SARS-CoV-2’s public health impact is worse than that of heart disease or cancer, study claims; others say the work may overestimate harm for the general population.

• 21 AUG 2023 BYCATHERINE OFFORD

Three-and-a-half years since SARS-CoV-2 spread around the world, scientists are still documenting the virus’ myriad effects on human health. What’s clear already is that those effects can continue long beyond the original infection.

Now, researchers have attempted to quantify this long-term harm using a massive database of U.S. veterans’ health records. They found a dramatically increased risk of dozens of conditions including heart failure and fatigue, sometimes years postinfection. Overall, the team estimates, COVID-19’s public health impact is more than 50% greater than that of cancer or heart disease.

Other researchers say the conclusions broadly reflect what physicians have seen. However, several cite drawbacks in the study’s statistical analysis that could have led it to overestimate harm to the general population. “The authors have done a good job in doing the analysis, but there are some limitations and those limitations are not small,” says Maarten van Smeden, a medical statistician at the University Medical Center Utrecht. “You have to take this with a little grain of salt.”

To assess COVID-19’s impact, Ziyad Al-Aly, a clinical epidemiologist at Washington University in St. Louis and the VA Saint Louis Health Care System, and colleagues analyzed data from the U.S. Department of Veterans Affairs. In previous studies of this data set, the same researchers identified an elevated risk of heart attack and mental health disorders up to a year after infection.

This time, the team looked at 80 health problems—from fatigue and other symptoms commonly associated with Long Covid to neurodegenerative disease—and general risk of death or hospitalization up to 2 years postinfection. They included data from about 140,000 people who tested positive for SARS-CoV-2 in 2020 and 6 million people with no record of infection that year.

In the 3 months postinfection, people who’d had COVID-19 had higher rates of death and many health conditions including heart failure, diabetes, Alzheimer’s disease, and depression. The differences between groups declined over time. Yet even among people who weren’t hospitalized, the risks for about one-third of the health problems studied remained elevated 2 years later, the researchers report today in Nature Medicine. These people had about a 13% increased risk of diabetes compared with the no-infection group, for example.

The figures were starker for people hospitalized with COVID-19: Two years postinfection, this group had elevated risks for about two-thirds of the outcomes studied. Compared with controls, they were about 50% likelier to suffer heart failure and more than twice as likely to receive an Alzheimer’s diagnosis.

“This is a rigorous study that matches what we’ve been hearing from [clinicians and patients] for years,” says Francesca Beaudoin, a clinical epidemiologist at the Brown University School of Public Health who was not involved in the work. Although COVID-19 is directly causing some of these health problems, she adds, it could exacerbate others or accelerate their onset.

Al-Aly’s team also translated its results into disability-adjusted life years (DALYs), a measure of healthy lifetime lost because of disease or disability. (The metric, although controversial because of assumptions it makes about how disability affects quality of life, is used in policymaking to quantify public health impact.) The team calculated that SARS-CoV-2 infection led to over 80 DALYs per 1000 people who weren’t hospitalized, and more than 640 DALYs per 1000 people who were. Cancer and heart disease each have DALYs of about 50, says Al-Aly, who consults for Gilead Sciences. “Eighty is an astronomically high number.”

However, some scientists say it’s hard to extrapolate from the study’s patient pool to the general population. People who caught SARS-CoV-2 in 2020 were unvaccinated and would have had earlier variants of the virus, not the ones circulating now. What’s more, the veteran population was about 90% male, 70% white, and had an average age over 60.

“[Older] men are some of the people most afflicted by acute COVID,” says Shawn Murphy, a neurologist and bioinformatician at Harvard Medical School and Massachusetts General Hospital. Although the authors tried to correct for this statistically, still, “We need to be careful about doing these extrapolations.”

There are other potential sources of bias, too. It’s unlikely everyone in the “uninfected” group—98% of people in the study—actually avoided infection all year, van Smeden says. Some probably had mild cases that didn’t require health care. If so, the “infected group” really means “people who had more severe infection,” which could magnify COVID-19’s lingering harms, he says.

The infected group may also have been on average less healthy than controls, says epidemiologist Justin Lessler of the University of North Carolina at Chapel Hill. He notes, for example, that hospitalized patients were five times likelier to be diagnosed with tobacco use disorders in the 3 months following infection. He doubts COVID-19 caused them to start smoking—some probably smoked before or would have started anyway. “What we may be seeing here is …  people with poor health who are having severe COVID and continue to have poor health afterward.”

Duke University statistician Fan Li adds that greater use of health care—a typical consequence of getting COVID-19—increases a person’s chances of being diagnosed with other conditions. This also skews apparent differences between infected people and controls, and isn’t adequately addressed in the study, she says.

Al-Aly argues these issues don’t substantially affect the study’s findings, adding the team controlled for as many differences between infected and uninfected people as possible. He says the elevated disease risks are “very likely the result” of SARS-CoV-2 infection and that, if anything, the work underestimated COVID-19’s effects.

Researchers did agree on the seriousness of the virus’ long-term harms, and the importance of studying them. The new paper highlights multiple conditions—including diabetes, neurodegenerative disorders, and muscle diseases—that merit further investigation in connection with SARS-CoV-2 infection, Murphy says.

Post-COVID-19 illness “is going to be a lingering effect for many years,” he adds. “We need to put enough work into [not only] finding treatments for it, but being able to respect the disability claims that people are going to have. Those claims are well founded.”