Death rates from acute COVID infections have fallen at this point in the pandemic, but there’s still a lot of worry about long COVID. And if you’ve been looking at recent headlines, you might assume that long COVID should very much be a concern for the average person today.
A new study is out in the prestigious journal Nature Medicine that appears, at first glance, to bolster that view. Headlines about it say it shows that long COVID is worse, on a population level, than heart disease or cancer, and that it can continue to cause issues two years after the initial infection.
Long COVID, which the World Health Organization defines as unexplained symptoms that are still a worry more than three months after the initial infection, is a public health concern. And yes, there is a chance that you will get COVID and have lingering symptoms. But what this study has to say about long COVID risk for most people is actually reassuring.
Here’s one way to summarize what the study shows: It compared people who had a record of a COVID-19 infection in 2020 with a control group, who had never been diagnosed with the disease, and found that years later people who’d had COVID-19 were still at an increased risk of a range of issues. Specifically, the researchers found, the risks of 25 out of 80 serious conditions were elevated even in people who had not been hospitalized with their initial COVID-19 infection when compared with an uninfected control. People who had been hospitalized saw even higher risks.
There are a few important things to keep in mind. First, this paper looked at what’s known as the U.S. Veterans Health Administration cohort. This is a group of generally older, mostly male patients who get health care from the VHA, and who are, on balance, much less healthy than your average person. Any issues seen in this cohort are very hard to generalize to the rest of the population. The paper also looked at people who were infected prior to the widespread availability of vaccines.
Further, to be included in the study, participants had to seek medical help, which means that any analysis on this data will be biased. The rate of hospitalization for COVID-19 was much higher than you’d see in the general population. By definition, people who had mild infections and didn’t need help from the health care system were less likely to be included in this cohort. This means that the paper’s estimates of post-infection issues are likely to be exaggerated. (On the other hand, the authors did not take into account potential unrecorded infections in the control group, which means that some people in this group may also have been experiencing long COVID symptoms.)
So, what does the study actually show? It shows that there is a large risk of long-term issues for people who have been hospitalized with COVID-19. This is entirely expected. If you get sick enough with any disease to the point where you require hospitalization, you are almost guaranteed some measure of long-term issues. Getting really sick is bad for your health.
For people in the study who weren’t hospitalized with their initial infection, the situation was quite different. There was a large increased risk of issues such as heart attacks and susceptibility to other respiratory disease that happened in the first three months after being infected. But after that, the long-term problems associated with COVID-19 dropped off rapidly. Two years after infection, the non-hospitalized people with COVID-19 were experiencing levels of various diseases that were virtually indistinguishable from the control.
To put the risk of long COVID in context, let’s look at a few examples of these risks from the non-hospitalized group. (A technical note: The authors used a metric called Disability-Adjusted Life Years, but for simplicity, I’ve converted that back into absolute counts.) Between 18 and 24 months after infection, about 18 in every 10,000 people in the control group developed diabetes. In the COVID-19 cohort, that figure was 20 for every 10,000 people. For fatigue—one of the most common issues associated with long COVID—the numbers were 26 in 10,000 for the control vs. 31 in 10,000 in the COVID-19 arm.
This study measured illness by looking at how many people received diagnostic codes—and to get a diagnostic code, you’d have to actually go to a doctor and get diagnosed, which people don’t always do. That means that these numbers are potentially lower than the true figures of these diseases in this cohort. At the same time, they are likely higher compared with the true figures for the rest of the population due to the nature of the people included in the study.
But these numbers do still give some context as to the rates of these issues in people infected with COVID-19. At a population level, long COVID represents an elevated risk that could be meaningful: An extra two cases of diabetes per 10,000 people over 60 would be quite a few people nationally. At an individual level, however, they are much less alarming. The chance that you will be one of—maybe—five people out of many thousands who have long-term fatigue two years after COVID-19 is quite low. It certainly isn’t as worrying as the health issues posed by heart disease or cancer.
Ultimately, this study does not contradict a broader picture of long COVID that is emerging. If you look at the population-wide rates of people reporting long-lasting issues associated with COVID-19, they have been on the decline for some time. This is likely because of vaccination and past infection. Immunity makes future infections much less severe, and at this point the vast majority of the world has several layers of immunity thanks in particular to our excellent COVID-19 vaccines.
None of this is to say that long COVID isn’t a nasty problem for many people. I personally know several people suffering tremendously years after infection. There are still tens of thousands of people likely disabled in the U.S. due to their coronavirus infections. For some of these people, the issues may be lifelong.
Long COVID is a serious issue, and we are rightly putting a fair bit of effort into researching it. That being said, I find the new evidence quite reassuring. As the severity of COVID-19 drops due to our ever-growing levels of immunity, the risks from the disease, including long COVID, drop as well. And for people who have relatively mild initial infections—which is, at this point, the vast majority of those who get the disease, especially if they’ve recently had a booster—the risks are generally very small. Although COVID-19 will remain a public health problem for years to come, and possibly forever, it is no longer an everyday fear for many people.
Update, Aug. 29, 2023: This piece has been updated to clarify that there are likely tens of thousands of people disabled due to COVID-19 in a country the size of the U.S., rather than globally. This figure is an estimate, extrapolated from a study done in the U.K. at a time of peak prevalence of inactivity due to long COVID.