Interventions:
The rationale for the choice of the interventions was clear since the addition of PPI to ASA represents a valid alternative to ASA alone. They are likely to represent relevant comparators in many settings.
Effectiveness/benefits:
The clinical estimates were derived from systematic reviews of the literature, the methods and conduct of which were described in part. The final sources used to estimate treatment effectiveness and other parameters of underlying disease were robust given the characteristics of their design. RCTs and literature reviews are usually considered good sources of clinical data. Uncertainty in some estimates was extensively addressed in the sensitivity analysis. The derivation of the benefit measure was clear, but it was unclear whether a discount rate was applied to the LYs.
Costs:
The analysis of the costs appears to have been consistent with the perspective adopted. The cost of hospitalisation was presented as a macro-category, reflecting the accounting system used in the authors’ setting. Alternative sources of drug costs were used; the authors also considered the perspective of large payers that may be able to achieve considerable discounts. The impact of variation in the costs was the focus of the analysis and was considered globally in the sensitivity analysis. The sources of the costs were reported, together with other characteristics of the analysis such as the price year and use of discounting.
Analysis and results:
The synthesis of the costs and benefits was appropriate. The authors provided an extensive presentation of the results of the analysis considering different combinations of patient age and risk of disease. The issue of uncertainty was addressed extensively in the sensitivity analysis, the alternative assumptions being mainly based on published evidence. Furthermore, the authors compared their findings with other published economic evaluations, showing similar results. Finally, limitations and strengths of the analysis were appropriately highlighted.
Concluding remarks:
The quality of the study methodology appears to have been satisfactory, with good reporting of the methods and sources, and clear presentation of the study results. The authors’ conclusions appear valid and robust.