Introduction: Striking a balance between maximizing the effectiveness of corticosteroids while minimizing its adverse effects in Systemic Lupus Erythematosus (SLE) treatment requires careful consideration of how SLE patients value improvements in their disease versus risks of treatment adverse effects. Hence, this study aimed to elicit SLE patients' benefit-risks preferences for corticosteroid use, and to examine variation in preferences. Methods: SLE patients with current/past experience with using corticosteroids were recruited from the Johns Hopkins Lupus Center, the University of Maryland Medical Center and PatientsLikeMe. Study participants completed a web-based survey that comprised three sections: 1) sociodemographic characteristics, 2) a discrete choice experiment (DCE) measuring benefit-risks preferences, and 3) self-rated disease severity using the Systemic Lupus Activity Questionnaire (SLAQ). The DCE assessed two attributes of treatment benefit (time between flares and activities limitations), and four attributes of treatment risks relating to corticosteroids (heart-related damage, bone fractures, weight gain, and unusual mood swings/irritability). Preference weights were estimated using a conditional logit model and used to calculate the relative importance scores for each attribute and maximum acceptable risk that patients would tolerate. We also evaluated variations in preferences using a latent class analysis, and a stratified analysis based on SLAQ scores. Results: When making trade-offs between the risks and benefits of corticosteroid use, our 141 study participants (95% Female, 61% White) viewed weight gain as the most important, followed by heart-related damage and activities limitations. SLE patients were willing to tolerate higher risks for improvements in activities limitations (4.5% risk of heart damage) compared to improvements in time between flares (3.1% risk of heart damage). Our latent class analysis demonstrated the relative importance of three key attributes varied significantly across patient segments: activities limitations, weight gain, and heart damage. Individuals with SLAQ score ≤ 21 (N=75) were more willing to accept treatment risks in exchange for improvements in activity limitations compared to individuals with SLAQ score > 21 (N=66). Conclusions: Our study sheds light on SLE patients' benefit-risk trade-offs for corticosteroid use. This information can be used to guide corticosteroid therapy in this population.