There is a considerable amount of evidence pointing to the existence of a weekend effect of increased in-hospital mortality in many countries. The underlying reason for the effect is now the subject of more forensic studies, but considerable care should be exercised to ensure analyses are not confounded or based on prior beliefs only. This study uses a data driven approach to identify factors associated with the weekend effect. In-hospital mortality was higher for patients admitted on a Sunday for elective admissions whereas mortality was similar for all days of admission for emergency patients. Cluster analysis identified a number of high mortality AR-DRGs associated with weekend admissions. However, when these were used in a separate analysis no weekend effect was found. Future studies using a data driven approach will gain more insight into the "weekend effect".