Long-term symptom profiles after COVID-19vsother acute respiratory infections: a population-based observational study (COVIDENCE UK)
Abstract
Background
Long COVID is a well recognised, if heterogeneous, entity. Acute respiratory infections (ARIs) due to other pathogens may cause long-term symptoms, but few studies compare post-acute sequelae between SARS-CoV-2 and other ARIs. We aimed to compare symptom profiles between people with previous SARS-CoV-2 infection, people with previous non-COVID-19 ARIs, and contemporaneous controls, and to identify clusters of long-term symptoms.
Methods
COVIDENCE UK is a prospective, population-based UK study of ARIs in adults. We analysed data on 16 potential long COVID symptoms and health-related quality of life (HRQoL), reported in January, 2021, by participants unvaccinated against SARS-CoV-2. We classified participants as having previous SARS-CoV-2 infection or previous non-COVID-19 ARI (≥4 weeks prior) or no reported ARI. We compared symptoms by infection status using logistic and fractional regression, and identified symptom clusters using latent class analysis (LCA).
Findings
We included 10,203 participants (1343 [13.2%] with SARS-CoV-2 infection, 472 [4.6%] with non-COVID-19 ARI). Both types of infection were associated with increased prevalence/severity of most symptoms and decreased HRQoL compared with no infection. Participants with SARS-CoV-2 infection had increased odds of taste/smell problems and hair loss compared with participants with non-COVID-19 ARIs. Separate LCA models identified three symptom severity groups for each infection type. In the most severe groups (including 23% of participants with SARS-CoV-2, and 21% with non-COVID-19 ARI), SARS-CoV-2 infection presented with a higher probability of memory problems, difficulty concentrating, hair loss, and taste/smell problems than non-COVID-19 ARI.
Interpretation
Both SARS-CoV-2 and non-COVID-19 ARIs are associated with a wide range of long-term symptoms. Research on post-acute sequelae of ARIs should extend from SARS-CoV-2 to include other pathogens.
Funding
Barts Charity.
Research in context
Evidence before this study
We searched PubMed and Google Scholar for studies on post-acute sequelae of COVID-19 and other acute respiratory infections (ARIs), published up to May 24, 2023. We used search terms relating to COVID-19 and other ARIs (“COVID-19”, “SARS”, “severe acute respiratory syndrome”, “Middle East respiratory”, “MERS”, “respiratory infection”, “influenza”, “flu”) and post-acute symptoms (“long COVID”, “post-acute”, “PACS”, “sequelae”, “long-term”). Previous studies have shown a wide range of post-acute sequelae for COVID-19, affecting people with all severities of the acute disease. The few studies that have compared long-term symptoms between people with COVID-19 and non-COVID-19 ARIs have generally found a higher symptom burden among people with COVID-19; however, these studies have been restricted to hospitalised patients or electronic health record data, and thus do not capture the full picture in the community. Research into long COVID phenotypes has been inconclusive, with some analyses classifying people with long COVID according to the types of symptoms experienced, and others classifying them according to the overall severity of their symptoms.
Added value of this study
In this population-based study of ARIs in the community, we observed high symptom burden among people with previous SARS-CoV-2 infection when compared with controls, highlighting the extensive reach of long COVID. Our finding of a similar symptom burden among people with non-COVID-19 ARIs suggests that post-acute sequelae of other ARIs may be going unrecognised, particularly given that the vast majority did not experience a severe acute infection. Latent class analyses of symptoms identified groupings based on overall symptom severity, rather than symptom types, for both SARS-CoV-2 infections and non-COVID-19 ARIs, suggesting that overall symptom burden may best characterise the experience of people with post-acute sequelae. Notably, among participants with the most severe symptoms, only half of those with previous SARS-CoV-2 infection attributed their symptoms to long COVID, suggesting they either did not believe the infection was the cause, or they did not consider their symptoms severe enough to qualify as long COVID.
Implications of all the available evidence
The long-term symptoms experienced by some people with previous ARIs, including SARS-CoV-2, highlights the need for improved understanding, diagnosis, and treatment of post-acute infection syndromes. As much-needed research into long COVID continues, we must take the opportunity to investigate and consider the post-acute burden of ARIs due to other pathogens.
Related articles
Related articles are currently not available for this article.