One-way sensitivity analyses were carried out on the following to test the cost per life year gained:
(a) reduction in CHD and stroke according to the 95% confidence interval in the meta-analysis;
(b) pre-treatment risk;
(c) values of risk factors;
(d) survival after different disease events;
(e) disease costs and treatment costs;
(f) quality of life assumptions were tested in the sensitivity analysis (but not as part of the main analysis) only to determine the direction of change when quality of life was included;
(g) discount rate varied between 0% and 10%.