This review discusses the statistical physics of vaccination, focusing on the role of human behavior and heterogeneous contact patterns in disease transmission. It traces the evolution of theoretical epidemiology from classical models assuming homogeneously mixing populations to more recent models incorporating behavioral feedback and spatial/social structures. Many methods used in this field originate from statistical physics, such as lattice and network models, and their analytical frameworks. The feedback loop between vaccination behavior and disease propagation forms a coupled nonlinear system with analogs in physics. The review also highlights the new paradigm of digital epidemiology, where digital data sources like online social media are used to gain high-resolution insights into individual behavior. With the tools and concepts of statistical physics and new digital data, models capturing nonlinear interactions between behavior and disease dynamics offer a novel way to model real-world phenomena and improve health outcomes. The review concludes by discussing open problems and promising directions for future research in the field. Keywords: epidemiology; vaccination; human behavior; complex networks; data.This review discusses the statistical physics of vaccination, focusing on the role of human behavior and heterogeneous contact patterns in disease transmission. It traces the evolution of theoretical epidemiology from classical models assuming homogeneously mixing populations to more recent models incorporating behavioral feedback and spatial/social structures. Many methods used in this field originate from statistical physics, such as lattice and network models, and their analytical frameworks. The feedback loop between vaccination behavior and disease propagation forms a coupled nonlinear system with analogs in physics. The review also highlights the new paradigm of digital epidemiology, where digital data sources like online social media are used to gain high-resolution insights into individual behavior. With the tools and concepts of statistical physics and new digital data, models capturing nonlinear interactions between behavior and disease dynamics offer a novel way to model real-world phenomena and improve health outcomes. The review concludes by discussing open problems and promising directions for future research in the field. Keywords: epidemiology; vaccination; human behavior; complex networks; data.