The treatment effect of exercise programmes for chronic low back pain.

RATIONALE: This paper summarizes evidence of long-term effectiveness of physiotherapy exercise therapy for chronic low back pain (LBP).
METHODS: A literature search was undertaken for experimental studies (2001-2007), which reported any post-intervention (follow-up) outcomes. Studies were critically appraised using the PEDro instrument. Comparative statistics were calculated, relative to the type of follow-up outcome data.
RESULTS: Fifteen moderate quality trials were included [mean PEDro score 7.7, SD 1.3 (range 5-10)]. Nine trials reported pain scales, and six reported LBP reoccurrence. Trials which reported on pain scales at 6-month follow-up found significant differences in favour of exercise [standardized mean differences -0.57, 95%CI -0.75 to -0.39 (555 participants)]. At 12-month follow-up, a small pain scale benefit from exercise persisted [standardized mean differences -0.25, 95%CI -0.44 to 0.06 (434 participants)]. There was unconvincing evidence of exercise effectiveness on pain scales after this time. Three of the four trials which reported dichotomous outcomes at 6-month follow-up demonstrated large clinical benefits of exercise (relative risk reduction of reoccurrence 45-246%, absolute risk reduction of reoccurrence 36-42 for every 100 patients; and number needed to treat approximating 3, to prevent one patient suffering a LBP recurrence). The effect of exercise on LBP reoccurrence was variably reported beyond 6 months.
CONCLUSION: Exercise programmes are effective for chronic LBP up to 6 months after treatment cessation, evidenced by pain score reduction and reoccurrence rates. The way in which follow-up data are reported assists clinical interpretation of research findings.