Global disparities in SARS-CoV-2 genomic surveillance

Anderson F. Brito, Elizaveta Semenova, Gytis Dudas, Gabriel W. Hassler, Chaney C. Kalinich, Moritz U. G. Kraemer, Joses Ho, Houriiyah Tegally, George Githinji, Charles N. Agoti, Lucy Matkin, Charles Whittaker, Benjamin P. Howden, Vitali Sintchenko, Neta S. Zuckerman, Orna Mor, Heather M. Blankenship, Túlio de Oliveira, Raymond Tzer Pin Lin, Marilda Mendonça Siqueira, Paola Cristina Resende, Ana Tereza Ribeiro de Vasconcelos, Fernando Rosado Spilki, Renato Santana Aguiar, Ivailo Alexiev, Ivan Ivanov, Ivva Philipova, Christine V. F. Carrington, Nikita Sahadeo, Céline Gurry, Sebastian Maurer‐Stroh, Dhamari Naidoo, Karin J. von Eije, Mark D. Perkins, Maria D. Van Kerkhove, Sarah C. Hill, Éster Cerdeira Sabino, Oliver G. Pybus, Christopher Dye, Samir Bhatt, Seth Flaxman, Marc A. Suchard, Nathan D. Grubaugh, Guy Baele, Nuno R. FariamedRxiv 2021 · DOI

Abstract

Genomic sequencing provides critical information to track the evolution and spread of SARS-CoV-2, optimize molecular tests, treatments and vaccines, and guide public health responses. To investigate the spatiotemporal heterogeneity in the global SARS-CoV-2 genomic surveillance, we estimated the impact of sequencing intensity and turnaround times (TAT) on variant detection in 167 countries. Most countries submit genomes >21 days after sample collection, and 77% of low and middle income countries sequenced <0.5% of their cases. We found that sequencing at least 0.5% of the cases, with a TAT <21 days, could be a benchmark for SARS-CoV-2 genomic surveillance efforts. Socioeconomic inequalities substantially impact our ability to quickly detect SARS-CoV-2 variants, and undermine the global pandemic preparedness.

Introduction

Genomic surveillance serves as the primary "radar" for pandemic management, allowing the global scientific community to track viral evolution and identify Variants of Concern (VOCs) such as Delta and Omicron. However, this radar is not distributed equitably across the globe. When a new variant emerges in a region with sparse surveillance, it can circulate and spread internationally long before it is formally identified. This study maps the global landscape of SARS-CoV-2 genomic surveillance to identify the socioeconomic factors driving these disparities and to establish a reliable benchmark for early warning systems.

Methodology

The researchers conducted a comprehensive spatiotemporal analysis of genomic data from 167 countries between February 2020 and March 2021. The study focused on two primary performance metrics:

  1. Sequencing Intensity: The percentage of confirmed COVID-19 cases that are sequenced and shared publicly.
  2. Turnaround Time (TAT): The duration (in days) between sample collection and the public submission of the genome sequence.

To identify the drivers of surveillance performance, the team performed correlation analyses with 20 country-level socioeconomic and health quality covariates, including GDP, Research & Development (R&D) spending, and pre-existing healthcare indices. Furthermore, they utilized a "ground truth" dataset from Denmark—noted for its exceptionally robust surveillance system—to run simulations. These simulations modeled the probability of detecting rare lineages (0–5% prevalence) under varying combinations of sequencing intensity and TAT.

Data Sources

The analysis integrated several large-scale public and socioeconomic datasets:

  • GISAID (EpiCoV): Over 4 million consensus genome sequences.
  • NCBI: Over 1.7 million sequences and 1.2 million high-throughput datasets.
  • Socioeconomic Data: Sourced from the World Bank, United Nations, UNESCO, and the Institute for Health Metrics and Evaluation (IHME).
  • Danish National Metadata: Used as a high-resolution benchmark for simulation modeling.

Key Insights and Results

The Global Surveillance Gap

The study revealed a stark "Rich Get Data" effect. High-income countries (HICs) shared, on average, 16.5 times more sequences per reported case than low- and middle-income countries (LMICs). While case counts were often comparable between these groups during the study period, the data output was vastly different. Only 16 countries (9.6%) managed to sequence 5% or more of their cases, while 100 out of 167 countries sequenced less than 0.5%.

The 0.5% / 21-Day Benchmark

Through simulation, the authors identified a critical performance threshold: sequencing at least 0.5% of cases with a TAT of under 21 days. This benchmark provides approximately a 20% probability of detecting a lineage before it reaches 100 cases. While a 20% detection rate may seem low, it represents a realistic and achievable goal for many nations currently lacking any meaningful surveillance capacity. In contrast, a low-performance scenario (0.05% sequencing, 35-day TAT) common in many LMICs sees that probability plummet to just 2.6%.

Socioeconomic Drivers and Infrastructure

Sequencing intensity was most strongly correlated with per capita expenditure on R&D ($r^2 = 0.47$). Interestingly, a country's success was not solely dependent on current wealth; pre-existing infrastructure for influenza surveillance was a major predictor of performance. Countries with established systems for tracking the flu were able to pivot to SARS-CoV-2 more effectively than those starting from scratch.

The "Swamping" Effect

The data showed a moderate negative correlation between disease incidence and sequencing percentage. When cases spike, laboratory and bioinformatic capacities are often overwhelmed. This creates a dangerous paradox: the global "radar" is at its weakest exactly when the virus is spreading most aggressively.

Discussion and Limitations

The global average TAT during the study was 48 days. While this improved in 2021 following the emergence of the Alpha variant, many regions still struggle with delays that render genomic data "historical" rather than "actionable."

The authors note that these findings are correlational. While wealth and R&D spending are linked to better surveillance, the study does not account for specific logistical hurdles such as the cost of imported reagents, supply chain disruptions, or the "brain drain" of skilled bioinformaticians. Additionally, the study assumes a random sampling strategy. In practice, many countries use opportunistic sampling (e.g., focusing on travel-related cases). Future research is required to determine if targeted, non-random sampling could provide a more cost-effective way for resource-limited countries to detect dangerous mutations.

Conclusion

Socioeconomic inequalities fundamentally undermine global pandemic preparedness. By establishing a clear benchmark of 0.5% sequencing intensity and a 21-day turnaround, the international community can better target investments to close the "blind spots" in our global viral radar.

Summary updated

Abstract

Genomic sequencing provides critical information to track the evolution and spread of SARS-CoV-2, optimize molecular tests, treatments and vaccines, and guide public health responses. To investigate the spatiotemporal heterogeneity in the global SARS-CoV-2 genomic surveillance, we estimated the impact of sequencing intensity and turnaround times (TAT) on variant detection in 167 countries. Most countries submit genomes >21 days after sample collection, and 77% of low and middle income countries sequenced <0.5% of their cases. We found that sequencing at least 0.5% of the cases, with a TAT <21 days, could be a benchmark for SARS-CoV-2 genomic surveillance efforts. Socioeconomic inequalities substantially impact our ability to quickly detect SARS-CoV-2 variants, and undermine the global pandemic preparedness.
BibTeX
@misc{brito2021global,
  author = {Anderson F. Brito and Elizaveta Semenova and Gytis Dudas and Gabriel W. Hassler and Chaney C. Kalinich and Moritz U. G. Kraemer and Joses Ho and Houriiyah Tegally and George Githinji and Charles N. Agoti and Lucy Matkin and Charles Whittaker and Benjamin P. Howden and Vitali Sintchenko and Neta S. Zuckerman and Orna Mor and Heather M. Blankenship and Túlio de Oliveira and Raymond Tzer Pin Lin and Marilda Mendonça Siqueira and Paola Cristina Resende and Ana Tereza Ribeiro de Vasconcelos and Fernando Rosado Spilki and Renato Santana Aguiar and Ivailo Alexiev and Ivan Ivanov and Ivva Philipova and Christine V. F. Carrington and Nikita Sahadeo and Céline Gurry and Sebastian Maurer‐Stroh and Dhamari Naidoo and Karin J. von Eije and Mark D. Perkins and Maria D. Van Kerkhove and Sarah C. Hill and Éster Cerdeira Sabino and Oliver G. Pybus and Christopher Dye and Samir Bhatt and Seth Flaxman and Marc A. Suchard and Nathan D. Grubaugh and Guy Baele and Nuno R. Faria and Nathan D. Grubaugh and Guy Baele and Nuno R. Faria},
  title = {{Global disparities in SARS-CoV-2 genomic surveillance}},
  year = {2021},
  doi = {10.1101/2021.08.21.21262393},
  howpublished = {medRxiv},
}
Figure