OBJECTIVE To develop a reproducible scoring system to identify patients who present with a doubtful diagnosis of appendicitis and who would benefit from diagnostic laparoscopy. METHODS Prospective observational study. METHODS Regional teaching hospital, The Netherlands. METHODS 577 consecutive patients during the period 1994-5, and 343 who presented during the period 1996-7. METHODS The variables that seemed to be predictive of acute appendicitis were abstracted from the earlier group. a logistic regression analysis applied, and score created. The score was validated on the 343 patients who presented during 1996-7, and then the groups were combined for further analysis. METHODS Reproducibility of the scoring system obtained by comparing odds ratios (OR) of the two groups; its effectiveness judged by comparing the delayed and normal appendicectomy rates. RESULTS The following variables were significantly correlated with the presence of acute appendicitis: white cell count 10 x 10(9)/L or more (score 3), rebound tenderness and male sex (score 2 each); and symptoms present for < 48 hrs and temperature 38 degrees C or more (score 1 each). The OR for the two groups were 1.80 and 1.76, respectively, indicating that score was reproducible. With a sensitivity of 93% and a specificity of 83% it would be at least as accurate as clinical judgment. The normal appendicectomy rate would be 7% instead of 9%. and the negative exploration rates (laparoscopy and primary appendicectomy) would both be 22%. The score would also result in a lower perforation rate (2% compared with 17%). CONCLUSIONS The score can indicate when there is an indication for laparoscopy in patients with suspected appendicitis.