Objective To investigate the proportion of patients with low back pain who receive an opioid analgesic prescription on hospital discharge, the proportion using opioid analgesics 4 weeks after discharge, and to identify predictors of continued opioid analgesic use at 4 weeks after an ED presentation in opioid‐naïve patients. Methods An observational cohort study nested within a randomised controlled trial in four EDs in New South Wales, Australia. Participants were adults who presented to the ED with non‐specific low back pain or low back pain with lower limb neurological signs and symptoms. Electronic medical records supplemented the patient‐reported pain and use of opioid analgesics at 4‐week follow up. Results Of the 104 patients included, 33 (31.7%, 95% confidence interval [CI] 22.9–41.6) received an opioid analgesic prescription at hospital discharge and 38 (36.5%, 95% CI 27.3–46.6) reported taking an opioid analgesic for pain 4 weeks after the ED presentation. Among opioid‐naïve patients (n = 85), older age (odds ratio [OR] 1.04, 95% CI 1.00–1.08, P = 0.031) was the only predictor for continued opioid analgesic use at 4 weeks post‐ED presentation. Conclusion About one‐third of patients who present to the ED with low back pain receive an opioid analgesic prescription on discharge and are taking an opioid analgesic 4 weeks later. These findings justify future research to identify strategies to reduce the risk of long‐term opioid use in patients who present to the ED with low back pain.