Background: Irrational use of antibiotics proves major concerns to the health systems globally. It results in antibiotics resistance and increases health care costs. In Iran, many years of research, appreciable efforts and policy making have been of little avail and indicators still show suboptimal use of antibiotics, pointing out an urgent need to alternative approach to understand the problem and generate new solutions. Applying the Complex Adaptive Systems theory, to explore and research in health systems and their challenges and has become popular. Therefore, this study was aimed to better understand the complexity of irrational use of antibiotic use in Iran and the potential solutions. Method: This research utilized a CAS observatory tool to qualitatively collect and analysis data. Many interviews with key informants were conducted. The data was enriched with documents reviews in order to fully understand the system. MAXQDA software was applied to organize and analyze the data. Result: We could identify several diverse and heterogeneous, yet highly interdependent agents in the antibiotic consumption system in Iran, operating at different levels. The network structure and its adaptive emergent behaviour, information flow, governing rules, feedbacks and values of the system and the way they interact were identified. The gaps and weakness of the system which needs redesigning or modification were recognized as well. Findings describe antibiotic use as an emerge behavior of the system which is formed by interplay of many factors and actors over time.Conclusion: The study suggests re-engineering the system by implementing several system level changes including establishing strong, timely and effective interactions between identified stakeholders which facilitate information flow and provision of on time feedbacks, create win-win rules in participatory manner with stakeholders and distributed control system.