To describe a generalizable stochastic-simulation model for schizophrenia treatment related with the cardiovascular associate risk of SGA.

A model to simulate the expected 10-year occurrence of all-type cardiovascular events (CVE) in a hypothetical cohort of 100.000 patients with schizophrenia treated with SGA drugs in Spain was developed. The model considered, as a baseline health state, outpatient treated with SGA with characteristics of patients enrolled in the CLAMORS study; a cross-sectional study in schizophrenia spectrum disorders aimed to ascertain prevalence of metabolic syndrome in such patients together with CHD. Three other states were considered: suffering a CVE, death due to CVE and death due to other causes. The CVE risk for each SGA drugs was estimated through a locally-adjusted Framingham risk equation. Treatment outcomes were simulated using the expected mean change of the cardiovascular (CV) risk factors from the CATIE clinical trial. Death by CVE or others causes were estimated from published literature.

The 10-year rate of CVE following SGA treatment was 0.181, 0.179, 0.176 and 0.172 for olanzapine, quetiapine, risperidone and ziprasidone, respectively. Relative risk was calculated relative to no-treatment, and the corresponding values were 1.03, 1.02, 1.00 and 0.97. The total estimated CVE were 25,269 events; 25,157; 24,883 and 24,514, respectively.

A generalizable, flexible model was developed through stochastic simulation of the CV risk for SGA drugs. The estimated clinical outcomes suggest different levels of CVE risk for each SGA drugs. Ziprasidone showed the lower rate with no association with increased risk for CHD.