In-hospital cardiac arrests, intensive care unit (ICU) admissions and unexpected deaths are commonly preceded by warning signs up to 24 hours prior to the event. As a result, some of these critical events are potentially preventable. Critical care physicians are increasingly familiar with patient care systems; trauma systems have become well established in most health services, and the chain of survival provides a system response to out of hospital cardiac arrests. We now need to build on experience with systems to extend critical care services to all hospital patients at risk, whatever their location and on a continuous basis to prevent these critical events from occurring. In fact, if critical care medicine is to take up the challenge and move forward into the 21st century, we need to engage in a re-orientation from individual to system thinking. We know that the majority of in-hospital cardiac arrests occurring on the general wards represent failures in the system. These events are not the fault of one or two individual practitioners that failed to provide adequate care, but a consequence of organisational factors that result in failures in recognition and response involving more than one department, professional group or area of the hospital. There is also potential to reduce morbidity. Morbidity caused by failure to adequately treat hypoxemia and hypovolemia on the wards, results in preventable cases of renal and respiratory failure, requiring prolonged, uncomfortable and expensive admissions to intensive care, along with the invasive therapy that ICU admission entails. The Medical Emergency Team (MET) system provides a potential solution.