Cardiovascular disease is the number one cause of death globally, and atherothrombosis is the underlying cause of most cardiovascular events. Several studies have shown that antiplatelet therapy, including aspirin (acetylsalicylic acid), reduces the risk of cardiovascular events and death. However, it is well-known that many patients experience cardiovascular events despite treatment with aspirin, often termed "aspirin low-responsiveness". This fact has caused considerable debate: does biochemical aspirin low-responsiveness have prognostic value? Can low-responders be reliably identified? And if so, should antithrombotic treatment be changed? Is the whole discussion of antiplatelet drug response merely a result of low compliance? Compliance should be carefully optimised, before evaluating the pharmacological effect of a drug. It is well-known that cardiovascular disease is multifactorial, and, therefore, total risk reduction is not feasible. Aetiological factors to the variable platelet inhibition by aspirin seem to include genetic factors, pharmacological interactions, smoking, diabetes mellitus, and increased platelet turnover. It is a captivating thought that antiplatelet therapy may be improved by individually tailored therapy based on platelet function testing. Ongoing studies are challenging the current one-size-fits-all dosing strategy, but the preceding evaluation of platelet function assays has not been adequate. The overall objective of this thesis was to evaluate the reproducibility of and aggreement between a number of widely used platelet function tests and to explore the importance of platelet turnover for the antiplatelet effect of aspirin in patients with coronary artery disease. In the intervention studies (studies 1, 3, and 4), optimal compliance was confirmed by measurements of serum thromboxane, which is the most sensitive assay to confirm compliance with aspirin. In study 1, platelet function tests widely used to measure the antiplatelet effect of aspirin were evaluated in healthy individuals and patients with coronary artery disease. Pharmaco-specific metabolites were measured in urine and serum to investigate the pharmacodynamic effect of aspirin and to enable the comparison with the more global tests of platelet function. Based on repeated duplicate measurements, we evaluated the reproducibility of each test. We found that reproducibility of the classical reference method was not impressive and that the newer, so-called point-of-care tests differed markedly on reproducibility. With coefficients of variation of about 3%, the VerifyNow Aspirin test was clearly the most reproducible test - even after correction of the official scale, which begins at about 350 aspirin reaction units and, therefore, results in artificially low coefficients of variation. Among the platelet function tests investigated, Multiplate was most sensitive for aspirin treatment. In study 2 we performed the hitherto largest study of newly released, immature platelets as a marker of platelet turnover. The study population included healthy individuals, patients with stable coronary artery disease, and patients with acute coronary syndromes. The main finding was an increased fraction of immature platelets in patients with ST-segment myocardial infarction, indicating an increased platelet turnover. Smoking and type 2 diabetes were identified as independent determinants of platelet turnover. In study 3 we explored the relationship between platelet turnover and the antiplatelet effect of aspirin in patients with stable coronary artery disease. The study results support the hypothesis that an increased platelet turnover reduces the antiplatelet effect of aspirin. The main findings were: 1) platelet turnover correlated with platelet aggregation measured by Multiplate and with sP-selectin, a marker of platelet activation. 2) Patients with diabetes mellitus type 2 had reduced antiplatelet effect of aspirin compared with patients without diabetes. 3) Widely used platelet function tests differ with respect to dependence on platelet parameters, including platelet count. 4) Smoking, diabetes mellitus type 2, and thrombopoietin were identified as independent determinants of platelet turnover. 5) The relative fraction of immature platelets has been employed in most previous studies, but in stable patients the absolute immature platelet count does not seem dependent on the total platelet count, and it has a stronger correlation with both platelet activation measured by sP-selectin and with platelet aggregation during treatment with aspirin. In study 4 we investigated platelet turnover and the antiplatelet effect of aspirin in a nested case-control study on patients with previous definite stent thrombosis. Patients with stent thrombosis were compared with patients without stent thrombosis, with whom they were matched at a 1:2 ratio with respect to risk factors for stent thrombosis: age, sex, stent type, and indication for percutaneous coronary intervention. The study showed that patients with previous stent thrombosis have reduced antiplatelet effect of aspirin and a tendency towards increased platelet turnover. In conclusion, widely used platelet function tests markedly differ on reproducibility, and the agreement between tests is relatively poor. An increased platelet turnover as suggested by the presence of newly formed immature platelets is important for the antiplatelet effect of aspirin, and, perhaps also for the development of acute coronary thrombosis. In the future, individually tailored antiplatelet therapy may potentially improve the benefit-risk ratio of antiplatelet therapy.