OBJECTIVES:  The objective was to examine the importance of treatment attributes on patient preferences for GLP-1 RAs among GLP-1-experienced and injection-naïve T2DM patients in Germany. METHODS:  Patients recruited through a panel completed an online, discrete choice experiment survey. Patients chose between hypothetical, blinded treatment profiles reflecting attributes of GLP-1 RAs. Sample consisted of injection-naïve T2DM patients with oral-only experience or those currently using exenatide QW or liraglutide QD. Survey included eight attributes: efficacy, SEs, device size, needle size, titration, injection preparation, evidence of long-term efficacy/safety, and dosing frequency (daily or weekly). Odds ratios (OR) and 95% CIs were calculated using a conditional logit model indicating the likelihood of choosing a treatment with a given attribute level versus a reference attribute level. RESULTS:  586 patients were surveyed, including injection-naïve patients (n=296, 51%) and GLP-1RA patients (n=290, 49%; 147 exenatide QW & 143 liraglutide QD). Among the injection-naïve group, mean age was 58.8 years, 70% were male, and mean time since diagnosis was 7.7 years. SEs (OR for level reflecting exenatide QW vs liraglutide QD=2.39, 95% CI=2.23-2.56; p<.001), efficacy (OR for -1.5% A1c vs -0.8%=1.92, 95% CI=1.73-2.14; p<.001), and dosing frequency (OR for weekly vs daily=1.71, 95% CI=1.60-1.83; p<.001) were most important. Among injection-experienced patients, mean age was 60.0 years, 50% were male, and mean time since diagnosis was 5.13 years; efficacy (OR for -1.5% A1c vs -0.8%=3.27, 95% CI=2.90-3.70; p<.001), SEs (OR for level reflecting exenatide QW vs liraglutide QD=2.91, 95% CI=2.68-3.15; p<.001), then dosing frequency (OR for weekly vs daily=2.83, 95% CI=2.60-3.07; p<.001) were most important. Needle size, device size, and titration were not important drivers of patients’ preference. CONCLUSIONS: Among injection-naïve and GLP-1 experienced patients, key drivers of treatment preference were SEs, efficacy, and dosing frequency. Understanding patient preferences is important for optimizing treatment decision-making and improving treatment adherence.