This review addressed a well-defined question in terms of participants, interventions, outcomes, and study design. However, there appeared to be a discrepancy between the inclusion criteria and what was actually included; this may have potentially affected the reproducibility and reliability of the review. The search included appropriate databases, but was restricted to English, so the potential for language bias could not be ruled out. Two reviewers independently selected the studies and assessed quality of the included studies using standard criteria. It appeared that the similar efforts to minimise error and bias were applied to data extraction. Potential sources of heterogeneity were explored; the methods used were appropriate.
Although the authors offered an appropriately cautious conclusion based on the limited evidence presented, the reliability of the review is potentially compromised by bias and error in the search and selection of studies.
Four of the authors disclosed financial links with Lundbeck Pharmaceuticals and/or Forest Laboratories (both manufacturers of memantine).