Long COVID: A narrative review and meta-analysis of individual symptom frequencies
Atchley-Challenner, R; Strasser, Z; Krishnamoorthy, A; et al., COVID, August 2024
Short Summary
This RECOVER study looked at how common 21 Long COVID symptoms were in adults before the Omicron wave of the COVID-19 pandemic. Researchers started by looking at 59 research papers. They then rated each paper based on information bias, which means how much the findings reported in the paper might have been affected by how the study was conducted or who participated in the study. Papers with less bias originated from studies that looked at groups of participants who represented the general population in a country, received survey responses from most participants, and included a comparison group of people who never had COVID.
Researchers found that the number of Long COVID symptoms people reported was different depending on how the study collected that information. For example, when people filled out surveys about their symptoms, about one in four people reported Long COVID symptoms like feeling very tired or having trouble breathing. But in studies using electronic health records (EHRs), fewer than one in thirteen people had Long COVID symptoms. This shows that many people may have had Long COVID symptoms that they did not report to their doctor.
This study is important because it shows that more people may have had Long COVID earlier in the pandemic than researchers thought. Future research should focus on grouping symptoms into clear patterns so doctors can better recognize Long COVID and choose the right treatment for people experiencing its symptoms.
This summary was prepared by the RECOVER Initiative.
Publication Details
DOI: 10.3390/covid4100106
Abstract
Objective: To summarize the prevalence of Long COVID symptoms among the general population reported in published articles from the pre-Omicron SARS-CoV-2 era. This narrative review examined 21 symptoms.
Methods: A PubMed/manual search returned 114 articles on general Long COVID symptoms. Manuscripts were excluded if they were not research studies, did not report symptom prevalence, or used a pediatric population. Ninety-eight studies were selected for review and fifty-nine met the criteria for inclusion. The risk of bias was assessed with the Hoy critical appraisal tool.
Results: After excluding studies with a high risk of bias, meta-analysis of prevalence for 21 symptom categories ranged from 2.6–28.7% in studies based on surveys to 0.3–7.1% in studies based on electronic health record data.
Conclusions: Long COVID symptom studies are limited by the variability in study design and representation of the general population. Further research is needed to effectively cluster symptoms in meaningful ways that enable focused treatment.
Authors
Rachel Atchley-Challenner, Zachary Strasser, Aparna Krishnamoorthy, Deepti Pant, Lori B Chibnik, Elizabeth W Karlson
Keywords
PASC; Long COVID; narrative review