The objective of the study was to develop a clinical prediction rule for identifying patients with low back pain, who improved with mechanical lumbar traction. A prospective, cohort study was conducted in a physiotherapy clinic at a local hospital. Patients with low back pain, referred to physiotherapy were included in the study. The intervention was a standardized mechanical lumbar traction program, which comprised three sessions provided within 9 days. Patient demographic information, standard physical examination, numeric pain scale, fear-avoidance beliefs questionnaire and Oswestry low back pain disability index (pre-and post-intervention) were recorded. A total of 129 patients participated in the study and 25 had positive response to the mechanical lumbar traction. A clinical prediction rule with four variables (non-involvement of manual work, low level fear-avoidance beliefs, no neurological deficit and age above 30 years) was identified. The presence of all four variables (positive likelihood ratio = 9.36) increased the probability of response rate with mechanical lumbar traction from 19.4 to 69.2%. It appears that patients with low back pain who were likely to respond to mechanical lumbar traction may be identified.