” and Iwasaki, an unusually telegenic lab scientist, is also uncommonly beloved by laypeople. (Siniscalchi, for instance, told me that Iwasaki’s involvement was one of the reasons she signed up for LISTEN in the first place.) “We’re not trained to do what they’re doing,” Harrington said.

There are other reasons to wonder whether the LISTEN model can work at scale. The study recruits from people who find the project on their own—in research parlance, a “convenience sample”—and its population of some three thousand is currently around seventy-five per cent female and eighty-five per cent white. N.I.H.’s RECOVER, in contrast, has enrolled more than twenty-nine thousand, including children and pregnant people, and has pursued nontraditional recruitment strategies to reach a diverse group of patients, including people who may not speak English, who aren’t as active online, or who have jobs that make it difficult to participate in scientific studies. If LISTEN operates like a highly functional community-board meeting, RECOVER is attempting to be something like the U.S. Census.

But Krumholz thinks of his work with Iwasaki as a demonstration project, showing proof of concept that research can be done differently. “It’s like we’re at Kitty Hawk,” he told me earnestly. “The thing’s flying, and I can see fleets of planes soon. This can work.” He continued, “In the old way of doing things, where actually I don’t look into their eyes, I don’t know who they are—they’re basically numbers on a sheet—and I’m just trying to dispassionately find a result . . . that’s a whole heck of a lot easier than when I’ve got an emotional connection with an entire community.”

When Krumholz was a young resident in San Francisco during the AIDS crisis, Anthony Fauci was managing the epidemic as director of the National Institute of Allergy and Infectious Diseases (NIAID). Fauci recalled attending a forum at the Lesbian and Gay Community Services Center (as it was then called) in the West Village in order to meet with AIDS patients and their community. This was a town hall where the audience couldn’t be muted and hidden from view; the relationship at the time between AIDS patients and the government was frankly antagonistic. Fauci remembers being confronted by “about a hundred very angry, hurting, frustrated and frightened young gay men.”

Ultimately, the friction yielded to progress, including both therapeutic breakthroughs and changes to the clinical-trials process that created roles for patients. When we spoke, Fauci called his collaboration with AIDS activists one of the most important things that he accomplished in his career (along with helping to develop effective AIDS therapies). But he admitted that the collaboration wasn’t summoned from thin air. “That really is a long, drawn-out process of the ultimate evolution of mutual trust,” he said. Gregg Gonsalves, an associate professor of epidemiology at Yale who, in the nineties, met with Fauci regularly as part of ACT UP and Treatment Action Group, spoke more bluntly: “You know, shit got real—there were shouting matches. I mean, it was not fun.”

Gonsalves and I talked about how a sense of urgency both drove the AIDS movement and, in some cases, led it astray. “In the early days of ACT UP, there was some idea that, you know, there’s a big velvet curtain at the F.D.A., and that behind it there were hundreds of drugs waiting for us to sort of try one by one,” he said. But, he said, “the initial assumption that [any] new drug was better than no drug at all, or that new drug is better than old drug, is not a way to save your life.”

H.I.V. and _SARS_-CoV-2 are very different viruses, and AIDS and Long COVID are very different illnesses. Nonetheless, groups like the P.L.R.C. are explicitly modelled on the citizen-scientist and patient-activist movements that have come before them, and the AIDS movement stands as the ur-example of a movement that changed not just the experience of patients, but the politics of knowledge production. Like ACT UP, contemporary patient advocates work at the intersection of medicine and activism; they petition and protest the same leaders and institutions that they depend on for treatment. The AIDS movement’s successes provide the benchmark for what can be accomplished in relatively short order when patients apply pressure and scientists yield in response. But that movement’s tensions—questions, for example, about urgency versus methodology—also hang over this one.

Only ten months after _LISTEN_’s launch, Krumholz felt moved to begin a LISTEN town hall with an apology. “I know all of you, probably, are as frustrated as we are, probably more frustrated that progress has been so slow,” he said. “We would like this study to go faster. . . . So let’s think together what it would take to make that happen.” He was reflecting the concerns of his patient-constituency, which has frequently expressed that Long COVID research has been slow to yield results and that therapeutic trials for post-viral conditions have been too long delayed.

Krumholz expects first results from the Paxlovid trial in June, and _LISTEN_’s first basic descriptive papers from the observational cohort are already available in preprint. “To manage your expectations, these aren’t going to be Nobel Prize-winning contributions,” Krumholz cautioned study participants in the summer. The analyses describe the demographics and self-perceived health burden of people with internal tremors, like the “seltzer” sensation Siniscalchi described, and the demographics and symptoms of people with “post-vaccination syndrome.” Both papers come to the modest conclusions that the patients surveyed describe themselves as very ill, and that more research, as ever, is needed.

Outside of her work with LISTEN, Iwasaki has already published extensively on Long _COVID_’s potential disease mechanisms. Krumholz told me that we are still far from the kind of breakthrough that will impact people’s daily lives. “It’s almost like everyone is overfitting or over-indexing on, like, one finding that they’ve got,” he said. He meant that, with clinical syndromes as complex as Long COVID or post-vaccine syndrome, it’s easy to misdiagnose a lot of noise as signal.

Iwasaki admitted that working in direct collaboration with her subjects, who are eager for solutions, can be emotional. “There’s a lot of anger and anxiety and frustration,” she said. “There’s always that disconnect between what patients expect and what we can deliver in a timely manner.” But, she continued, “I would rather be criticized for being late than put out something that’s half-baked, whereas I think patients, because of their suffering—they’re really expecting something to happen much quicker.”

When I asked Gonsalves about what Krumholz and Iwasaki are doing with LISTEN, he was admiring, but candid. “Their motivation and the desire to do it is commendable, but it’s gonna be messy as hell,” he said. I laughed, and when I repeated the quote to Krumholz, he laughed, too. ♦