Variation in global COVID-19 symptoms by geography and by chronic disease: A global survey using the COVID-19 Symptom Mapper

Balasundaram Kadirvelu, Gabriel Burcea, Jennifer K Quint, Céire Costelloe, A. Aldo FaisalEClinicalMedicine 2022 · DOI

TL;DR

This study reveals significant geographical and chronic disease-related variations in COVID-19 symptoms, challenging the simplistic "triad" often communicated. Understanding these differences is crucial for accurate diagnosis, risk assessment, and targeted treatment, especially as new variants emerge.

What Problem Does This Paper Address?

The prevailing public health messaging around COVID-19 often simplifies the illness to a triad of symptoms: cough, fever, and loss of taste/smell. However, anecdotal evidence and early observations suggested that symptom presentation could vary. This paper addresses the critical gap in understanding how COVID-19 symptoms differ across diverse geographical locations and among individuals with various underlying chronic health conditions. The authors highlight that a generalized understanding of symptoms might lead to underdiagnosis, delayed treatment, and potentially contribute to the spread of the virus, particularly among vulnerable populations with pre-existing conditions.

Methods

The researchers employed an exploratory study design using statistical and machine learning methods. They utilized a large-scale online survey, the "COVID-19 Symptom Mapper," to collect self-reported symptom data. Key analytical techniques included:

  • Descriptive Statistics and Visualization: Histograms and heatmaps were used to characterize the respondent cohort. Radar plots were employed to visually compare symptom profiles across different groups.
  • Comparative Analysis: Chi-squared tests were used to compare symptom proportions between groups.
  • Multivariable Logistic Regression: This was used to assess the association between symptoms and factors like age, gender, chronic disease, and country of residence, while controlling for confounding variables.
  • Hierarchical Clustering: This technique was applied to group countries and chronic disease profiles based on their symptom similarities, helping to identify patterns and relationships.

Datasets

The study utilized data from the COVID-19 Symptom Mapper survey. This dataset comprised responses from 175,566 individuals across 190 countries, collected between April 9, 2020, and September 22, 2020. For the primary analysis, the researchers focused on 78,299 responders who had either tested positive for COVID-19 (n=7980), tested negative (n=5620), or reported symptoms but had not been tested (n=64,699).

Key Insights and Ideas

  • Beyond the Triad: The study strongly supports the idea that COVID-19 symptoms are far more diverse than the commonly communicated triad of cough, fever, and loss of taste/smell.
  • Geography Matters: Symptom profiles show significant variations based on the country of residence, suggesting that geographical location, potentially influenced by cultural perceptions, environmental factors, or even early variant differences, plays a role.
  • Chronic Conditions as Modifiers: Underlying chronic diseases significantly alter the symptom presentation of COVID-19, making it crucial to consider a patient's health history for accurate assessment.
  • Clustering Reveals Patterns: Hierarchical clustering effectively grouped countries and chronic diseases by symptom similarity, providing a visual and quantitative understanding of these relationships.
  • Public Health Messaging Implications: The findings underscore the need for more nuanced public health communication about COVID-19 symptoms to avoid misinterpretation and ensure timely diagnosis.

Key Results

  • Geographical Differences: For example, India reported a lower proportion of headache and itchy eyes compared to other countries.
  • Chronic Disease Impact: COVID-19 positive responders with asthma were more likely to report shortness of breath compared to those without underlying chronic disease.
  • Country-Specific Symptoms: Brazil and Mexico showed higher rates of itchy eyes and headache. India and Pakistan reported fewer cases of nasal congestion and muscle ache.
  • UK Specifics: Responders from the UK reported a higher incidence of loss of smell and taste, muscle ache, shortness of breath, sputum, sore throat, chills, cough, fatigue, and headache compared to other countries.
  • Multivariable Analysis: After adjusting for age and gender, factors like asthma and hypertension were associated with higher odds of reporting shortness of breath. Similarly, being from the UK was associated with higher odds of shortness of breath, while Brazil and India were associated with lower odds.

Limitations and Future Work

  • Self-Reported Data: The study relies on self-reported data, which can be subject to recall bias and interpretation issues.
  • Sampling Bias: The use of a web-based survey means the sample may not be representative of the entire population, potentially skewing towards younger or more digitally connected individuals. The authors acknowledge that educational factors and socioeconomic status could also play a role.
  • Lack of Symptom Definition: Symptoms were not explicitly defined in the survey, which could lead to varied interpretations by respondents.
  • Test Type Variability: The study did not distinguish between different types of COVID-19 tests, which could affect accuracy.

Future Work: The authors suggest that future research should focus on symptom profile variations in emerging SARS-CoV-2 variants. This is crucial for accelerating clinical diagnosis, predicting prognostic outcomes, and targeting treatments more effectively.

Summary updated

Abstract

We acknowledge funding to AAF by a UKRI Turing AI Fellowship and to CEC by a personal NIHR Career Development Fellowship (grant number NIHR-2016-090-015). JKQ has received grants from The Health Foundation, MRC, GSK, Bayer, BI, Asthma UK-British Lung Foundation, IQVIA, Chiesi AZ, and Insmed. This work is supported by BREATHE - The Health Data Research Hub for Respiratory Health [MC_PC_19004]. BREATHE is funded through the UK Research and Innovation Industrial Strategy Challenge Fund and delivered through Health Data Research UK. Imperial College London is grateful for the support from the Northwest London NIHR Applied Research Collaboration. The views expressed in this publication are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care.
BibTeX
@inproceedings{kadirvelu2022variation,
  author = {Balasundaram Kadirvelu and Gabriel Burcea and Jennifer K Quint and Céire Costelloe and A. Aldo Faisal},
  title = {{Variation in global COVID-19 symptoms by geography and by chronic disease: A global survey using the COVID-19 Symptom Mapper}},
  booktitle = {EClinicalMedicine},
  year = {2022},
  doi = {10.1016/j.eclinm.2022.101317},
}
Figure