Summary Background Predictors for successful treatment are important for personalized medicine. Predictors for drug survival of biologics in psoriasis have been assessed, but not split for different biologics or for the reason of discontinuation. Objectives To compare long‐term drug survival between the outpatient biologics adalimumab, etanercept and ustekinumab in patients with psoriasis, and to elucidate predictors for overall survival and drug discontinuation due to ineffectiveness and side‐effects for each biologic separately. Methods Ten years of data were extracted from the prospective, multicentre, long‐term BioCAPTURE registry. Kaplan–Meier survival analyses and confounder‐corrected multivariate Cox regression analysis for drug survival (MCR‐DS) were performed to compare drug survival between biologics. To elucidate the predictors for different reasons of discontinuation for each biologic, univariate Cox regression analyses and multivariate Cox regression analyses for predictors (MCR‐P) with backward selection were performed. Results In total, 526 treatment episodes – 186 adalimumab, 238 etanercept and 102 ustekinumab – were included covering 1333 treatment years. MCR‐DS showed a significantly higher overall survival for ustekinumab compared with adalimumab and etanercept. MCR‐P showed that higher body mass index (BMI) was a predictor for discontinuation due to ineffectiveness for etanercept and ustekinumab and that female sex was a predictor for discontinuation due to side‐effects for adalimumab, etanercept and ustekinumab. Conclusions Ustekinumab has the highest confounder‐corrected long‐term drug survival in psoriasis treatment, compared with adalimumab and etanercept. Higher BMI is a predictor for discontinuation due to ineffectiveness in etanercept and ustekinumab, and female sex is a consistent predictor for discontinuation due to side‐effects in all three outpatient biologics. What's already known about this topic? Drug survival of biologics for psoriasis has been analysed, but few publications used prospective multicentre daily‐practice data. These studies found that ustekinumab had the highest confounder‐corrected overall drug survival, but they did not split survival analysis for the different biologics or for different reasons of discontinuation. Different predictors for overall drug survival were found in prospective and retrospective studies. What does this study add? This study reports on a longer period of drug survival than previous studies. Ustekinumab has a higher confounder‐corrected drug survival and higher survival for discontinuation due to ineffectiveness and side‐effects than adalimumab and etanercept. Higher body mass index (BMI) predicts discontinuation due to ineffectiveness in etanercept and ustekinumab, and female sex is a consistent predictor for discontinuation due to side‐effects in all three biologics. Linked Comment: Egeberg. Br J Dermatol 2016; 175:247–248 Plain language summary available online