OBJECTIVES: To analyse the cost-effectiveness of CAD in breast cancer diagnostic in comparison to normal procedure from the perspective of statutory health insurance (SHI). METHODS: To compare the effectiveness with and without CAD, total costs of diagnostic measures were calculated by a Markov-model. Model structure, transition possibilities, procedures within therapies and complications were ascertained by a Delphi-panel. Subsequently, costs of therapy per patient with and without CAD were calculated. Fur-thermore, costs of successive therapy of undetected cancer without CAD were consid-ered. Based on literature, an increase of 19.5% in detecting breast cancer with CAD was determined. Moreover the assumption was made, that with CAD 19.5% of cancers could be detected at an earlier stage. RESULTS: Based on perspective of the SHI, diagnostic and therapy of 10,000 mammography patients from the Markov-cohort caused total costs in amount of 2,298,048€ without CAD (229.80€ per patient) and 2,352,635€ with CAD (235.26€ per patient). By consideration of the effectiveness parameter (number of detected breast cancers per 10,000 patients, 0.01912 without CAD, 0.02285 with CAD), the effectiveness-adjusted costs amounted to: without CAD 12,019€, with CAD 10,296€. Thus, the implementation of CAD proves to be more cost-effective due to a higher sensitivity of the diagnostic procedure.Subsequently, two sensitivity-analyses were conducted to test robustness of this model for cost effectiveness and for costs per patient relative to the price for CAD. CONCLUSIONS: Diagnosis costs per patient are higher with CAD compared to normal procedure. However, more breast cancers can be detected and treated at an earlier stage. Therapeutic costs per patient are lower; therefore implementation of CAD is more cost-effective. As far as Germany is concerned, 2.691 additional breast cancers can be detected every year if CAD would be included in breast cancer diagnosis.