Cost-effectiveness analysis of alternative statin prescribing strategies for the secondary prevention of cardiovascular disease at a South African public sector tertiary hospital

Strokes and ischaemic heart disease are among the top ten causes of death in South Africa. Given that burden of disease, it is important to establish whether interventions aimed at preventing cardiovascular disease are not only effective, but cost effective too. Cost-effectiveness analyses compare interventions in terms of both their costs and consequences and are a useful tool for policymakers. Statins reduce the risk of cardiovascular events such as myocardial infarctions and strokes, by lowering low density lipoprotein cholesterol (LDL-C) concentrations. Several studies, mostly conducted in Europe or North America, have demonstrated that statins are cost effective, particularly when used to reduce the risk of further cardiovascular events in patients who already have cardiovascular disease (secondary prevention). Despite their widespread use, there are no published cost-effectiveness analyses of statins for the secondary prevention of cardiovascular disease in South Africa. There are also only limited local efficacy data from clinical trials and no costing data of cardiovascular events from a public healthcare sector perspective. There is some debate regarding the optimal statin dose. Some guidelines recommend increasing statin doses until target LDL-C concentrations are achieved, while others recommend prescribing statins at a fixed high dose without monitoring LDL-C. Monitoring LDL-C is relatively expensive compared to the cost of statins, but there is limited evidence that it might improve adherence. I compared the costs (from a provider perspective) and outcomes (life years), of increasing statin doses based on regular measurement of LDL-C concentrations, to achieve a target LDL-C concentration of <1.8 mmol/L; prescribing atorvastatin 80 mg without LDL-C monitoring; and the status quo, simvastatin 20 mg without LDL-C monitoring. I constructed a Markov model with annual cycles; a five-year timeline; starting age of 60 years; and the following health states: ≤1 year after first cardiovascular event, ≤1 year after subsequent cardiovascular event, >1 year after any cardiovascular event, and dead. I estimated transition probabilities using published literature. I estimated the costs of hospitalisation for myocardial infarctions, strokes, unstable angina pectoris and coronary revascularisation procedures using health services utilisation and expenditure data from a sample of patients at a public sector hospital. I discounted costs and outcomes at 3% per year; and explored alternative scenarios and timelines in sensitivity analyses. Atorvastatin 80 mg without LDL-C monitoring, was both the cheapest and most effective option over a five-year period. It remained the most effective option over a lifetime period, but with an incremental cost-effectiveness ratio (ICER) of $146.94 per life year gained relative to the status quo. Treat to target was as effective as atorvastatin 80 mg if I assumed adherence rates of 80% and 60% respectively, but with an ICER of $54 930.96. Treat to target would dominate atorvastin 80 mg only if the frequency of LDL-C monitoring was reduced from 3-monthly to 6-monthly until targets were reached, and the cost of LDL-C monitoring decreased by $9.25 (84%). Fixed-dose statin treatment without cholesterol monitoring is the most cost-effective option for providing statins for the secondary prevention of cardiovascular disease. The costs of regular LDL-C monitoring currently make a treat to target strategy unaffordable in our setting. These results might be used to help guide policy regarding secondary prevention of cardiovascular disease in South Africa.

Grantor dc:publisher.institution

Health Economics Unit

Year dc:date.issued

2016

Author dc:creator

• De Waal, Reneé

Advisors dc:contributor.advisor

• Cleary, Susan
• Steyn, Krisela
• Levitt, Naomi S

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/11427/22752

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/22752