OBJECTIVES: Topical steroids are the corner-stone of current treatment for atopic dermatitis (AD), a chronic fluctuating inflammatory skin disease. However, steroids carry a risk of local and systemic side effects limiting their long term use and effectiveness. The objective was to assess the incremental cost-effectiveness ratio (ICER) of the new topical immunomodulator Tacrolimus in moderate to severe AD. Tacrolimus has shown significant clinical improvement, maintained with long term intermittent treatment up to four years. METHODS: A Markov model was developed in MS-Excel. Model health states represent severe, moderate, mild, and virtually cured AD as defined by the Eczema Area and Severity Index (EASI). Based on prevalence data, 82% start with moderate, 18% with severe AD. The model simulates monthly severity fluctuations. Transitions among health states were calculated from two 1-year observational trials (Tacrolimus n = 93, current care n = 120). Tacrolimus consumption was obtained from the clinical trial, other resource utilisation from a two-round Delphi consensus panel (n = 8). Unit costs from the Belgian health care payers perspective were applied. Effects are expressed in 'disease controlled days', defined as days with mild or virtually cured AD. The time horizon was from 1 (basecase) to 3 years. RESULTS: By the end of the first year, Tacrolimus increases the number of 'disease controlled days' with 59 days at an IC of €60, implying an ICER of 1€/disease controlled day. Better ICERs are obtained in severe AD, when taking into account UV light therapy costs in the steroid arm, currently not reimbursed in Belgium. Due to the long term clinical improvement, the ICER decreases with treatment duration and Tacrolimus ointment is estimated to become cost-saving as from 2 years treatment duration. CONCLUSION: Tacrolimus is a cost-effective treatment for moderate to severe AD, providing potential savings to the health care system during long term maintenance treatment.