Understanding and controlling the evolution of antibiotic resistant strains is one of the biggest public health challenges of our time. Despite a vast amount of data gathered and models being developed, coexistence of antibiotic resistant and sensitive genotypes within the same bacterial pathogen is still an unresolved problem. Simple epidemiological models predict the dominance of either of the two strains while more complex models suffer from generality. Using empirical evidence, I set out to resolve this problem by coupling within-host pathogen dynamics and between-host transmission of bacteria. First, stochastically modelling the within-host system I will develop predictions for the rate of resistance emergence and abundance of sensitive and resistant individuals in hosts with or without antibiotic treatment. While resistant bacteria thrive under antibiotic treatment, the sensitive strain has an advantage in invading and colonising untreated hosts. The outcomes help to get a more detailed understanding of the within-host dynamics, e.g. identification of optimal treatment strategies to confine the evolution of antibiotic resistance. Feeding these results into the dynamics on the population level, the between-host level, will result in a within-between-host feedback. Fitting and confronting the model to empirical data on prevalence and resistance emergence in Streptococcus pneuomoniae and Escherichia coli will conclude this project. The mechanistic implementation of the dynamics can immediately be linked to data which is of great importance given the increasing amount of empirical studies in the field of epidemiology. Through the theoretical and applied results, the study will add new insights and predictions in the field of infectious disease evolution and be able to identify factors enabling the stable coexistence of antibiotic resistant and sensitive bacteria.