Two possible strategies for diagnosis and staging of patients with suspected biliopancreatic cancer and obstructive jaundice were tested in a decision analysis. One strategy was called 'surgical' strategy and consisted of only Ultrasonography and (spiral) CT scan and exploratory laparotomy in all patients without irresectable disease or distant metastases according to this work-up. The other strategy was called 'non-surgical' strategy and consisted of the above strategy, but also included ERCP and endoscopic internal drainage (stent) and diagnostic laparoscopy and laparoscopic ultrasound. In the decision analysis incidences of the outcomes after the two strategies were estimated based on the literature and our own experience, and a utility (0-1) was attributed to each separate outcome, according to the expected quality of life as judged by the team. Total utility of each strategy was calculated. It could be concluded that a non-surgical strategy would lead to a somewhat higher total utility in case the yield (i.e. prevention of unnecessary laparotomies) could be high and supposed that a beneficial effect of pre-operative internal biliary drainage was present. In case of low yield of diagnostic laparoscopy the superiority of the 'non-surgical' strategy is doubtful.