The review question was supported by population and outcome criteria. Several databases were searched, although it was not clear whether the search was restricted by publication type or language. Publication bias was not assessed, but the small number of included studies would make this difficult. The authors did not report whether appropriate measures were taken to minimise the possibility of error and/or bias at any stage of the review process.
Study quality did not appear to have been assessed. There was a lack of study and patient details, which limited interpretation of the results. Results were calculated from aggregated study data. As highlighted by the authors, the review did not differentiate where drugs were taken in overdose versus where they were present incidentally.
Given that the evidence was based on a small number of observational studies (of uncertain quality) that rather crudely aggregated results to determine the review aims, in addition to limitations in the review process, the authors' conclusions should be interpreted with substantial caution.