Skip to main content

Network meta-analysis for a vaccine

Our technical team conducted a meta-analysis for a vaccine.

What we did

Project background

The client was launching a new vaccine and wanted to gain optimal market access in global markets for this product. Asc academics was approached with the request to perform a network meta-analysis that compares the safety profiles of the client’s vaccine with those already on the market, in order to demonstrate the clinical and economical value of the product.

Challenges faced

The main challenge was to deal with the uncertainty around the estimates. The robustness of the results of a network meta-analysis depends on factors such as the amount of available literature and the sample size of the trials. In this case, the literature for the comparators of the new vaccine was scarce and the sample sizes of the trials were relatively small.

Our solution

The solution was to use the Bayesian methods instead of the frequentist ones. Bayesian statistics have the benefit of being flexible and going further than simple binary hypothesis testing, as opposed to the frequentist method. With their use, it is possible to provide some claims on the likelihood of various performance metrics by proving rank probabilities and median ranks with uncertainty intervals, which are more tangible and interpretable in terms of intervention performance. Therefore, Bayesian statistics were the perfect solution for a network meta-analysis involving a small number of studies and larger uncertainty.

Our impact

The odds ratio for twenty different endpoints was created through this analysis in order to compare the safety profile of several vaccines and provide insight into the optimal market access of the client’s product. The success condition for this project was proven by a request for a manuscript from the client.

Meet the experts

Timon  Louwsma, MSc

Timon Louwsma, MSc

Giorgia  Tiozzo, MSc

Giorgia Tiozzo, MSc

Services used

Do you have a question? just asc