Exercise beats usual care for chronic low back pain, by a margin patients can feel

Low back pain has been the leading cause of disability worldwide for at least three decades, draining healthcare budgets and productivity on an enormous scale. For the many adults whose pain is chronic and has no specific identifiable cause, the practical question is which conservative treatment actually helps. Exercise is widely recommended, but recommendations are only as good as the trial evidence behind them. This Cochrane review set out to weigh exercise against no treatment, usual care, placebo and every other conservative option, on both pain and daily function.

The reviewers gathered 249 randomised trials of exercise treatment for adults with chronic non-specific low back pain lasting more than 12 weeks, drawn from Europe, Asia, North America and the Middle East. Participants averaged 43.7 years old and 59% were female. Pain and functional limitation scores were rescaled to a common 0 to 100 scale, and the reviewers set thresholds in advance: a 15-point difference in pain and a 10-point difference in function would count as clinically important. Risk of bias and certainty of evidence were formally assessed.

Against no treatment, usual care or placebo, exercise reduced pain by 15.2 points, meeting the clinically important threshold, with moderate certainty. Function improved by 6.8 points, which fell short of the 10-point bar. Against other conservative treatments, exercise improved pain by 9.1 points and function by 4.1 points, effects too small to be clinically important overall. Subgroup analyses suggested exercise probably beats education alone and non-exercise physical therapy for pain, but showed no difference against manual therapy. Reported adverse effects were mostly minor, such as muscle soreness, and occurred at similar rates in exercise and comparison groups.

This is one of the largest syntheses of exercise for chronic back pain, and it lands with nuance: exercise clearly helps pain compared with doing nothing or usual care, but its advantage over other active conservative treatments is modest, and manual therapy matched it. Most trials carried a risk of bias because participants inevitably knew whether they were exercising, which the reviewers accounted for in their certainty ratings. Trials identified after April 2018 had not yet been incorporated, so the picture will be refined in future updates.

For anyone managing long-running, non-specific back pain, this is a reason to see exercise as a genuinely effective treatment for pain rather than a vague lifestyle suggestion, with mostly minor downsides like muscle soreness. It is also a reason to hold realistic expectations: the effect on day-to-day function was smaller than the pain relief, and exercise was not clearly superior to every alternative, matching manual therapy in particular. The evidence supports exercise as a first-line conservative option, not a cure.