Predictors of chronic COVID-19 symptoms in a community-based cohort of adults

Abstract

Background: COVID-19 can cause some individuals to experience chronic symptoms. Rates and predictors of chronic COVID-19 symptoms are not fully elucidated.

Objective: To examine occurrence and patterns of post-acute sequelae of SARS-CoV2 infection (PASC) symptomatology and their relationship with demographics, acute COVID-19 symptoms and anti-SARS-CoV-2 IgG antibody responses.

Methods: A multi-stage observational study was performed of adults (≥18 years) from 5 US states. Participants completed two rounds of electronic surveys (May-July 2020; April-May 2021) and underwent testing to anti-SARS-CoV-2 nucleocapsid protein IgG antibody testing. Latent Class Analysis was used to identify clusters of chronic COVID-19 symptoms.

Results: Overall, 390 adults (median [25%ile, 75%ile] age: 42 [31, 54] years) with positive SARS-CoV-2 antibodies completed the follow-up survey; 92 (24.7%) had ≥1 chronic COVID-19 symptom, with 11-month median duration of persistent symptoms (range: 1-12 months). The most common chronic COVID-19 symptoms were fatigue (11.3%), change in smell (9.5%) or taste (5.6%), muscle or joint aches (5.4%) and weakness (4.6%). There were significantly higher proportions of ≥1 persistent COVID-19 symptom (31.5% vs. 18.6%; Chi-square, P = 0.004), and particularly fatigue (15.8% vs. 7.3%, P = 0.008) and headaches (5.4% vs. 1.0%, P = 0.011) in females compared to males. Chronic COVID-19 symptoms were also increased in individuals with ≥6 acute COVID-19 symptoms, Latent class analysis revealed 4 classes of symptoms. Latent class-1 (change of smell and taste) was associated with lower anti-SARS-CoV-2 antibody levels; class-2 and 3 (multiple chronic symptoms) were associated with higher anti-SARS-CoV-2 antibody levels and more severe acute COVID-19 infection.

Limitations: Ambulatory cohort with less severe acute disease.

Conclusion: Individuals with SARS-CoV-2 infection commonly experience chronic symptoms, most commonly fatigue, changes in smell or taste and muscle/joint aches. Female sex, severity of acute COVID-19 infection, and higher anti-SARS-CoV-2 IgG levels were associated with the highest risk of having chronic COVID-19 symptoms.

Fig 1. Cohort design.

Fig 2. Chronic COVID-19 symptoms.

Proportions of chronic COVID-19 symptoms were determined overall (A) and stratified by (A) sex, (B) age and (C) level of SARS-CoV-2 IgG antibody levels. Chi-square tests were performed comparing the frequency of chronic COVID-19 symptoms by sex, age and level of SARS-CoV-2 IgG antibody levels. P-values are presented in the graphs.

Fig 3. Duration of chronic COVID-19 symptoms.

Median duration of chronic COVID-19 symptoms were determined overall and stratified by individual symptoms. Mann-Whitney U tests were performed comparing the duration of chronic COVID-19 symptoms between those with vs. without a history of acute COVID-19 symptoms. P-values are presented in the graphs.

Fig 4. Latent class analysis of patterns of chronic COVID-19 symptoms.

Latent Class Analysis was used to examine patterns of binary variables of chronic SARS-CoV-2 symptoms in adults. LCA used the observed binary data to identify homogeneous patterns, i.e. n = 4 latent classes. Conditional probabilities were estimated using maximum likelihood to characterize the latent classes. (A) Conditional probability plots are presented, where probabilities closer to 0 or 1 indicate lower or higher chances, respectively. The horizontal black lines represent the overall distribution of SARS-CoV-2 symptoms. The black solid fill represents Class 1. The white solid fill represents Class 2. The gray solid fill represents Class 3. The diagonal black lines represent Class 4. The proportion of respondents who are members of these classes is presented. B. Chi-square tests were performed comparing (B) age, sex and SARS-CoV-2 IgG antibody levels with class membership.