Box 1.

Main results of the review on statins.

Methods	Thirteen studies included a cost-utility analysis,20–27,29,31,33–35 four studies a cost-effectiveness analysis28,30,32,36 and the remaining two both the techniques together.37,38 Seventeen studies took a third-party payer’s viewpoint,20–32,34–36,38 only two a societal perspective33,37 – one of them without including indirect costs.30 Eleven of the 13 cost-utility analyses and one of the four cost-effectiveness analyses were based on Markov models20,22,23,25–27,29,31,33–36 – such as the two studies including both the techniques37,38 – while the remaining five21,24,28,30,32 used other models. Eleven studies adopted a lifetime horizon,20,24–26,29,31,33–37 five were designed over long-term periods21–23,27,38 and only three were short-term.28,30,32
Costs	All the studies estimated the costs of drugs and management of cardiovascular disease events (hospitalisations, follow-up treatments and monitoring procedures). Five of the 19 included the costs of surgical interventions20,30,31,34,37 and six of those related to death too;20,25,27,29,35,38 three extended the estimates to direct non-healthcare costs (i.e. travel).32,33,35 The Spanish study28 was the only one that estimated the costs relating to adverse drug effects.
Funding	Fifteen studies were funded by industry and all concluded in favour of the sponsored products.20–25,30,32–38 Three of the remaining studies concluded in favour of a statin treatment,26,28,29 while the last – focused on patients at low risk in primary prevention – was the only one that was unfavourable.27
Statin resistance	Only four studies took account of resistance to statins. The German25 and Finnish33 cost-utility analyses in secondary prevention mainly differed in the efficacy sources (respectively, one short-term clinical trial and various clinical studies selected through a systematic literature search) and the single statins assessed (only simvastatin in the former, also atorvastatin and rosuvastatin in the latter). Both studies concluded in favour of an association of actives (simvastatin + niacin/laropiprant in the former, simvastatin + ezetimibe in the latter) – all drugs marketed by the (same) sponsor. In the Spanish cost-effectiveness analysis,28 all statins and combinations with cholestyramine/ezetimibe were compared with no treatment in primary prevention, while the Swedish cost-effectiveness analysis32 limited the comparison to eight different statin therapies for high-risk patients with hypercholesterolaemia. Both studies estimated the efficacy of therapies through a meta-analysis and concluded in favour of rosuvastatin (the sponsor’s drug in the Swedish study), the Spanish study recommending combination therapies too when greater reductions in low-density lipoprotein cholesterol are required.