OBJECTIVES:To determine whether individualised manual therapy plus guideline-based advice results in superior outcomes to advice alone in participants with clinical features potentially indicative of lumbar zygapophyseal joint pain. DESIGN:Multi centre parallel group randomised controlled trial. SETTING:14 physiotherapy clinics in Melbourne, Australia. PARTICIPANTS:Sixty-four participants with clinical features potentially indicative of lumbar zygapophyseal joint pain. INTERVENTIONS:10-weeks of physiotherapy comprising individualised manual therapy based on pathoanatomical, psychosocial and neurophysiological barriers to recovery plus guideline-based advice (10 sessions) or advice alone (two sessions). MAIN OUTCOME MEASURES:Primary outcomes were activity limitation (Oswestry Disability Index), and separate 0 to 10 numerical rating scales for leg pain and back pain. Measures were taken at baseline and 5, 10, 26 and 52-week. RESULTS:Between-group differences for back pain favoured individualised manual therapy over advice for back pain at 5 (1.0; 95% CI 0.6 to 2.0), 10 (1.5; 95% CI 0.5 to 2.4) and 26-weeks (1.4; 95% CI 0.4 to 2.3) as well as for activity limitation at 26 (8.3; 95% CI 2.6 to 14.2) and 52-weeks (8.2; 95% CI 2.3 to 14.2). There were no significant between-group differences for leg pain. Secondary outcomes and responder analyses also favoured individualised manual therapy at almost all time-points. CONCLUSIONS:In participants with clinical features potentially indicative of lumbar zygapophyseal joint pain, individualised manual therapy led to greater reduction in back pain at 5, 10 and 26-week follow-up as well as activity limitation at 26 and 52-weeks. Between-group differences were likely to be clinically important. TRIAL REGISTRATION:ACTRN12609000334202.