Crohn's disease is a chronic inflammatory bowel disease of unknown aetiology which affects around 35,000 people in the UK (population 56.8 million). The potential for onset in early adult life, disease chronicity and a need for hospitalisation and surgery mean that the disease can be associated with substantial healthcare costs. Cost-of-illness studies focusing on direct medical costs have identified that over half the average costs associated with the disease relate to hospital costs. Estimates of the contribution of drug costs to the total direct economic burden have varied between 4.6 and 25%. Figures for average annual direct costs per patient in the US have been put at between US dollars 6561 (1990 values) and US dollars 12,417 (1994 values), whereas European studies have given much lower cost estimates (US dollars 655, 1994 values). However, all studies have highlighted that much of the total cost of illness relates to extensive interventions required by a small proportion of severely affected individuals. Indirect costs associated with reduced productivity in Crohn's disease can be high, with long periods of absenteeism and early disability. However, most patients (90%) remain in the workforce and life expectancy is relatively normal. A variety of drugs are employed for the treatment of Crohn's disease, both in an attempt to induce clinical remission in active disease and to maintain remission once this has been achieved. Comparative data on cost effectiveness is lacking, though crude estimates based on randomised trials suggest that the frequently prescribed aminosalicylates, which have only modest efficacy, are a relatively costly drug option. The costs associated with adverse drug effects, particularly for corticosteroids, have not been formally quantified. Despite high costs, new drug therapies for more severe disease, such as anti-tumour necrosis factor (TNF-alpha) antibodies, may prove a cost-effective option if the need for hospitalisation is reduced. In a modelling exercise, a US group estimated that if a theoretical new drug was introduced which was capable of reducing non-drug costs (including hospitalisation) by a fifth despite doubling the overall drugs bill, there would still be a reduction in the overall costs of Crohn's disease by 13%. Although surgical therapy is costly, there may be prolonged post-surgical remission following resection of localised disease and early surgery may represent a cost-effective option for selected patients. Without formal cost-effectiveness analyses, or (better still) clinical trials incorporating cost data, decisions about the relative efficiency of treatment alternatives for Crohn's disease remain subjective and more research is clearly required in this area.