474 patients with a lumbar disc herniation were randomized to either undergo surgical treatment via a lumbar discectomy or a nonoperative intervention in order to assess long term outcomes and determine outcome predictors. Patients who refused randomization were enrolled in an observational cohort and contributed follow-up data based on subsequent treatment. Between the randomized groups, differences after 8 years in primary outcomes of pain and function did not significantly differ between operatively and nonoperatively managed patients, though secondary outcomes of leg pain, the Sciatica Bothersome Index, and patient-reported satisfaction and perceived improvement favoured the operative group. In the analysis of possible predictive factors; sequestered fragments, high intensity pain for a period of more than 6 months and a work status including students, homemakers, unemployed, or retired were characteristics suggestive of greater benefit with surgery.