COVID-19 is a slippery customer. A quick look through the comment section of any social media post with a photo of that dreaded positive test will tell you that everyone’s journey with this disease is different. One thing that’s become clear during this pandemic is that, in some unlucky people, the initial infection takes a long time to clear – sometimes over a month. Now, a new analysis from the University of Oxford suggests these persistent infections could be more common than we'd hoped.

The team of scientists used data from the UK’s Office For National Statistics COVID Infection Survey (ONS-CIS), which gathered information from over 90,000 participants in all. In the period November 2020 to August 2022, there were 3,603 participants who provided two or more positive samples that then underwent viral sequencing. This was the pool of data that the researchers were working with.

Of these over 3,000 people, 381 tested positive for an identical infection over a period of one month or longer. Fifty-four of these had a persistent infection that went on for at least two months. Extrapolating from these data, the researchers estimate that between 0.1 and 0.5 percent of all infections could persist for at least 60 days.

Some of these individuals remained infected with strains of the virus that had subsequently gone extinct in the wider population. Reinfection with the same strain after recovery, however, was very rare, likely due to immune memory and the fact that older variants start to be superseded by newer ones over time. 

The risk of prolonged COVID has previously been recognized in immunocompromised people, whose bodies can struggle to clear the initial infection. Some of these patients are thought to have been the source of the new variants that have emerged over the years we’ve been grappling with this pandemic, as the virus can build up mutations and genetic variation as it incubates. 

Certain individuals in this study did show high levels of mutations in their viral samples, including at sites that could affect the virus’s ability to evade prior immunity – genes that antibodies target, for example, or genes that encode parts of the all-important spike protein. 

However, the study authors say that these mutations were not evident in most of the cases they considered, so it’s not likely that any and all persistent infections could give rise to new variants of concern.

The idea of being constantly sick for two months is bad enough on its own, but it doesn’t stop there. People with persistent infections were also 55 percent more likely than those with typical disease to report having long COVID symptoms 12 weeks after the onset of infection. 

“Although the link between viral persistence and long COVID may not be causal, these results suggest persistent infections could be contributing to the pathophysiology of long COVID,” said co-lead author Dr Katrina Lythgoe in a statement. Other factors, such as inflammation and organ damage have also been implicated in long COVID, and scientists are still working hard to unpick this complex condition. 

Overall, this study provides an important insight into the prevalence of persistent COVID infections within the community, something that was a bit of a data blind spot until now. 

“Our observations highlight the continuing importance of community based genomic surveillance,” said co-lead author Dr Mahan Ghafari, “both to monitor the emergence and spread of new variants, but also to gain a fundamental understanding of the natural history and evolution of novel pathogens and their clinical implications for patients.”

The study is published in Nature.