Group testing performance evaluation for SARS-CoV-2 massive scale screening and testing.

Area of Science:

• Epidemiology
• Biostatistics
• Public Health

Background:

• Current molecular testing for COVID-19 lacks high-throughput capacity for community-level scans.
• Shallow contact tracing may miss silent transmission clusters crucial for disease spread.
• Group testing offers a feasible solution for resource-scarce settings with low disease prevalence.

Purpose of the Study:

• To evaluate the efficiency of sparse random pooling in group testing for COVID-19 surveillance.
• To determine the potential for scaling up testing capabilities in low prevalence scenarios.
• To compare sparse random pooling with conventional group testing methods.

Main Methods:

• Utilized sparse random pooling with conventional group test decoding algorithms.
• Performed simulations for both exact and inexact recovery scenarios.
• Investigated efficiency gains for very sparse prevalence regimes.

Main Results:

• Simulations demonstrated significant reductions in per-case test numbers.
• Achieved potential scale-up factors of 15X-20X compared to current COVID-19 group testing schemes.
• Identified further improvements possible in sparse prevalence scenarios for outbreak avoidance.

Conclusions:

• Sparse random pooling offers superior efficiency gains over conventional group testing and Reed-Solomon codes.
• Specialized designs can be developed for different scenarios to optimize testing.
• Significant scaling up of COVID-19 testing capabilities is achievable.