A large number of studies show that hospitals are unacceptably noisy. Up to date no study has measured noise levels in intensive care units or neonatal wards that comply with the WHO recommendations. Furthermore, sound levels in hospitals have risen since the 1960s. The noise origins mainly from: (1) operational activities generated by the staff in their care giving activities and communication, (2) medical equipment and alarms and (3) structural sounds from the building such as ventilation and closing doors. While some sounds are unavoidable, many are totally or partially unnecessary. Noise in hospitals has been suggested to increase the risk for cardiovascular response, pain, intensive care delirium, fragmented sleep and reduced recuperation. For patients, the cause of these outcomes is multi-factorial, however the impact of the sound environment can, as opposed to most other factors, be abated. For the personnel, noise may cause annoyance, stress, tiredness and lead to more errors however these outcomes are less well investigated. The paper will give a summary of what is known today, specifically focusing on the outcomes from intervention studies of the physical environment and point to the most important areas for further improvements in research.