The profile of liver disease admissions and associated deaths in a district general hospital was studied to determine whether patients with end stage liver disease are appropriately referred for consideration of liver transplantation. Admission details were provided by the Office of Population Censuses and Surveys (OPCS) and their accuracy was assessed by case note analysis. According to OPCS, 77 patients with liver disease were admitted on 113 occasions between 1 January 1987 and 31 December 1989. The case notes of 74 (96%) were retrieved and examined. Only 64 (86%) had primary liver disease. Twenty four (31%) died of liver failure. Alcohol was the aetiological agent in two thirds. According to accepted criteria, 11 patients were suitable for liver transplant assessment but only three had been referred to a transplant centre. Of the remaining eight, five died during the study period. Two of the three patients referred died without transplantation; one underwent transplant and survived. There is discrepancy between OPCS data and true disease aetiologies, with approximately 40% under reporting of alcoholic liver disease. If this population is representative of the situation nationally, substantial numbers of patients with end stage liver disease might benefit from liver transplantation, but are not referred to a centre.