Admissions to the intensive care unit (ICU) from the wards have a higher mortality when compared to patients admitted from the operating theatres/recovery and accident and emergency department.1 Suboptimal care may contribute to morbidity and mortality of patients admitted from the ward.2 Failure to appreciate physiological derangements of breathing and mental status has been demonstrated in patients who subsequently suffered cardiac arrest, and these events may have been apparent up to 8 h prior to the event.3 4 The Early Warning Score (EWS) was developed as a simple scoring system to be used at ward level utilising routine observations taken by nursing staff.5 Deviations from the normal score points and a total is calculated. The EWS was evaluated prospectively for 1 month. The score was then modified to include urine output, to make temperature deviations less sensitive and to include normalised blood pressure (Table 8). We then evaluated this prospectively for 9 months. A total score of 4 or more resulted in the patient being reviewed by ward medical staff and help sought from the intensive care team if appropriate. Over a 9-month period 206 patients on two general surgical wards were put on the scoring system, of these 26 were admitted to the ICU. The APACHE II scores of these patients was 16.6 (± 7.3). Eleven patients were admitted to the ICU from the surgical ward who had not been monitored on the modified EWS and their admission APACHE II scores were 23.5 (± 4.1). This compares with admission APACHE II scores of 22.3 (± 5.5) in 43 patients admitted from surgical wards in the 9-month period prior to introduction of the system. The introduction of the system has appeared to lead to earlier referral to the intensive care unit.