Interventions:
The choice of comparators appears to have been appropriate for the authors' setting. The dosages were provided.
Effectiveness/benefits:
The methods used to identify the studies for the effectiveness estimates were not reported, which makes it difficult to determine if the best available evidence was used. A meta-analysis was used to estimate the probability of developing a wound infection and this should ensure the validity of that estimate. The primary outcome measure was QALDs, which was appropriate.
Costs:
The perspective was clearly stated and the relevant costs appear to have been included. The resource quantities and unit costs were not reported separately. Most of the cost estimates were from the literature, but some of the drug costs were from the authors' institution and might not be generalisable to other settings. The price year was reported and the costs were appropriately standardised, using the consumer price index. Discounting was not performed and was not necessary given the time horizon.
Analysis and results:
The authors conducted an appropriate incremental analysis and the full results were presented. The uncertainty was satisfactorily addressed through a series of one-way and probabilistic sensitivity analyses. The authors discussed one limitation of their analysis, which was the lack of utility data for the postoperative period.
Concluding remarks:
On the whole, the methods seem to have been appropriate and were reasonably well reported. The conclusion reached by the authors reflects the scope of their analysis.