The purpose of this study is to determine if the addition of lumbar manipulation will improve function, decrease pain, and recurrence in adolescents with low back pain. The secondary aim of this study is to determine if a modification of a clinical prediction rule by Flynn will be effective in identifying adolescent patients with low back pain who would benefit from lumbar manipulation.

This is a double blind, randomized controlled, parallel group study. The investigators will consider adolescent patients with a primary complaint of low back pain referred to physical therapy.

The purpose of this study is to determine if the addition of lumbar manipulation will improve function, decrease pain, and recurrence in adolescents with low back pain. The secondary aim of this study is to determine if a modification of a clinical prediction rule by Flynn will be effective in identifying adolescent patients with low back pain who would benefit from lumbar manipulation.

Prior to randomization, patients will complete several self-report measures and then receive a standardized history and physical examination performed by the treating physical therapist. The information collected will include age, sex, duration and nature of symptoms. Standardized physical examination measures include lumbar active range of motion assessment for quality and mobility, lumbar segmental mobility, hip internal rotation range of motion, straight leg raise test, and prone instability test. Treating physical therapists were trained in all evaluation measures before data collection began. Patients are assessed on each item of a modified clinical prediction rule which consists of 4 items; duration of symptoms, symptoms distal to knee, Hip Internal Rotation >35 degrees, and presence of lumbar segmental hypomobility. The fear avoidance beliefs questionnaire work sub-scale component was removed since the score would likely not be valid in the adolescent population. Meeting 3 of the 4 items was considered to be positive on the rule in this population.

The manipulating therapist who performs the intervention will be aware of treatment allocation, whereas the treating physical therapist and patients are blinded to treatment group allocation. To assess if the patients are successfully blinded to group allocation, the investigator will ask each patient to guess which intervention was performed on them following discharge from therapy.

Randomization is determined by checking a randomly generated computer list that tells the manipulating therapist the assigned group. The computer list was generated by an aide not involved in the study who blindly drew 52 cards that placed an equal number of patients into either the manipulation or sham group.

To address the issue of safety, patients who experience a clinically significant decrease in functional ability or increase in pain will be classified as having an adverse reaction. The minimally clinical important difference for the Patient Specific Functional Scale is >= 6. While the minimal clinical important difference of the Numeric Pain Rating Scale is 2 points. Following the manipulation or sham intervention patients will be asked if they experienced pain or discomfort during or following the intervention. An interim analysis is planned at the midpoint of the study to assess the safety and efficacy of the intervention. If >1/3 of patients in the manipulation group have a worsening of symptoms at any follow up, the study will be terminated for safety reasons.

Inclusion Criteria:

• patients 13 to 17 years old
• duration of low back symptoms less than 90 days

Exclusion Criteria:

• contraindication to manipulation. These included "red flags" for physical therapy, previous lumbar surgery, those who had signs consistent with nerve root compression (positive straight-leg test of < 45 degrees, diminished reflexes, sensation, or lower extremity strength), those who were pregnant, and those diagnosed with, or suspected of having a spondylolysis or spondylolisthesis.