Interventions:
The comparators were the two most commonly prescribed statins, as shown by their official market shares from 1997 to 2010 (70% of the market). They are likely to be relevant comparators in other settings.
Effectiveness/benefits:
The clinical data were from two published cost-effectiveness analyses, which obtained treatment effect data from two head-to-head randomised clinical trials. These trials should have had high internal validity, but they could have been described in more detail. The authors acknowledged that they assumed perfect compliance, and it was unclear whether this could have favoured either drug. Extensive sensitivity analysis was conducted on some clinical parameters. QALYs were an appropriate benefit measure, as they capture the effect of the treatments on survival and quality of life, which are both affected in patients on statins. The sources for the utility weights were not reported.
Costs:
The economic analysis included all those costs relevant to the perspective of the payer. The authors stated that a broader perspective, such as that of society, would have improved the cost-effectiveness of atorvastatin due to the greater reduction in cardiovascular events compared with simvastatin. The unit costs and quantities of resources were not reported, except for the unit costs of drugs, which changed over time with generic or branded prices. Some sources for the costs of cardiovascular events were described and these appear to have been cost-effectiveness analyses. The impact of alternative costs was assessed in the sensitivity analysis. Reflation exercises are feasible as the price year was explicitly stated.
Analysis and results:
The model outcomes were clearly presented. The costs and benefits were appropriately synthesised, using an incremental approach. A deterministic approach was used to assess uncertainty, and the findings were clearly reported. The model used to estimate the long-term costs and benefits was not described; a published model might have been used. The authors stated that other economic evaluations had shown the cost-effectiveness of atorvastatin, but these usually used placebo as a comparator and its branded price. These results were specific to the US setting.
Concluding remarks:
A valid cost-effectiveness framework was used and, despite limited reporting of some data sources, the authors’ conclusions appear to be robust.