Recent studies suggest that epilepsy that is unresponsive to medical therapy is likely to be refractory from the onset. Identifying such patients early and treating them with nonpharmacologic therapies may improve their outcome. We hypothesized that patients who had adjunctive therapy with vagus nerve stimulation (VNS) earlier in the course of their epilepsy would have a better response compared with patients who had VNS therapy instituted later in the course. Patients in the VNS patient outcome registry who were more than 5 years post onset of their seizure disorder at implantation and had seizure frequency data available at both baseline and 3 months comprised the control group (n = 2785). These data were obtained retrospectively. Patients who were implanted between August 15, 2000 and July 31, 2001 who had epilepsy for 5 years or less at implantation or who had tried four or fewer standard antiepileptic drugs (AEDs) before implantation, and who were evaluated at baseline and at 3-month intervals for seizure frequency and quality of life, comprised the early adjunctive registry (EAR group; n = 120). This group was identified prospectively by participating physicians at multiple centers. The data describe patient demographics, medical history, seizure frequency, and physician-graded quality of life measures. The two populations were demographically similar except for statistically significant differences in age, duration of epilepsy, institutionalized patients, and seizure type (partial and generalized). Although the median reduction in seizure frequency for all patients at 3 months was similar between groups (48.2% control versus 50.0% EAR), 15.0% of the patients in the EAR group reported no seizures at 3 months compared with 4.4% of those in the control group (p < 0.001). In addition, significantly more patients in the EAR group (20% versus 8%; p < 0.001) reported no seizures with alteration or loss of consciousness, and 32% of EAR patients reported no complex partial seizures compared with 17% in the control group (p = 0.002). Improvements in all areas of quality of life were reported by both populations, but more patients in the EAR group were reported as "much better/better" for postictal state (p = 0.030) and seizure clustering (p = 0.002). Typically, 5% of patients report having no seizures after 3 months of VNS therapy. The proportion increased threefold, from 5% to 15%, for patients who received VNS therapy earlier in the treatment process. Patients reported even higher rates of no seizures when simple partial seizures were excluded from the analysis or when only complex partial seizures were considered. Although these results are preliminary, they offer promise of success in achieving seizure control among patients with refractory seizures who have been diagnosed with epilepsy for less than 5 years or who have tried four or fewer AEDs. We suggest future prospective studies evaluating VNS therapy versus best medical therapy after the first two to three AEDs have failed, which typically occurs within 2 years of seizure onset.