Chronic obstructive pulmonary disease (COPD) is a major cause of mortality and morbidity. Relatively few pharmacoeconomic studies have been conducted on this disease. This article reviews available information about the utilisation of healthcare resources and cost of care, and the cost or cost effectiveness of therapeutic interventions reported for this disease. Burden-of-illness data indicate that hospital care, medications and oxygen therapy were the major cost drivers in these studies. Mean annual Medicare expenditures in the US were $US11,841 (2000 values) for patients with COPD compared with $US4,901 for all covered patients. Utilisation was skewed; the most expensive 10% of the Medicare beneficiaries accounted for nearly 50% of total expenditures for this disease. Costs are associated with health status, age, physician specialty, geographic location and type of insurance coverage. Six types of interventions were assessed in the literature--pharmacotherapy, oxygen therapy, home care, surgery, exercise and rehabilitation and health education. The studies used different analytic strategies (e.g. cost-minimisation and cost-effectiveness analyses) and even within the realm of cost-effectiveness analyses, no uniformity existed as to how outcome was measured. Patient severity was not always delineated, and the length of the follow-up period, while quite short, varied. Only 11 of the 34 evaluations were based on randomised controlled trials. Cost-minimisation studies generally found no significant difference in the cost of antimicrobial treatment for first-line, second-line and third-line agents. Studies of bronchodilators indicated that ipratropium bromide alone or in combination with salbutamol (albuterol) was the preferred medication. The major area for achieving cost savings is by reducing hospital utilisation. As the annual rate of hospitalisation is relatively low, large patient samples will be required to demonstrate an economic advantage for a new therapy. The major challenges will be financing such a study, and selecting an outcome measure that satisfies both clinical and economic conventions.