We still don’t have a really good understanding of how many people get long covid, or why. And the data could be changing over time. But a study from the University of Southern California queried 8,000 people about the first year of the pandemic and came up with some possible answers.
“Long COVID and symptom trajectory in a representative sample of Americans in the frst year of the pandemic.” Scientific Reports, July 8, 2022, https://www.nature.com/articles/s41598-022-15727-0.pdf
The study noted that depending on the definition that is used, anywhere from 10% to 90% of people infected with the SARS2 virus go on to have something that can be called ‘long covid.’ In this study, long covid was defined by symptoms that started during the initial infection, and which persisted 4 months. Based on that definition, 23% of patients developed long covid.
Of course, in some people, the symptoms are going to be less severe, while in others, the condition will be debilitating. Deciding how to draw a line to separate the people who have a few minor issues from those who are dramatically impacted is a subjective game.
Perhaps surprisingly, there was only one pre-existing condition that predicted long covid at the time of infection: obesity. People who were significantly overweight when they caught the virus were 5.44 times more likely to develop symptoms of long covid compared to people who had a lower body mass index. Also curious: they did not find that gender affected the risk of long covid (this finding is in conflict with several other studies that found that women were more likely to have long covid than men). Age, race/ethnicity, smoking status, and pre-existing medical conditions did not seem to significantly influence the risk. Looking into the data, I notice that people with chronic lung disease and autoimmune conditions might have been at increased risk (1.83x and 3.05x higher risk than average) - but this did not hit statistical significance. This may point to a study that did not have enough participants in those categories.
Two other interesting points: 1) people who had hair loss in the acute phase of their infection were much more likely to get long covid (6.94x more likely), and 2) People with a severe sore throat in the acute phase were also more likely to go on to long covid (3.56x more likely). It’s worth noting that the Omicron BA.5 variant is that became dominant recently (summer of 2022) is somewhat more likely to involve a sore throat compared to previous variants, but we don’t yet know for sure if that will translate to a higher risk for Long Covid.
My Overall Thoughts
This study adds to the discussion, but should not get too much attention.
The study’s findings on gender might mean that several other studies are wrong - or this study may be off.
A poll or survey is an estimate. Questionnaires can be tricky things. First, the way that questions are asked can have a big effect on the results. Researchers try to avoid those types of issues, but it is not always easy. Second, the way that people are recruited for the study also can also have a big effect on the outcome. Again, pollsters try to think about these things and avoid them, but survey research involves art as well as science.