Knowledge of cardiovascular risk factors is increasing. At the same time, risk estimation becomes more and more difficult. The need for a more comprehensive, but more individually based approach is evident. To achieve this aim, we propose a systems biology approach in cardiovascular risk assessment. This means that a large amount of health data, describing many aspects of the health-status of patients, is collected and computed and the results are compared with existing knowledge. Finally, a clinical model is created, which can be the first step in ongoing research protocol, aimed at assessing cardiovascular risk. By using this approach, all potentially relevant risk factors can be identified on a small sample. Moreover, risk groups can be more specifically defined, based on the "natural" clustering of data, according to their predictive load. We tested this possibility on an example of hyperhomocysteinemia which is a well-known complex cardiovascular risk factor.