Objectives: To study the predictors of absenteeism in adult patients treated for moderate-to-severe atopic dermatitis in Portugal. Methods: Data were drawn from a cross-sectional study conducted in community pharmacies designed to characterize the adult population being treated for moderate-to-severe atopic dermatitis (AD) in Portugal and AD related absenteeism. The questionnaire was self-administered but the data on AD medications was provided by pharmacists. A zero-inflated negative binomial model was used to estimate the predictors of absenteeism related to AD. Results: From the 622 eligible patients, 66.2% were employed, from which 81.8% reported not having missed work due to AD in the previous 6 months. The mean number of absenteeism days was 0.88, with a standard deviation of 3.88. Results of the zeroinflated negative binomial model reveal that patients followed in private setting and patients followed by general practitioners had a higher chance of not missing work. The second part of the model indicates that patients with higher levels of embarrassment and patients on systemic therapy have more days of absenteeism due to AD. Combining both models, we estimated that patients with high levels of embarrassment miss on average 1.46 (CI 90%: 0.41-2.51) days more than patients that don't report any level of embarrassment. Patients who are on systemic therapy miss on average 0.77 (CI 90%: 0.11-1.44) days more than those that are not on systemic therapy. Conclusions: AD is a disease that affects daily life, resulting in productivity losses of patients. The results indicate that the level of embarrassment and the type of medication are predictors of the absenteeism in patients with moderate-to-severe AD.