Power and sample size considerations for test-negative design with bias correction: a case study on the world first malaria vaccine
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Test-negative design (TND) studies are increasingly common in evaluating vaccine effectiveness (VE) for various infectious diseases. TND studies are susceptible to bias due to disease outcome misclassification caused by imperfect test sensitivity and specificity. Several bias correction methods have been proposed. However, sample size or power considerations for TND studies incorporating bias correction for such misclassification have not yet been investigated.
Methods
Motivated by the world’s first malaria vaccine, we investigated how bias correction influences statistical power and sample size for VE estimation using Monte Carlo simulations. Under realistic assumptions about the proportion of vaccinated individuals in the general population, VE against clinical cases, and the probability of malaria diagnosis in unvaccinated individuals, we estimated the power to detect VE across different malaria vaccination statuses, with and without bias correction, at diagnostic test sensitivities of 60%, 80%, and 95%, and a specificity of 98%.
Results
The results demonstrated that using imperfect diagnostic tests reduces statistical power in both observed data-based VE and bias-corrected VE. The magnitude of power loss was substantially influenced by the sensitivity of the tests.
Conclusions
During the design phase of a TND study, researchers should conduct power calculations accounting for correcting the bias due to outcome misclassification. To achieve this, researchers need to collect comprehensive data, including the expected effect size of VE, the sensitivity and specificity of the diagnostic tests, the proportion of the vaccinated group, and the case ratio of the target disease.
