To systematically review the evidence for health coaching for patients with low back pain and describe the diversity of health coach training and interventions.Electronic databases Medline, CINAHL, EMBASE, PsycINFO, AMED and the Cochrane Central Register of Controlled Trials were searched to 24 June 2013 using individually devised strategies. Randomised or quasi randomised controlled trials (RCTs) of health coaching for adults with low back pain of any duration were considered. The overall quality of the body of evidence was assessed using the GRADE approach. Treatment effects were presented as the difference in mean scores with 95% confidence intervals and standardised mean difference at follow-up between health coaching and control groups. Health coaching interventions were compared narratively by their theoretical principles and the training and assessment of heath coaches.Five publications describing three RCTs and one cluster RCT met the criteria for review. A rating of very low was assigned to the body of evidence using the GRADE approach. One RCT found significant improvements in lifting capacity and exercise compliance in favour of the health coaching group at both follow-up points with a large and moderate standardised mean difference. All included studies based health coaching interventions on the transtheoretical model of change however, the content of counselling programmes varied between studies and measures of treatment fidelity were inconclusive.Variability in health coaching interventions and a lack of assessment of treatment fidelity in addition to the very low rating of the overall body of evidence identified in the current review renders any estimates of the effect of health coaching on low back pain uncertain. Well-designed RCTs of patients with sub-acute low back pain are required that incorporate clearly described protocols for health coaching interventions and include standardised measures of treatment fidelity.