Interventions:
No formal justification was provided for the selection of the two treatments from among all the available drugs. However, it appears that they were chosen because they had been compared in the primary RCT.
Effectiveness/benefits:
The source of clinical evidence was highly appropriate as the authors carried out a head-to-head study of the two drugs. Although little information on the RCT methodology was provided, as it had been published elsewhere, its randomised and blinded design ensures a high internal validity. The patient inclusion and exclusion criteria were reported. Furthermore, the study groups were comparable at baseline in terms of socio-demographic and clinical characteristics. The clinical endpoints were adjusted by the baseline scores. These aspects of the analysis enhance the validity of the clinical estimates. The authors justified their choice of a disease-specific benefit measure, which was the relevant outcome for clinicians and the health care payer. However, a disease-specific measure cannot be directly compared with the benefits of other health care interventions.
Costs:
The authors explicitly stated the study perspective, and all the relevant cost categories appear to have been included. Extensive information on the unit costs, quantities of resources used, and the price year was provided, which enhances the transparency of the economic evaluation. However, the sources of data were reported only for the drugs. The approach used to derive the prices of the other health services was not reported. The resource use data reflected the actual consumption of health care services. Conventional statistical analyses of the costs were carried out in the sensitivity analysis. The authors noted that due to the ‘controlled’ nature of the study, visits to the health care professionals which were a planned part of the study, were not included in the cost analysis. Thus, the number of visits might have been underestimated.
Analysis and results:
The incremental approach used to combine the costs and benefits was appropriate. However, incremental cost-effectiveness ratios were not calculated given the dominance of one treatment over the other. The issue of uncertainty was addressed by means of a probabilistic approach. Only the price of CIT was varied in the deterministic sensitivity analysis, as at the time of the study, ESC was not marketed in France. Two alternative assumptions were made: either that the two drug prices were the same at the branded price of CIT, or that ESC had the branded price of CIT and CIT had its generic price. Finally, the authors noted some potential limitations of their analysis such as the use of patient self-reported data, which might affect the validity of these estimates given the potential for recall bias.
Concluding remarks:
The study methodology was of good quality and was well presented. The conclusions appear to be valid.