Fatigue is among the most critical forms of deterioration damage that occurs to steel bridges. It causes a decline of the safety level of bridges over time. Much research has been accomplished on a reliability level, which focuses on evaluating the service life of the selected detail or the whole structure. However, the decisions aiming to keep the existing bridges in service should be further considered, i.e. whether interventions are needed, what kind of interventions could be applied, which intervention is the optimal choice to be both sustainable and economical within a budget restraint considered. Therefore a risk-based decision making framework is necessary for bridges service life extension. In this paper, a risk-based decision support framework is formulated aiming for rational decisions, which is presented as a decision tree. It presents possible intervention actions, evaluates the corresponding probability of failure and integrates them with possible consequences. A case study of the Rautasjokk Bridge in Sweden is incorporated to demonstrate the framework. A separation of the load bearing structure in subsystems is suggested for risk analysis and decision optimization. The expected result and how to prioritize the calculated utilities are discussed. And the results are compared with the reliability analysis to check if they are consistent.