Introduction:
The COVID-19 pandemic challenged health systems worldwide to collect data broadly, rapidly, and safely to better understand the virus and respond dynamically to new developments. As new variants require new vaccines, adaptive and scalable study solutions were needed to test vaccine effectiveness (VE) and drug safety under real-world conditions. We investigated the feasibility of a digital study app to collect data on COVID-19 in Germany with a focus on VE.

Methods:
In a digital prospective cohort study, we collected self-reported data between May 2021 and April 2022 to assess test-positive rates, incidences, suspected adverse events after vaccination, and VE. We used binomial regression models adjusted for confounders to estimate VE.

Results:
A total of 9,592 individuals participated in our study. Incidence and test-positive rates in our cohort reflected the course of the pandemic in Germany. 1,768 unvaccinated participants were compared with 5,368 participants who had received at least two doses of any COVID-19 vaccine and 2,433 participants who had received three doses. Adverse events after vaccination were consistent with previously published research. Preliminary analyses revealed VE against infections across all vaccine brands waning from 89.1% (95% CI 68.2-96.2) at week 4 to 45.5% (95% CI 29.6-57.8) at week 28 after the second dose. Third vaccine dose restored VE to 67.5% (95% CI 53.6-77.1) after 4 weeks. We observed similar preliminary results for two and three doses of the BNT162b2 vaccine.

Conclusion:
We successfully tracked the course of the pandemic in the German population using self-reported data from a small, self-selected cohort. Registration of COVID-19 VE with self-reported data in our study app was feasible but presented numerous challenges and potential biases that require further refinement for successful use as a digital public health tool.