Although inhaled corticosteroids are commonly used to treat patients with chronic obstructive pulmonary disease (COPD), their effect on clinical outcomes such as exacerbation and mortality is unknown. This systematic review was conducted to determine whether inhaled corticosteroids improve clinical outcomes for patients with stable COPD. All placebo-controlled randomized trials of inhaled corticosteroids given for at least 6 months for stable COPD were identified by searching MEDLINE (1966-2000), EMBASE (1980-2001), CINAHL (1982-2000), SIGLE (1980-2000), the Cochrane Controlled Trial Registry, and the bibliographies of published studies. We independently extracted data from each of the studies using a specified protocol, and determined the summary risk ratios (RRs) and 95% confidence intervals (CIs) for exacerbations and deaths. Nine randomized trials (3976 patients with COPD), including four with a systemic steroid run-in phase, were identified. Use of inhaled corticosteroid therapy reduced the rate of exacerbations (RR = 0.70; 95% CI: 0.58 to 0.84), with similar benefits in those who were and were not pretreated with systemic steroids. Inhaled corticosteroid therapy was also associated with increased rates of oropharyngeal candidiasis (RR = 2.1; 95% CI: 1.5 to 3.1), skin bruising (RR = 2.1; 95% CI: 1.6 to 2.8), and lower mean cortisol levels. No effects were seen on all-cause mortality (RR = 0.84; 95% CI: 0.60 to 1.18) in the five trials that measured this outcome. This systematic review demonstrates a beneficial effect of inhaled corticosteroids in reducing rates of COPD exacerbation. Further research is required to define the long-term effects of these medications and the benefit/risk ratio for patients with COPD.