OBJECTIVES: Atazanavir/r (ATV/r) and darunavir/r (DRV/r) are commonly used to suppress HIV in Germany. Compared to lopinavir/r, both have demonstrated similar relative efficacy in randomized clinical trials (RCT) in previously untreated patients; whereas annual drug acquisition costs are 250 euro more for DRV/r. However, the objective was to estimate how these would translate into costs and effects, after 5 and 25 years. METHODS: The cost-effectiveness was forecasted using a microsimulation-model (monthly cycles) for previously untreated HIV-patients Response to medication was modeled as a reduction in HIV-RNA viral-load. In accordance with literature, increase in CD4+ could persist up to 5 years while having a response; whereas no response resulted in the opposite. The occurrences of AIDS defining events and non-AIDS defining malignancies were linked to the current CD4+ and independent of therapy. Adverse events (AE) on the other hand were treatment-specific and the following were considered: 1) diarrhea; 2) dizziness; 3) jaundice; 4) nausea; and 5) rash. Toxicities associated with the long-term use of agents considered in the model included: 1) cardiovascular events; 2) renal insufficiency; and 3) hepatic failure. Drug efficacy and AE incidences were based on published RCTs - compared via an indirect comparison - whereas, other data was based on published literature.  RESULTS: After 5 years, initiation with ATV/r use was found to be associated with a lower total cost compared to DRV/r. In the long-term (i.e. 25 year time horizon) the use of ATV/r was found to be associated with an increase in survival of 16.0 (discounted) years; 0.31 life years and 0.55 QALYs compared to DRV/r. This resulted in an ICUR of € 11,241 per QALY gained. Results were most sensitive to changes in virological parameters, market shares for future lines and cost per CD4-level. CONCLUSIONS: The model forecasts ATV/r to be cost-effective compared to DRV/r in Germany.