The emergence of antibiotic resistances due to antibiotic residues in urban sewage systems is becoming an increasingly important issue. This paper presents a model for the spatiotemporal analysis of antibiotic inputs to derive spatiotemporal distribution patterns which are the basis for later predictions of future antibiotic inputs into the sewer system. To identify spatiotemporal distribution patterns of antibiotic prescriptions data statistical and GIS methods like time series and spatial cluster analysis are used. In order to find possible interrelationships the prescription data is combined with other influencing parameters (e.g. cases of respiratory infections) and tested for statistical correlations. Results show a pronounced seasonal course for three antibiotics of the macrolide group which also show high correlations with cases of respiratory infections in the study area. Further, results show that weekly data of respiratory infections by Google Flu Trends may be used as predictor variable to derive forecasts of future antibiotic inputs into the sewer system