Alliance for Pandemic Preparedness
October 22, 2020
Is “No Test Is Better than a Bad Test”? Impact of Diagnostic Uncertainty in Mass Testing on the Spread of COVID-19
Category: Article Summary
Topic: Modeling and Prediction
Keywords (Tags): testing
- Gray et al. suggest that tests for SARS-CoV-2 with poor sensitivity and specificity could be potentially worse than not implementing tests at all in a model-based analysis. The model scenarios include no lockdown, immunity passports, and incremental relaxation. Tests, even if highly sensitive, are not sufficient to prevent widespread infection in the no lockdown scenario. Antibody testing, even with high sensitivity, is potentially risky in low-prevalence settings in the immunity passport scenario if it is poorly targeted. Tests must be paired with other strategies, such as effective contact tracing, in the relaxation scenario.
Gray et al. (Oct 21, 2020). Is “No Test Is Better than a Bad Test”? Impact of Diagnostic Uncertainty in Mass Testing on the Spread of COVID-19. PLOS ONE. https://doi.org/10.1371/journal.pone.0240775