Symptom-provoking active movements can be useful assessments of patients with low back pain. They may give an early indication of the likelihood of chronicity and can also assist the clinician in selecting treatment interventions. Reliability data for such movements is incomplete. While symptom-provoking flexion and extension movements have consistently shown acceptable reliability and results for sidebend tests have suggested moderate reliability, results for symptom-provoking active rotation are variable and only one study has investigated the reliability of combined movements. The purpose of this study was to further investigate the interexaminer reliability of symptom-provoking active sidebend, rotation and combined movement assessments. Symptom-provoking movements were identified on the basis of aggravation or reproduction of the patient's pain complaint, that is, the symptoms for which the patient was seeking treatment. Pairs of therapists from a pool of four therapists independently examined patients with low back pain during a routine clinic visit. Percent agreement and Kappa values (95% Confidence Interval) for the movement assessments were sidebend 81.4% and 0.60 (0.40; 0.79), rotation 70% and 0.17 (-0.08; 0.42), sidebend-rotation 64.3% and 0.29 (0.06; 0.51), flexion-sidebend-rotation 70% and 0.39 (0.18; 0.61), and extension-sidebend-rotation 67.1% and 0.29 (0.06; 0.52). The majority of schemes described for the physical examination of patients with low back pain include symptom-provoking active sidebend, and findings from this study support the ongoing clinical use of this assessment. While symptom-provoking active rotation and combined movements were found to be unreliable, these assessments have been shown to have potential diagnostic value. Further research is warranted to investigate procedural changes that may improve reliability and to further investigate their diagnostic accuracy.