The role of testing in Covid-19

Stanca Ciupe

Virginia Tech, USA

IMAG/MSM Working Group

November 4, 2021

Can we quantify the amount/type/frequency of tests needed by one community?

Credit: VDH

Aim: find a tradeoff between test sensitivity, test frequency = cost, and return delay

Step 1: Connect test sensitivity with virus load inside an individual

• Within-host model

• Temporal human data

Credit: Ke et al. MedRxiv 2020

Connect test sensitivity with virus load inside an individual

Step 2: Build the transmission model

S(t)

Testing

Delay in test return

Testing

Isolation

Isolation

Testing rate

Results: model dynamics

• Duke University dashboard

15% testing daily

• University of Virginia dashboard

8.8% daily testing

• Cornell University dashboard

22% daily testing

​

​

​

Results: model dynamics

A rapid test that detects virus above 100 vRNA per swab, 1,000

vRNA/swab and 10,000 vRNA/swab outperforms a PCR with a

return delay of 4.2, 3.3, and 2.2 days.

Rapid tests are cheaper – one can increase test frequency for the same budget!

Results: increased testing frequency

A rapid test that detects virus above

100,000 vRNA/swab reduces total

cumulative cases from:

• 54% to 1.3% for the same budget.
• 54% to 3.8% for half the budget.
• 54% to 5.4% for third of the budget.

How do vaccines influence testing strategies?

Modeling testing under vaccination – ongoing research

Testing

Isolation

S(t)

Testing

Delay in test return

Isolation

Delay in test return

Isolation

R(t)

Rv(t)

Accounting for the Delta Variant

Alpha

Delta

Bernal et al. NEJM 2021

Breakthrough cases at day 100

Alpha variant Delta variant

Testing and Vaccination - Alpha

Testing and Vaccination - Delta

Role of transmission coefficient

Role of vaccine waning

Conclusions - Testing and Vaccination

• Testing and isolation combine to make a powerful tool for outbreak prevention, especially early in an epidemic.
• The value of testing diminishes as vaccination becomes more widespread with higher infectivity of emerging variants being the highest factor in determining where peak of test efficacy
• Vaccination provides better long terms control
• Ideal distribution of public health resources is a combination of testing and vaccination
• The precise optimal mixture will change depending on the success of vaccine deployment and the state of the epidemic

Caveats:

• Epidemic characteristics bias cross-sectional viral concentrations: how do we compare virus levels during different times in the epidemic?

​

​

​

​

​

​

​

• We considered the effect of vaccines on blocking transmission. What about severe disease?

​

​

​

​

​

​

​

​

• Priority for symptomatic testing, high risk groups, essential workers
• Test sensitivity above the limit of detection varies according to symptoms status
• Worst case scenario = other PPE were not considered

​

​

Li et al. medRxiv

Chia et al. medRxiv

Acknowledgements

Questions?