Treatment of coronary bifurcation lesions remains challenging; a simple strategy has been preferred as of late, but the disadvantage is ostium stenosis or even occlusion of the side branch (SB). Only a few single-center studies investigating the combination of a drug-eluting stent in the main branch followed by a drug-eluting balloon in the SB have been reported. This prospective, multicenter, randomized study aimed to investigate the safety and efficacy of a paclitaxel-eluting balloon (PEB) compared with regular balloon angioplasty (BA) in the treatment of non-left main coronary artery bifurcation lesions. Between December 2014 and November 2015, a total of 222 consecutive patients with bifurcation lesions were enrolled in this study at ten Chinese centers. Patients were randomly allocated at a 1:1 ratio to a PEB group (n = 113) and a BA group (n = 109). The primary efficacy endpoint was angiographic target lesion stenosis at 9 months. Secondary efficacy and safety endpoints included target lesion revascularization, target vessel revascularization, target lesion failure, major adverse cardiac and cerebral events (MACCEs), all-cause death, cardiac death, non-fatal myocardial infarction, and thrombosis in target lesions. The main analyses performed in this clinical trial included case shedding analysis, base-value equilibrium analysis, effectiveness analysis, and safety analysis. SAS version 9.4 was used for the statistical analyses. At the 9-month angiographic follow-up, the difference in the primary efficacy endpoint of target lesion stenosis between the PEB (28.7% ± 18.7%) and BA groups (40.0% ± 19.0%) was -11.3% (95% confidence interval: -16.3% to -6.3%, Psuperiority <0.0001) in the intention-to-treat analysis, and similar results were recorded in the per-protocol analysis, demonstrating the superiority of PEB to BA. Late lumen loss was significantly lower in the PEB group than in the BA group (-0.06 ± 0.32 vs. 0.18 ± 0.34 mm, P < 0.0001). For intention-to-treat, there were no significant differences between PEB and BA in the 9-month percentages of MACCEs (0.9% vs. 3.7%, P = 0.16) or non-fatal myocardial infarctions (0 vs. 0.9%, P = 0.49). There were no clinical events of target lesion revascularization, target vessel revascularization, target lesion failure, all-cause death, cardiac death or target lesion thrombosis in either group. In de novo non-left main coronary artery bifurcations treated with provisional T stenting, SB dilation with the PEB group demonstrated better angiographic results than treatment with regular BA at the 9-month follow-up in terms of reduced target lesion stenosis. ClinicalTrials.gov, NCT02325817; https://clinicaltrials.gov.