Evaluating the Impact of Testing and Timely Diagnosis on Disease Transmission

(Figure 1) A probabilistic model on the expected reduction in postinfection transmission risk for vaccinated versus unvaccinated individuals.

(Image credit: Study's authors)

A fundamental aspect of any program focused on testing and timely diagnosis of communicable diseases is its effectiveness in reducing transmission. A recent study, “Modeling the Transmission Mitigation Impact of Testing for Infectious Diseases,” published in Science Advances, introduces the concept of Testing Effectiveness (TE)—the fraction by which testing and postdiagnosis isolation reduce transmission at the population scale. Utilizing TE, the investigators give their guide recommendations on the optimal usage of today's rapid diagnostics to control various respiratory pathogens, including influenza A, respiratory syncytial virus (RSV), and omicron-era severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).

As the authors note, in the 4 years since the emergence of SARS-CoV-2, researchers have made significant developments in the test usage approach. Despite the documented successes of routine screening in nursing homes, college campuses, and nations, elective testing after known exposures or symptom onset has become more common. The array of available diagnostics has also expanded, with rapid diagnostic tests (RDTs) for RSV and influenza A, and new tests for SARS-CoV-2 using exhaled aerosols. Furthermore, multiplex RT-qPCR and rapid antigen lateral flow "triple tests" simultaneously detect these viruses.

“Moreover, because testing guidelines are only as effective as human behaviors allow them to be, it would be valuable for models to incorporate key behaviors such as imperfect participation and compliance and imperfect adherence to post-diagnosis isolation,” the authors wrote.

Estimating the population-scale impact of testing directly is challenging and often only possible retrospectively, making mathematical modeling a valuable tool for predicting effectiveness from first principles. Our model incorporates test specifications, within-host pathogen dynamics, and human behaviors to estimate TE.

To demonstrate the model's utility, the researchers applied it to study RSV, influenza A, and SARS-CoV-2. The model evaluates differences between pathogens and testing strategies, including optimal testing timing after respiratory symptoms appear and the trade-offs between test sensitivity and turnaround time (TAT). It also assesses the costs and benefits of postdiagnosis testing to shorten isolation times.

“To examine the impact of testing on community transmission, we developed a probabilistic model that integrates 4 key elements: (i) the properties of a particular diagnostic test; (ii) a strategy for its administration; (iii) the time-varying profiles of infectiousness, symptoms, and detectability over the course of an infection; and (iv) the key behaviors of participation (whether or not one refuses to test), compliance (whether or not one takes a recommended test), and isolation length,” the researchers wrote. (Figure 1.)

The model the authors created “was necessary because empirical evaluation of the impact of a testing program or behavior is difficult, with compelling analyses nevertheless lacking formal controls or requiring enormous scale.”

The study’s findings reveal that the timing of testing is critical. For RSV and influenza A, testing should begin immediately upon symptom onset to maximize TE. However, for SARS-CoV-2 in the omicron era, delaying testing by up to 2 days may be more effective. This delay accounts for the variable gap between symptom onset and the first detectability by RDTs. When fewer tests are available, strategic delays can increase TE by improving the probability of diagnosis while maintaining a high impact per diagnosis.

For SARS-CoV-2, the model indicates that rapid tests (RDTs) were superior to RT-qPCR for the founder strain due to RT-qPCR's longer turnaround times. However, changes in viral kinetics and rapid test sensitivity in the omicron era have reversed this trend. RT-qPCR now shows higher TE despite its longer turnaround times.

The model also evaluates the costs associated with test consumption and isolation days. Using post-diagnosis testing to shorten isolation times can reduce the average isolation period without significantly affecting TE. For example, elective post-symptom testing using 1 or 2 RDTs daily followed by a fixed isolation period or a test-to-exit (TTE) policy shows that TTE programs can decrease isolation days while maintaining TE.

The scientists suggest that no single testing strategy is optimal for all respiratory viruses due to differences in their within-host dynamics. Therefore, testing guidelines should be pathogen-specific and consider the timing and availability of tests. For instance, immediate testing is recommended for RSV and influenza A, while strategic delays may be more effective for SARS-CoV-2 in the omicron era.

Moreover, the study’s model underscores the importance of considering human behaviors, such as adherence and postdiagnosis isolation, in developing testing policies. Effective use of rapid diagnostics, paired with appropriate behavioral guidelines, can substantially increase the mitigation impact of testing.

This study highlights the importance of flexible and pathogen-specific modeling in guiding testing strategies for communicable diseases. By quantifying the impact of testing on disease transmission, the model provides valuable insights into optimizing test usage to control the spread of respiratory pathogens. The introduction of TE as a measure of testing effectiveness allows for a more nuanced understanding of how testing can reduce transmission at the population level.