The selection of the comparators was appropriate, as the widely used leukotriene-receptor antagonist montelukast was compared against second-generation antihistamines, including the recently approved levocetirizine. The authors stated that over-the-counter drugs, such as loratadine and cetirizine were not included as they were not relevant to health plan decision makers.
A systematic search of the literature was appropriately carried out to identify the relevant clinical inputs for the model. The method used to calculate the quality of life improvements for the efficacy of each treatment was reported and compared with an alternative method used in other studies. An indirect comparison between treatments was required, with placebo as the common comparator. The sources were RCTs, which are generally considered to be a valid source of evidence due to their methods. The benefit measure was justified by the authors and was clinically relevant, but might be difficult to compare with the benefits of other health care interventions.
The categories of costs appear to have been appropriate for the perspective of the third-party payer. No information on the unit costs and resource quantities was provided, limiting the transparency of the analysis. The data were from commonly used US sources and the approach used to derive the costs for levocetirizine was described. The authors noted that the costs for hospitalisations and emergency visits were not included as they were very rare in the population analysed.
Analysis and results:
The results were clearly reported and an incremental approach was appropriately used to synthesise the costs and benefits of the treatments. This approach allowed the identification of the dominant strategy. The Monte Carlo simulation showed the high level of uncertainty in the findings. The authors acknowledged that the main limitation of their analysis was the low number of RCTs found for some drugs, which led to high uncertainty in the cost-effectiveness results.
Overall, the methods were valid and the study was well reported. The authors’ conclusions are appropriate, but the high uncertainty in the findings should be considered.