OBJECTIVE: This study aimed to assess the long-term outcome and prognostic factors of vagus nerve stimulation (VNS) for drug-resistant epilepsy (DRE) using real-world data. METHOD: We included 189 DRE patients who underwent VNS implantation between 2005 and 2018 at nine national hospitals in Korea. Seizure-frequency data obtained quarterly one year before and after surgery and annually up to four years after surgery were collected from medical records. Health resource utilization trends over the four years preceding and following surgery were assessed through linkage with national health insurance data. We performed interrupted time series analysis to examine the trend in seizure frequency before and after one year following surgery. RESULTS: The seizure frequency exhibited a decreasing trend in 27.5 % and an increasing trend in 3.8 % during the first year following VNS implantation without a significant change in efficacy over the subsequent three years. Patients with focal spikes with secondary bilateral synchrony (SBS) in electroencephalography had a higher responder rate (adjusted odds ratio (aOR)= 3.06 [1.36-6.90]), whereas those with Lennox-Gastaut syndrome had a lower responder rate (aOR=0.38 [0.15-0.94]). One-year seizure-freedom was achieved in 6.0 % of patients at some point during the four-year follow-up. Over an eight-year period, the number of antiseizure medications (ASMs) tended to increase before surgery and remained at a median of 5 [4-6] after surgery. While the total medical and epilepsy-related costs tended to decrease after surgery, the ASM cost continued to increase. CONCLUSION: VNS was substantially beneficial for one in four patients with DRE, offering the chance of seizure-freedom. However, the efficacy of VNS fell within the efficacy range of recently introduced medical treatments and did not lead to a decrease in the ASM burden. Focal spike with SBS is a potential biomarker for a favorable response to VNS.