Fifty consecutive critically ill patients transported between hospitals by a mobile intensive care team were assessed prospectively using a modification of the acute physiology and chronic health evaluation (APACHE II) sickness scoring system. Assessments were made before and after resuscitation, on return to base, and after 24 hours of intensive care. No patient died during transport. Twenty two patients died subsequently in hospital and 28 survived to return home. The mean score for the non-survivors before resuscitation was 21.7 and for the survivors 12.2 (p less than 0.0005). Among the non-survivors there was a significant fall in score with resuscitation but this did not alter their subsequent outcome. Neither group deteriorated during transport. The sickness score is a powerful method for determining prognosis, and employed longitudinally it may be useful in the assessment of treatment. It has important implications for the administration and organisation of regional intensive care services.