Background: Regular paracetamol is the recommended first-line analgesic for acute low-back pain; however, no high-quality evidence supports this recommendation. We aimed to assess the efficacy of paracetamol taken regularly or as-needed to improve time to recovery from pain, compared with placebo, in patients with low-back pain. Methods: We did a multicentre, double-dummy, randomised, placebo controlled trial across 235 primary care centres in Sydney, Australia, from Nov 11, 2009, to March 5, 2013. We randomly allocated patients with acute low-back pain in a 1:1:1 ratio to receive up to 4 weeks of regular doses of paracetamol (three times per day; equivalent to 3990 mg paracetamol per day), as-needed doses of paracetamol (taken when needed for pain relief; maximum 4000 mg paracetamol per day), or placebo. Randomisation was done according to a centralised randomisation schedule prepared by a researcher who was not involved in patient recruitment or data collection. Patients and staffat all sites were masked to treatment allocation. All participants received best-evidence advice and were followed up for 3 months. The primary outcome was time until recovery from low-back pain, with recovery defined as a pain score of 0 or 1 (on a 0-10 pain scale) sustained for 7 consecutive days. All data were analysed by intention to treat. This study is registered with the Australian and New Zealand Clinical Trial Registry, number ACTN 12609000966291. Findings:550 participants were assigned to the regular group (550 analysed), 549 were assigned to the as-needed group (546 analysed), and 553 were assigned to the placebo group (547 analysed). Median time to recovery was 17 days (95% CI 14-19) in the regular group, 17 days (15-20) in the as-needed group, and 16 days (14-20) in the placebo group (regular vs placebo hazard ratio 0.99, 95% CI 0.87-1.14; as-needed vs placebo 1.05, 0.92-1.19; regular vs as-needed 1.05, 0.92-1.20). We recorded no difference between treatment groups for time to recovery (adjusted p=0.79). Adherence to regular tablets (median tablets consumed per participant per day of maximum 6; 4.0 [IQR 1.6-5.7] in the regular group, 3.9 [1.5-5.6] in the as-needed group, and 4.0 [1.5-5.7] in the placebo group), and number of participants reporting adverse events (99 [18.5%] in the regular group, 99 [18.7%] in the asneeded group, and 98 [18.5%] in the placebo group) were similar between groups. Interpretation: Our findings suggest that regular or as-needed dosing with paracetamol does not affect recovery time compared with placebo in low-back pain, and question the universal endorsement of paracetamol in this patient group. Funding: National Health and Medical Research Council of Australia and GlaxoSmithKline Australia.