Interventions:
The selection of the comparators was appropriate as the proposed antibiotic-impregnated shunts were compared with the standard shunts and infection prevention.
Effectiveness/benefits:
A valid approach was used to derive the clinical estimates. The methods and conduct of the literature review were clearly presented. The search was conducted by two independent reviewers. The inclusion and exclusion criteria were listed. The methods of the included studies were reported, as well as the study design. A random-effects meta-analysis was appropriate. Publication bias was investigated and discussed. The results were clearly presented for each study and for multi-institutional versus single-institution studies. The clinical part of the study was satisfactorily investigated. The main outcome measure was disease-specific and might not be easily comparable with the benefits of other health care interventions.
Costs:
Limited information was given on the economic data, but this is reasonable as the primary focus of the study was on the clinical meta-analysis. The perspective was not reported, but might have been that of the third-party payer or hospital. The costs were from published US studies and from the authors’ institution, and are therefore likely to be specific to the USA. The price year was not reported and the costs were not varied in sensitivity analysis. Note: information received from the author subsequent to this abstract being published indicates that the price year was 2010.
Analysis and results:
The results were extensively reported. The number-needed-to-treat was used to synthesise the clinical impact of the two procedures. No specific approaches were used to investigate uncertainty. The authors acknowledged that the main limitation of their analysis was the lack of good published randomised controlled trials and the evidence was mainly from retrospective cohort studies. Similar results may be found in settings with similar clinical practice and costs of procedures and hospitalisations.
Concluding remarks:
A valid approach was used to assess the comparative efficacy of the two procedures. The authors’ conclusions appear to be robust, but the costs were not extensively reported.