Fatigue, headache and muscle pain are among the most frequent symptoms of long COVID. Credit: Catherine McQueen/ Moment/ Getty Images.
Long COVID, a potential complication of COVID-19, can last for months or years, affecting various organs and reducing quality of life.
In a new study1, Manlio De Domenico, a physicist at the University of Padova, proposes a method for estimating how common long COVID is, and what effect vaccination has had on it, starting from publicly available data. The results say that vaccines can decrease the prevalence of long COVID among adults in the United States by nearly 21%, and by nearly 16% among all who had COVID-19 in 158 countries.
“My goal was to offer a flexible and relatively simple tool to public health officials to allocate resources for treatments and healthcare services,” says De Domenico, “and also to monitor how its prevalence changes as new variants emerge.”
A controlled study conducted in the United Kingdom estimated that long COVID prevalence was around 10% among people who caught the Delta variant, which emerged in June 2021, and 5% after an infection with Omicron, which appeared in November of the same year. But such controlled studies require a lot of resources and planning.
In contrast, national surveys provide data that are publicly available and continually updated, such as the Coronavirus Infection Survey in the UK or the Household Pulse Survey in the US. However, their results cannot be directly compared due to differences in the age groups. To overcome this issue, De Domenico devised a statistical technique that allows mapping the age groups of the UK survey against those of the US one, considering the demographic structure of the two populations.
The study found that in both countries the prevalence of long COVID among adults who had at least a COVID-19 infection varies with age, and that for all age groups the prevalence in the United States is larger than in the United Kingdom. Overall, long COVID is currently affecting about 1 in 10 children aged 0 to 12 and 1 in 5 people aged 13 to 59. The prevalence is at least twice as high between 60 and 89 years old.
To estimate the prevalence in the other 158 countries the author uses the results of a meta-analysis published last October.
This approach should be seen as complementary to controlled studies, to uncover patterns that can then be explored in more detail, says De Domenico.
The scientist also assessed the impact of vaccinations, looking at how the prevalence of the syndrome varies across US states with different vaccine coverage. The results suggest that in a population completely vaccinated, long COVID is nearly 21% less frequent than in an unvaccinated population. “The estimate is in good agreement with cohort studies conducted in the US and in Italy,” De Domenico says.
