CRD Summary: The review assessed the effects of inhaled budesonide on the risk of pneumonia in patients with chronic obstructive pulmonary disorder. The authors concluded that budesonide treatment did not increase the risk of pneumonia and was safe. The small number of events in the analyses, coupled with limitations in review methodology, means the authors' conclusions should be interpreted with caution.

CRD Commentary: The review addressed a clear question, supported by appropriate inclusion criteria. However, it was unclear why a trial which failed one of the criteria was still included in the review. A search to identify studies in any language was conducted using electronic databases. Although the authors reported use of the Jadad scale to assess study quality, this appears not to be a very informative assessment for a review restricted to including only double-blinded, randomised trials. Two reviewers independently selected trials for inclusion, but details of the processes used for quality assessment and data extraction were not reported; this possibly subjected the review to risk of reviewer error and bias. Despite thorough details being provided for most aspects of the included trials, it was still unclear which comparator treatments were used (placebo or formoterol), making it difficult to interpret individual trial results. Appropriate methods appeared to be used to pool data, although the authors did not state whether they used a fixed-effect or random-effects model. Statistical heterogeneity between trials was not formally assessed. The authors noted that none of the trials were powered specifically to detect pneumonia (out of the 14 treatment arms, only three reported occurrence of more than 20 adverse events). It appears likely that the trials were too small to reliably address the authors' objectives. In light of this, and other limitations of the review, the authors' conclusions should be interpreted with caution.