Test-Negative Designs with Multiple Testing Sources

Area of Science:

• Epidemiology
• Vaccinology
• Biostatistics

Background:

• Test-negative designs (TNDs) are common for assessing infectious disease interventions like vaccines.
• TNDs traditionally reduce confounding from healthcare-seeking behavior by testing symptomatic individuals.
• Recent challenges include bias from aggregating symptomatic and asymptomatic test results, particularly for diseases like COVID-19 and Ebola.

Purpose of the Study:

• To address the 'multiple reasons for testing problem' in TNDs.
• To propose a method for estimating vaccine efficacy using combined symptomatic and asymptomatic test results.
• To assess if vaccine efficacy is consistent across different testing sources.

Main Methods:

• Utilized a modified TND approach for an Ebola Viral Disease (EVD) vaccine trial.
• Incorporated testing of close contacts of symptomatic, test-positive individuals.
• Developed a statistical method to estimate common vaccine efficacy from dual testing sources.

Main Results:

• The study proposes a method to estimate vaccine efficacy from symptomatic and asymptomatic cases.
• The approach allows for assessing if efficacy differs between these two groups.
• While the EVD trial concluded early, the methodology remains relevant for future vaccine efficacy studies.

Conclusions:

• A refined TND approach can mitigate bias from multiple testing reasons.
• Accurate assessment of vaccine efficacy requires accounting for diverse testing scenarios.
• The proposed method is crucial for future infectious disease vaccine trials, especially during outbreaks.