Validity of the Glasgow Aneurysm Score and the Hardman Index in predicting outcome after ruptured abdominal aortic aneurysm repair (Br J Surg2005; 92: 570–573)

Sir We have recently finished an almost identical study. Like the authors, we noted that the two scoring systems were poor predictors of postoperative mortality following ruptured abdominal aneurysm repair, but furthermore, they also failed to correlate with each other. The original articles on the Glasgow Aneurysm Score and the Hardman Index1,2, reported specific subject populations, yet the two populations were very different, even in respect of basic demographic proportions (for example, the M : F sex ratios in the two studies were, respectively, 2 : 1 and 5·5 : 1). This indisputably influenced outcomes and has implications to all scoring systems, since no two populations are ever identical, given that they are in continual flux. In other words, scoring systems are intrinsically biased towards a specific population at a single point in time. The implication is that while a scoring system may hold true for a specific population at a particular point in time, this cannot automatically be assumed for other populations without appropriate revalidation. Going further, continual population flux implies that this revalidation process also applies for the index population from which the original scoring system was drawn. The hope of clinicians to develop predictive scoring systems that ‘withstand the test of time’ is intrinsically flawed since periodic revalidation is necessary to keep them accurate.