For more than 30 years, several attempts to quantify severity of injuries on a numerical scale have been developed and applied. These score systems play an important role in research but also in comparative analyses in quality assessment. Besides the Injury Severity Score which is by far the most frequently applied score, several more sophisticated approaches have been developed to better describe injury severity and to improve outcome prediction. Besides a description of the most important scores, quality measures for scoring systems such as accuracy, ROC (receiver operating characteristics) analysis, and goodness of fit are presented.The main areas for application in quality assessment are, improvement of triage decisions, the identification of patients with unexpected outcomes, the generation and validation of audit filter, and, most importantly, the use as an objective basis for external and internal outcome comparisons. A score-based prognosis for a group of patients can be regarded as an external standard or norm to which the own results can be compared.But although score systems are an excellent tool for objective risk adjustments, results can still be influenced by different types of bias which have to be regarded. The main sources of bias are the application to different types of patients, or the selective use in only a specific subgroup of patients. The choice of a simple-to-assess score with high reliability and low interrater variability is a further prerequisite for receiving valid results.Critical evaluation of outcome is a central point in quality assessment, and the careful application of score systems provides an accepted basis for risk adjustment.Scores are not the key elements of quality improvement efforts, but without them, many approaches simply become impossible.