Included studies needed to be randomised or non-randomised trials that measured the effect of walking on pain levels and quality of life (QoL) in patients with low back pain. Participants could have low back pain that was acute (pain less than six weeks), sub-acute (pain six weeks to three months) or chronic (pain more than three months). Walking needed to be the main intervention or an adjunct to other interventions. Independent walking without use of equipment and walking on a treadmill were included. Outcome measures needed to be validated measures of pain or disability. Studies were excluded if participants had pathologies such as fractures, osteoporosis, infections, systemic musculoskeletal disorders, cancer and pregnancy.
The included studies varied in design, intervention, comparator and outcome measures. Three studies used a treadmill and one considered free walking. Sample size ranged from 11 to 208. Three studies used the visual analogue scale to assess pain and the other used Brief Pain Inventory (BPI).
Four reviewers independently initially assessed titles and abstracts for the review. Full-text retrieved articles were independently reviewed by two reviewers. A third reviewer acted as arbitrator where necessary.