Long COVID—and the increased risk of death, disability, and hospitalization it brings—can persist for two years, according to landmark study published on August 21 in Nature Medicine.

It’s the first study to look at a broad range of potential health effects stemming from the virus in the two years after infection. Most previous studies had only examined the initial year after infection, or a more narrow range of health effects in a period slightly longer than a year.

For those who contracted the virus in 2020 and were hospitalized with it, the risk of both death and hospitalization remained “significantly elevated” for two years, according to researchers with the U.S. Department of Veterans Affairs and Washington University.

Among those who contracted the virus the same year and weren’t hospitalized during their initial infection, the risk of death remained statistically significant for six months, researchers found. The risk of hospitalization remained elevated for about a year and a half.

“The findings highlight the substantial cumulative burden of health loss due to [long COVID] and call for attention to the care needs of people with long-term health effects” due to the virus, the study’s authors wrote.

A longer road for those who were hospitalized

Researchers examined the Department of Veterans Affairs medical records of nearly 140,000 individuals who survived COVID during 2020, as well as nearly 6 million who weren’t known to have contracted the virus that year. They followed them for two years to gauge their risk of death from all causes, as well as the incidence of 80 conditions known to be post-acute sequelae of COVID (PASC), frequently referred to as long COVID.

At the two-year mark, the risk of most of those health conditions—69%—was insignificant for those who hadn’t been hospitalized with the virus. But “substantial” risk remained “impacting several major organ systems.” The chance of developing blood, lung, gastrointestinal, or musculoskeletal conditions remained elevated, in addition to fatigue and diabetes, suggesting a longer-lasting risk for these ailments, the authors stated.

In a August 21 blog entry, Dr. Eric Topol, a professor of molecular medicine at Scripps Research and founder and director of the Scripps Research Translational Institute, wrote that the statistic was the “only reassuring finding for non-hospitalized people in the study.”

For those who had been hospitalized with the virus, the risk of most of conditions—65%, affecting all organ systems examined, and including cardiovascular, blood, endocrine, gastrointestinal, kidney, mental health, musculoskeletal, and pulmonary issues —remained significant for the entire two years. The findings are a nod to the “the difficult and protracted road to recovery among those whose disease was sufficiently severe to necessitate hospitalization during the acute phase of infection,” researchers wrote.

The study’s findings “show that while risks of many (but not all) post-acute sequelae decline and become non-statistically significant over time, the decline is less pronounced among those who were hospitalized in the acute phase of infection,” the authors wrote.

The good news for everyone: Researchers found no increased risk of cancer among those who had experienced COVID, hospitalized or not.

While significant, the study had its limitations, as the authors point out. All participants were veterans, and most were older males. Long COVID as a whole may look different—in both duration and symptoms—in a primarily female, or younger, population. As Topol pointed out, the study’s prototypic participant, a 61-year-old male, is far different from a female in her thirties—the demographic for which long COVID is thought to be most prevalent.

Further, all participants were infected during the first year of COVID, before variants like Delta and Omicron evolved. While thought to be less common, long COVID from later strains may feature important differences.

What’s more, those with a record of COVID infection were compared to those with no record of a COVID infection during the same year. But some of them may have had COVID without knowing, or telling their doctor, skewing rates of death, hospitalization, and disability in the COVID crowd for better or worse.

Long COVID’s long-term impacts on the immune system

Another new study, published Aug. 18 in the journal Cell, details the long-term immune system changes that severe COVID can trigger. The findings help elucidate why some with long COVID have symptoms tied to prolonged inflammation, like lung and kidney damage and neurological changes—and may have implications for anyone who has experienced the virus, regardless of severity.

Researchers with Weill Cornell Medicine in New York City and other institutions examined data from the blood of tens of patients—those who had recovered from severe COVID, and those who had recovered from other types of critical illness. In particular, they were able to isolate and analyze a rare type of stem cells found in blood—CD34+ hematopoietic stem and progenitor cells—thanks to a new technique they developed that made bone marrow biopsy unnecessary.

Among their findings: Monocytes—a type of white blood cell produced every few days from stem cells—showed changes in epigenetic programming up to a year after severe COVID infection.

Epigenetic programming refers to the epigenome—comprised of factors like chemicals, stress, diet, drugs, and disease that modify DNA, telling it “what to do, where to do it, and when to do it,” according to the U.S. National Institutes of Health. Those changes, to what could be casually referred to as DNA’s “packaging,” can be passed down from cell to cell as they divide, and from generation to generation.

Researchers also found that stem cells of those who had experienced severe COVID were more likely to allow activation of inflammation-associated genes. Such cells were also more likely to create a type of blood cell that serves as a “first responder” to infection.

Stem cells “can pass their epigenetic ‘memories’ on to their progeny immune cells, changing those cells’ inflammatory programs,” Dr. Steven Josefowicz, an associate professor of pathology and laboratory medicine at Weill Cornell Medicine, told Fortune. “So, when they see another pathogen, they respond in a different way than they would if they came from…cells that hadn’t seen inflammation to the same extent.”

Such changes to the immune system may persist longer than a year, Josefowicz said, adding that the study only lasted a year. And they may also occur—at least to some extent—in those who had more mild cases of COVID, though further study will be needed to tell.

What is long COVID?

With more than 200 symptoms identified—from lingering cough and fatigue to ear numbness and a sensation of “brain on fire”—long COVID is undoubtedly not one but multiple conditions, experts say.

True long COVID, some contend, is best defined as a chronic-fatigue-syndrome-like condition that develops after a COVID infection, similar to other post-viral syndromes that can occur after an infection with herpes, Lyme disease, and Ebola, among others.

Other post-COVID complications like organ damage should not be defined as long COVID, and better fit into the larger umbrella category of PASC, some experts say. Also known as post-acute sequelae of COVID-19, the term is used to encompass a wide variety of COVID consequences, from chronic-fatigue-like symptoms and subsequent heart disease to lasting lung damage and odd new symptoms like urinary incontinence, itching, and skin lesions.

As of Jan. 16, 15% of U.S. adults reported having long COVID symptoms at some point in the pandemic, and 6% reported lingering symptoms, according to a Jan. 26 report by the Kaiser Family Foundation, citing data from the U.S. Centers for Disease Control and Prevention.

The percent of Americans who’ve experienced COVID and still report long COVID symptoms dropped from 19% in June to 11% in January, according to the report.

Who’s most at risk for long COVID?

Factors like age, gender, BMI, and preexisting conditions may put individuals at higher risk for long COVID, according to a study published in March in the Journal of the American Medical Association Internal Medicine.

The U.K.-based study found that certain groups of people are at a significantly higher risk of developing the post-viral condition, thought to affect millions around the world. They include:

 Women

 Over 40

 People with obesity

 Smokers

 Those who were immunosuppressed before COVID

 People who were hospitalized with COVID

 People who had the following conditions before COVID:

 anxiety or depression

 diabetes

 asthma or COPD

Researchers examined the results of 41 published studies, with a combined total of more than 860,000 patients. They found that the aforementioned conditions were strongly associated with a higher risk of long COVID symptoms persisting three or more months after infection.

The results bolster the case that female gender and older age serve as risk factors for developing long COVID. A potential common thread among several risk categories: preexisting inflammation, which may extend the acute phase of COVID “even after recovery.” In the case of females, hormones might play a role in inflammatory status, while obesity shares a pro-inflammatory profile with long COVID, the authors write.

That’s not such great news for a giant swath of the population. There is good news, however: At least two doses of COVID vaccination seemed to lower the risk of developing long COVID, researchers found. Other studies have come to similar conclusions, they noted. They include a report from the U.K. Office of National Statistics, which found that those with two doses of COVID vaccine had a 42% lower risk of developing the potentially disabling condition.