Interventions:
The selection of the comparators was appropriate and they appear to be transferable to other settings. No dosage details were given.
Effectiveness/benefits:
A valid approach was used to identify the relevant sources of evidence. The key details of the reviews were reported and various sources were identified for the clinical inputs. Few details of these sources were given, but most of them (meta-analyses, literature reviews, and clinical trials) appear to have been valid. Wide ranges of values were used in the sensitivity analyses to consider variability in these data. The authors stated that conservative assumptions were made, where high quality data were not available. Life-years were a valid benefit measure, given the impact of cardiovascular disease on survival. The authors stated that not assessing quality-adjusted life-years (QALYs) biased their results in favour of aspirin alone, by overestimating the quality of life after recurrent MI.
Costs:
The cost categories reflected the third-party payer perspective, as stated. The resource use and unit costs were not presented separately; only total costs per item were reported, reducing the ability to reproduce the analysis. Typical US sources were used, including Medicare and drug price lists. The cost data were extensively varied in the sensitivity analysis, which showed the impact of the PPI drug price. Other details, such as the discount rate and price year, were reported.
Analysis and results:
The results were clearly presented, and the costs and benefits were synthesised, using an incremental approach. The uncertainty was satisfactorily investigated, using various approaches, and the key findings were reported. The authors stated that only MI was considered as a cardiovascular event and this might have underestimated the benefits of increased aspirin adherence with the PPI. The analysis appears to be transferable to settings with similar costs.
Concluding remarks:
The methods were valid and the authors’ conclusions appear to be robust.