Completed Economic Analysis Worksheet for Evidence-Based Purchasing
Johannesson M, JÃ¶nsson B, Kjekshus J, Olsson AG et al. Cost effectiveness of simvastatin treatment to lower cholesterol levels in patients with coronary heart disease. New England Journal of Medicine. 1997;336:332-336.
Are the results of this economic analysis valid?
Did the analysis provide a full economic comparison of health care strategies?
Kind of. Direct health costs as well as lost income from illness were included. Costs not included were explained (quality of life costs, extra health costs from years of life gained). However, the study did not compare costs with status quo. It only broadly compared the costs of treating cholesterol with statins with those of treating other common conditions.
Were the costs and outcomes properly measured and valued?
Yes. The Scandinavian Simvastatin Survival (4S) Study was used for effectiveness information, itself a reasonably reliable multicentre randomised controlled trial. Follow up was 5 years. The unit being assessed were years of life in people with existing heart disease taking simvastatin. Costs were assessed using hospital cost data and income from patients' work status, recorded every 6 months (page 334). The discount rate was 5%.
Was appropriate allowance made for uncertainties in the analysis?
Yes. Sensitivity analysis (table 5) was conducted to see what effects different assumptions might have on the results, including assumptions about non-health costs, follow up costs and differences between Swedish and US drug prices.
Are estimates of costs and outcomes related to the baseline risk in the treatment population?
Yes - Table 1.
Are the valid results of this economic analysis important?
What were the incremental costs and outcomes of each strategy?
The authors combine Swedish direct and indirect costs for first and subsequent years at a discount rate of 5%. Costs are calculated at the 1995 exchange rate with the $US for different aspects of care for three age groups and both sexes, made explicit (for comparison in other settings) in Tables 2-5.
Do incremental costs and outcomes differ between subgroups?
Yes. Costs were calculated for different risk groups. Older people (who have not so many years of life to gain) and women (at lower risk of events) are more expensive.
How much does allowing for uncertainty change the results?
See sensitivity analysis - Table 5. Different assumptions change the range of costs a lot, but not so much as to reverse the conclusions.
Can you apply this valid, important evidence about prognosis in caring for your patient?
Are the treatment benefits worth the harms and costs?
Probably - although it would be important for purchasers to consider the costs within their own country.
Could my patients expect similar health outcomes?
Probably, unless the rate of coronary heart disease in the population is much lower (or higher) than those reported here, in which case the cost effectiveness per year would be lower (or higher).
Could I expect similar costs?
In Western Europe, probably. US (more expensive) drug prices and health care costs would need to be calculated from the Tables given to see whether the cost effectiveness was still as great. Table 5 includes a sensitivity analysis for US drug prices.
The authors note that the results only apply to a high risk population - i.e. those with a history of heart disease - as the cost effectiveness falls as absolute risk of cardiac events falls.