Background: Cardiovascular diseases are among the most common non-communicable diseases and are one of the leading causes of death worldwide. In addition, they cost the EU around €282 billion annually. To facilitate early detection and prevention of cardiovascular risk, risk scores are used internationally to estimate an individual's risk of disease over a 5- to 10-year period. In Austria, the ARRIBA tool is currently in use but has not been officially integrated into the preventive medical check-up (PMCU).

Methods: In accordance with PRISMA guidelines, three databases (Medline, Embase, The Cochrane Library) were searched for systematic reviews published between 2015 and 2025. Seven systematic reviews, one validation study, and five guidelines identified through manual search, were included. Quality assessment was conducted using ROBIS, PROBAST, and AGREE-II.

Results: The predictive accuracy of the scores varies considerably across populations. In the general population, most scores demonstrate moderate discriminatory ability, with a possible over- or underestimation of risk. In populations with specific pre-existing conditions (e.g. rheumatoid arthritis, type 1 diabetes), risks are frequently underestimated. Evidence for clinical benefit remains limited. All parameters required for the scores are already collected within the PMCU; however, implementation requires specialised software, regional calibration, training, and regular recalibration of the models. European guidelines recommend risk assessment in apparently healthy adults, though recommended scores, age thresholds, and treatment thresholds vary.

Discussion: The available evidence shows no consistent indication of improved long-term outcomes through the use of risk scores. The generalisability of models to different population groups is limited, and the evidence base remains heterogeneous. Nevertheless, risk scores can serve as tools for risk communication and decision support in a preventive medicine context.

Conclusion: The available data are insufficient to clearly recommend a specific score. No score has been validated in Austrian populations. SCORE2 with European calibration appears most suitable, with ARRIBA as an alternative. Further evaluation is required for widespread implementation.