Objectives: We hypothesised that, in the subgroup of seriously injured patients who receive early critical care in the operating theatre or intensive care unit, there would be a greater actual survival rate than that statistically predicted using trauma scoring techniques.
Methods: 1031 seriously injured patients on a national trauma database were analysed. The numbers of survivors in 3 initial destination groups [intensive care unit (ICU), theatre and ward] were compared with the average number of survivors statistically predicted for similar groups of patients using the TRISS methodology. W statistics were then used to test for statistical significance.
Results: 77/122 patients admitted to an ICU survived (predicted number 66, W stat 8.8 [2.6-15.0]). 129/178 patients transferred to theatre survived (predicted number 113, W stat 8.8 [4.2-13.5]). 296/348 patients admitted to a ward survived (predicted number 292, W stat 1.3 [-1.9-4.4]).
Conclusions: The number of patients who survive after severe injury is significantly greater than the number predicted to survive by current trauma scoring methods if the patient is sent directly to theatre or are admitted directly to the ICU. Survival is as predicted if the patient is sent initially to a ward.