Additionally, at least three other questions need to be considered when contemplating inhaled corticosteroids for the patient with COPD: first, is the benefit of these drugs most likely to be obtained when used in combination with long-acting bronchodilators; second, do these agents reduce mortality; and third, are they safe? A number of studies utilizing combinations of either budesonide and formoterol or fluticasone and salmeterol have been performed, and a Cochrane systematic review11 of these studies has suggested that while the combination of an inhaled corticosteroid and a long-acting β-agonist is more effective than placebo, there is heterogeneity between studies as to whether or not combination therapy reduces exacerbations or improves quality of life or lung function when compared to inhaled corticosteroids used alone. With regard to the question of mortality reduction, retrospective studies, including a previously published report9 from the ISEEC study group, have suggested that inhaled corticosteroid monotherapy reduces all-cause mortality in COPD by approximately 25% when compared with placebo. Whether this effect is due to a reduction in exacerbations or is mediated by alterations in systemic inflammation or other factors remains unknown, and the reduction in mortality observed in these secondary analyses remains to be confirmed by prospective studies. When published, the results of the now-completed Towards a Revolution in COPD Health study,12 a 3-year study with > 6,000 participants assessing the effect of placebo, fluticasone, salmeterol, and fluticasone/salmeterol combination on all-cause mortality in COPD, will provide prospectively obtained data on this important outcome. This study12 should also add to our understanding of the long-term safety of both fluticasone and salmeterol, particularly with regard to steroid-related adverse effects such as reduced bone mineral density, and increased fracture and cataract risk.