[Mapping Regional Differences in Infection Rates for the Coronavirus (COVID-19): Results of a Bayesian Approach to Administrative Districts of Bavaria].

Background: Since the beginning of the COVID-19 pandemic, thematic maps showing the spread of the disease have been of great public interest. From the perspective of risk communication, those maps can be problematic, since random variation or extreme values may occur and cover up the actual regional patterns. One potential solution is applying spatial smoothing methods. The aim of this study was to show changes in incidence ratios over time in Bavarian districts using spatially smoothed maps.

Methods: Data on SARS-CoV-2 were provided by the Bavarian Health and Food Safety Authority on 29.10.2021 and 17.02.2022. The demographic data per district are derived from the Statistical Report of the Bavarian State Office for Statistics for 2019. Four age groups per sex (<18, 18-29, 30-64,>64 years) divided into 16 time periods (01/28/2020 to 12/31/2021) were included. Maps show standardized incidence ratios (SIR) spatially smoothed by Bayesian hierarchical modelling.

Results: The SIR varied remarkably between districts. Variations occurred for each time period, showing changing regional patterns over time.

Conclusion: Smoothed health maps are suitable for showing trends in incidence ratios over time for COVID-19 in Bavaria and offer the advantage over traditional maps in giving more realistic estimates by including neighborhood relationships. The methodological approach can be seen as a first step to explain the regional heterogeneity in the pandemic, and to support improved risk communication.