Exercise therapy for treatment of non-specific low back pain

Sixty-one randomized controlled trials (6390 participants) met inclusion criteria: acute (11), subacute (6) and chronic (43) low-back pain (1 unclear). Evidence was found of effectiveness in chronic populations relative to comparisons at all follow-up periods; pooled mean improvement was 7.3 points (95% CI, 3.7 to 10.9) for pain (out of 100), 2.5 points (1.0 to 3.9) for function (out of 100) at earliest follow-up. In studies investigating patients (i.e. presenting to healthcare providers) mean improvement was 13.3 points (5.5 to 21.1) for pain, 6.9 (2.2 to 11.7) for function, representing significantly greater improvement over studies where participants included those recruited from a general population (e.g. with advertisements). There is some evidence of effectiveness of graded-activity exercise program in subacute low-back pain in occupational settings, though the evidence for other types of exercise therapy in other populations is inconsistent. There was evidence of equal effectiveness relative to comparisons in acute populations [pain: 0.03 points (95% CI, -1.3 to 1.4)].

Limitations: This review largely reflects limitations of the literature, including low quality studies with heterogeneous outcome measures, inconsistent and poor reporting, and possibility of publication bias.