MRI’s are often prescribed by doctors and specialists to help determine “the cause of your back pain”.
Research clearly shows (Kent and Keating, 2004; Chou, 2007) that there is often a poor correlation between people’s clinical presentation and the scan results. In low back pain sufferers research shows up to 85% of people have an MRI that does not match their clinical presentation (their history about what aggravates and eases their symptoms, combined with their mechanical signs).
Many people with no low back pain show significant degenerative changes and multiple levels of disc bulging. Other people with disc problems can show a “normal MRI” as the disc protrusion does not reveal itself in the MRI taken in the lying position.
The best approach is a through MDT (mechanical diagnosis and therapy, also known as McKenzie Method) evaluation. Those that respond are rapidly identified, and importantly those that will not respond are quickly identified also, usually within the first session. Often the need for expensive imaging is avoided, and given to those that truly need it who are not responding to any form of conservative management.

References:
Chou, R. (2007) ‘Diagnosis and Treatment of Low Back Pain: A Joint Clinical Practice Guideline from the American College of Physicians and the American Pain Society’, Annals of Internal Medicine, 147(7), p. 478. doi: 10.7326/0003-4819-147-7-200710020-00006.

Kent, P. and Keating, J. (2004) ‘Do Primary-Care Clinicians Think That Nonspecific Low Back Pain Is One Condition?’, Spine, 29(9), pp. 1022-1031. doi: 10.1097/00007632-200405010-00015.

Kikuchi, S. (2001) ‘Values and Problems in MR Imaging for the Evaluation of Low Back Pain (Orthopedic Surgeon’s View)’, Semin Musculoskelet Radiol, 05(02), pp. 127-128. doi: 10.1055/s-2001-15665.