Abstract
Background Restrictions on social interaction and movement were implemented by the German government
in March 2020 to reduce the transmission of coronavirus disease 2019 (COVID-19). Apple's
“Mobility Trends” (AMT) data details levels of community mobility; it is a novel resource
of potential use to epidemiologists.
Objective The aim of the study is to use AMT data to examine the relationship between mobility
and COVID-19 case occurrence for Germany. Is a change in mobility apparent following
COVID-19 and the implementation of social restrictions? Is there a relationship between
mobility and COVID-19 occurrence in Germany?
Methods AMT data illustrates mobility levels throughout the epidemic, allowing the relationship
between mobility and disease to be examined. Generalized additive models (GAMs) were
established for Germany, with mobility categories, and date, as explanatory variables,
and case numbers as response.
Results Clear reductions in mobility occurred following the implementation of movement restrictions.
There was a negative correlation between mobility and confirmed case numbers. GAM
using all three categories of mobility data accounted for case occurrence as well
and was favorable (AIC or Akaike Information Criterion: 2504) to models using categories
separately (AIC with “driving,” 2511. “transit,” 2513. “walking,” 2508).
Conclusion These results suggest an association between mobility and case occurrence. Further
examination of the relationship between movement restrictions and COVID-19 transmission
may be pertinent. The study shows how new sources of online data can be used to investigate
problems in epidemiology.
Keywords
COVID-19 - generalized additive model - disease - mobility
