Background The aim of this study was to evaluate the efficacy of acupuncture in treating non-specific acute low back pain, and to identify the different components of the therapeutic effect of the technique. Methods Randomized controlled multicentre prospective study, with four Branches, aimed at comparing real semi-standardized acupuncture at specific Points Group A), acupuncture at non-specific points (Group B), placebo acupuncture (Group C) and conventional treatment (analgesics, antirugs, inflammatory drugs, myorelaxants, postural advice) (Group D). The patients were blinded for the specific, non-specific and placebo acupuncture treatments. Evaluation of the patients and analysis of the results were carried out by experts who were blinded as regards the allocation to the different treatment groups. The patients selected were of working age and presented symptoms of acute non-specific low back pain; they had been referred by their GPs at Primary Healthcare Clinics to one of the four clinics participating in the study, all of which belonged to the Andalusian Public Health System (Spain). The participants were evaluated before randomization and at the end of weeks 3, 12 and 48 from the start of treatment. The primary outcome measure was significant clinical improvement (SCI) at 3 weeks after randomization, defined as a reduction of 35% of more in the RMQ score. Results 275 patients were randomized; mean duration of 6 days of the low back pain episode. 59% were women, with a mean age of 42.7 years. 75% of the participants were in paid work, and over 50% of these said they were satisfied with it. The patients suffered a high mean baseline pain intensity level (70.4 mm on a scale of 100 mm), an index of fear and avoidance conduct related to physical activity of 18.8 (SD 5.6) and a corresponding work-related index of 24.3 (SD 9.8). For 44% of the patients, this was their first episode of back pain, while 61.7% were on sick leave due to the back pain being suffered at the outset of the study. Over half of the participants (57.8%) remarked that their pain was work related. By ITT analysis, the SCI at 3 weeks from the start of the treatment was obtained for 77.1% of the participants in Group A, for 73.5% of those in Group B, for 62.3% of those in Group C, and for 42.9% of those in Group D. The differences between Groups A, B and C with respect to Group D were statistically significant, as were those between Groups A and C (see table). Conclusion Acupuncture, both at specific points and at non-specific ones or as a placebo, achieves a rate of significant clinical improvement that is superior to that obtained by conventional treatment for acute non-specific back pain, and this difference is statistically significant. No differences were found between the different types of acupuncture, as regards specific or non-specific points, for this type of pain.