A wide variety of TTs were in use, with little evidence of reliability, validity and utility. Sensitivity was poor, which might be due in part to the nature of the physiology monitored or to the choice of trigger threshold. Available data were insufficient to identify the best TT.

Summary We analysed the physiological values and early warning score obtained from 1047 ward patients assessed by an intensive care outreach service. Patients were either referred directly from the wards (n = 245, 23.4%) or were routine critical care follow‐ups. Decisions were made to admit 135 patients (12.9%) to a critical care area and limit treatment in another 78 (7.4%). An increasing number of physiological abnormalities was associated with higher hospital mortality (p < 0.0001) ranging from 4.0% with no abnormalities to 51.9% with five or more. An increasing early warning score was associated with more intervention (p < 0.0001) and higher hospital mortality (p < 0.0001). For patients with scores above one (n = 660), decisions to admit to a critical care area or limit treatment were taken in 200 (30.3%). Scores of all physiological variables except temperature contributed to the need for intervention and all variables except temperature and heart rate were associated with hospital mortality.

Simple physiological observations identify high-risk hospital inpatients. Those who die are often inpatients for days or weeks before death, allowing time for clinicians to intervene and potentially change outcome. Access to critical care beds could decrease mortality.

Early identification of patients at risk, both before admission and after discharge from the ICU, may allow treatment to decrease mortality. Research and resources may be best directed at patients who die, despite a relatively low predicted mortality. Although these patients are a small percentage of the low-risk admissions, they constitute a large number of ICU deaths. Many patients die after discharge from ICU and this mortality may be decreased by minimizing inappropriate early discharge to the ward, by the provision of high-dependency and step-down units, and by continuing advice and follow-up by the ICU team after the patient has been discharged. Intervention before ICU admission and support of patients after discharge from the ICU should be part of the effort to decrease mortality for ICU patients. Inadequate provision of resources for critically ill patients may result in excess intensive care mortality that is not detected with ICU outcome prediction methods.

A 'patient-at-risk team', established to allow the early identification of seriously ill patients on hospital wards, made 69 assessments on 63 patients over 6 months. Predefined physiological criteria were not able to reliably predict which patients would be admitted to the intensive care unit. The incidence of cardiopulmonary resuscitation before intensive care admission was 3.6% for patients seen by the team and 30.4% for those not seen (p < 0.005). Of admissions seen by the team, 25% died on the intensive care unit compared with 45% of those not seen (not significant, p = 0.07). Among those not seen by the team, mortality was 40% for those who did not require resuscitation and 57% for those who did (not significant). Many critically ill ward patients had abnormal physiological values before intensive care unit admission. Identification of critically ill patients on the ward and early advice and active management are likely to prevent the need for cardiopulmonary resuscitation and to improve outcome.