Background: Low back pain (LBP) is a major health problem. Effective treatment of acute LBP is important because it prevents patients from developing chronic LBP, the stage of LBP that requires costly and more complex treatment. Physiotherapists commonly use a system of diagnosis and exercise prescription called the McKenzie Method to manage patients with LBP. However, there is insufficient evidence to support the use of the McKenzie Method for these patients. We have designed a randomised controlled trial to evaluate whether the addition of the McKenzie Method to general practitioner care results in better outcomes than general practitioner care alone for patients with acute LBP. Methods/design: This paper describes the protocol for a trial examining the effects of the McKenzie Method in the treatment of acute non-specific LBP. One hundred and forty eight participants who present to general medical practitioners with a new episode of acute non-specific LBP will be randomised to receive general practitioner care or general practitioner care plus a program of care based on the McKenzie Method. The primary outcomes are average pain during week 1, pain at week 1 and 3 and global perceived effect at week 3. Discussion: This trial will provide the first rigorous test of the effectiveness of the McKenzie Method for acute non-specific LBP.