Around 90% of antibiotics used in human medicine are prescribed in primary care. At least half of these prescriptions are either unnecessary or inappropriate. Multifaceted interventions have proved useful in improving antibiotic prescribing, but they should be adapted to the context (‘One size does not fit all’ and ‘No magic bullet’ concepts). Behavior change theories are helpful for designing interventions that target healthcare professionals. Recent evidence related to various interventions targeting healthcare professionals are described here (Table 1): education, guidelines, clinical decision support systems, delayed prescribing, patient materials, public commitment, POC diagnostic tests, selective susceptibility reporting, quality indicators, audit and feedback, and restrictive prescribing measures. Interventions targeting the healthcare system have been overlooked so far, even though their impact can be large and sustained. Some examples are given here, including dispensing of antibiotics, financial incentives, and public reporting. Most published studies come from Europe or North America. We need studies from a much greater range of settings. Most interventions have focussed on GPs. In the future, all healthcare professionals should be targeted. Most studies have focussed on RTIs ; we need to develop a better understanding of how much improvement is possible for other conditions and how well suited our current interventions are to achieving this. Patient-centered outcomes remain largely underdeveloped and underreported. Innovative strategies, including those targeting the system organization, should be tested.