People who endured even mild cases of covid-19 are at heightened risk two years later for lung problems, fatigue, diabetes and certain other health problems typical of long covid, according to a new study that casts fresh light on the virus’s true toll.

The analysis, published Monday in the journal Nature Medicine, is believed to be the first to document the extent to which an array of aftereffects that patients can develop — as part of the sometimes debilitating syndrome known as long covid — linger beyond the initial months or year after they survived a coronavirus infection.

According to the findings, patients who suffered bouts of covid severe enough to put them in the hospital are especially vulnerable to persistent health problems and death two years after they were first infected. But people with mild or moderate cases are not spared from the consequences when compared with those who never had covid, showing an elevated risk of two dozen medical conditions included in the analysis.

The study highlights the burden that continues to confront millions of people in the United States and the nation’s health-care system even though the federal government canceled the coronavirus public health emergency three months ago and the World Health Organization has declared the pandemic no longer a public health emergency of international concern.

“A lot of people think, ‘I got covid, I got over it and I’m fine,’ and it’s a nothingburger for them. But that’s not everything,” said the study’s senior author, Ziyad Al-Aly, a clinical epidemiologist at Washington University School of Medicine in St. Louis. After a couple of years, “maybe you’ve forgotten about the SARS-CoV-2 infection … but covid did not forget about you. It’s still wreaking havoc in your body,” said Al-Aly, chief of research at the Veterans Affairs St. Louis Health Care System.

Long covid remains a murky syndrome. Investigators participating in a growing body of research define it by different symptoms and different time frames, and some clinicians still do not always regard patients’ complaints as a serious phenomenon. According to Al-Aly, just two other known studies have used two-year time horizons, but they focused on a narrow group of symptoms, such as effects on the nervous system.

Estimates vary of how many people go on to suffer significant aftereffects. One analysis of nearly 5 million U.S. patients who had covid, based on a collaboration between The Washington Post and electronic health records company Epic, found that about 7 percent of those patients sought care for long-covid symptoms within six months of their acute illness. At the time, about 200 million people in the United States were known to have had covid, according to federal estimates, so that percentage translated into about 15 million with symptoms typical of long covid.

The new study is based on electronic medical records from VA databases of nearly 139,000 military veterans diagnosed with covid early in the pandemic, from March 2020 through the end of that year. They were compared with a group of nearly 6 million veterans not known to be infected with covid during that time. Both groups were tracked every six months to the two-year mark, looking at whether those who had been infected had higher rates of about 80 conditions typical of long covid. The study also looked at hospitalizations and deaths.

For the relatively small share of covid survivors who had been hospitalized, they had a heightened risk two years later of death, subsequent hospitalization and two-thirds of the medical conditions included in the analysis. Among those conditions: cardiovascular issues, blood clotting trouble, diabetes, gastrointestinal problems and kidney disorders. The survivors and the uninfected had started out in similar health, Al-Aly said, so the findings suggest the virus actually produced the heightened risk of lingering medical problems.

For the bulk of covid survivors in the study with milder cases, their long-term risks were less but not entirely gone. By six months after having tested positive, they were no more likely to die than people uninfected by covid. And their elevated risk had virtually disappeared by then for two-thirds of the conditions measured in the study, though they still displayed greater odds after two years of medical problems involving some organ systems, including cardiovascular and gastrointestinal trouble and blood clotting, along with diabetes, fatigue and lung issues.

Francesca Beaudoin, an emergency room physician and clinical epidemiologist who directs Brown University’s long-covid initiative, said the findings “capture what we are hearing at the narrative level from patients — that … the systems [affected after recovery from covid’s acute phase] are varied, that it results in loss of quality of life, loss of work and school.” Beaudoin said patients send her updates, reporting they still cannot walk one block without becoming worn out.

The study’s good news is that some people with milder covid cases do have fewer aftereffects over time, said Eric Topol, the director of the Scripps Research Translational Institute, who has been immersed in coronavirus research. But he said, “you don’t see a lot of optimism in these data. It’s basically a continuation of what we see at one year.”

The Nature Medicine paper noted that the covid survivors tracked in the analysis are not entirely representative of who is most likely to develop long covid. Because the patients in the study are veterans, the group is older than typical, and 9 out of 10 are men, while women account for more than half of long-covid patients in the general population.

Topol pointed out that because the study included only patients infected in 2020 — which allowed a two-year time span to follow them — they had the virus before coronavirus vaccines were widely available and before antiviral treatments such as Paxlovid had been developed. That also was a period before people tended to have built up immune defenses from one or more coronavirus infections.

“The whole landscape has evolved,” Topol said. Compared with people infected later in the pandemic, those in the study were “a defenseless population.”

Al-Aly said he and co-authors are working on a three-year analysis and plan to assess the same patients five years and a decade after they first developed covid.

“Obviously, we can’t predict the future,” said Akiko Iwasaki, an immunologist at the Yale School of Medicine who researches long covid. But she noted that omicron — a coronavirus variant that has spawned subvariants and dominated since late 2021 — is known to cause long covid. “We would expect some sort of parallel” with the study’s findings, she said. “It’s not a different virus, even though it’s a variant.”

Covid is not the only viral outbreak that has produced long-term aftereffects. Topol noted that people who survived the 1918 influenza pandemic had an elevated risk of developing Parkinson’s disease years later, while some people who had polio in the first half of the 20th century developed a constellation of symptoms known as post-polio syndrome decades afterward.

The study was funded by the Department of Veterans Affairs. It was not part of a $1.2 billion long-covid initiative of the National Institutes of Health called RECOVER, which patient advocates and some researchers wish had produced treatments in the two years since it began.

Slightly more than 1 million people in the United States have died of covid, according to the Centers for Disease Control and Prevention. The WHO reports more than 103 million confirmed cases in this country.