CRD Summary: This review found that pioglitazone was not associated with increased risk of cardiovascular events. Insufficient information presented and the conduct of the review made it difficult to draw any conclusions about the evidence presented and the reliability of the authors' conclusions.

CRD Commentary: The review addressed a question that was broad in scope. The search was limited to one electronic database, but there was also a search for unpublished trials. Steps taken to minimise bias were reported for some parts of the review process. Standard statistical methods were used to pool the data and potential sources of heterogeneity were explored. There appeared to be substantial clinical heterogeneity between patients in the included trials, as patients were included with Type 2 diabetes and other conditions. There was no information given on dose regimens of the study interventions and comparators. Given the clinical heterogeneity and the lack of information about the interventions and comparators, pooling these data may not have been appropriate. The included studies appeared to be of middling quality, given the inadequacy of blinding and randomisation procedures. Some data was only presented graphically in the review, which limited the interpretation of some of the findings. Some reporting errors, the lack of information presented and the conduct of the review made it difficult to draw conclusions from the evidence presented, and the reliability of the authors' conclusions is unclear.