Résumé : Objectives This 5-year follow-up study aimed to assess clinical outcomes of HIV-1 infected adults treated with atazanavir (ATV) in clinical practice in Belgium, to describe patient profiles and characteristics, as well as treatment safety. Methods A multicenter, non-interventional, non-comparative, retrospective cohort study was performed in HIV-1 positive adult patients treated with ATV between 2006 and 2012. Data were collected from 8 AIDS reference centers' databases. All analyses were on-treatment. Sub-analyses were carried out in unboosted ATV treated patients and in females. The primary endpoint was defined as the time-to-treatment-discontinuation. Furthermore, virological suppression, immunological response, time to loss of virological response, reasons for ATV initiation, and discontinuation were also assessed. Results 2264 ARV-naive and ARV-experienced patients (median age: 41 years) were included. Females and non-Caucasians were broadly represented (40 and 45%, respectively). The probability to remain on treatment was 0.78 (CI: 0.76; 0.78) for the first and 0.69 (CI: 0.66; 0.71) for the second year and was similar between males and females. Overall, 771 patients (34.1%) discontinued ATV over time, the median (Q1-Q3) time to discontinuation being 0.8 (0.3-1.5) year. In unboosted ATV-treated patients, results were comparable to the overall ATV population, except for a higher rate of discontinuation-over-time (45.1%). Conclusions Clinical and safety data from this 5 year-cohort study show that the vast majority of patients remained on ATV treatment for the first and second years, overall as well as patients treated with unboosted ATV and females.