The purpose of this study was to systematically review the evidence examining effects of walking interventions on pain and self-reported function in individuals with chronic musculoskeletal pain.
Randomized and quasi-randomized controlled trials in adults with chronic low back pain, osteoarthritis or fibromyalgia comparing walking interventions to a non-exercise or non-walking exercise control group were reviewed. Twenty-six studies (2384 participants) were included and suitable data from 17 were pooled for meta-analysis with a random effects model used to calculate between group mean differences and 95% confidence intervals. Data were analyzed according to length of follow-up (short-term: ≤8 weeks post randomization; medium-term: >2 months – 12 months; long-term: > 12 months). Interventions were associated with small to moderate improvements in pain at short (mean difference (MD) -5.31, 95% confidence interval (95% CI) -8.06 to -2.56) and medium-term follow-up (MD -7.92, 95% CI -12.37 to -3.48). Improvements in function were seen at short (MD -6.47, 95% CI -12.00 to -0.95), medium (MD -9.31, 95% CI -14.00 to -4.61) and long-term follow-up (MD -5.22, 95% CI 7.21 to -3.23).
Evidence of fair methodological quality indicates that there is an association between walking and significant improvements in outcome compared to control interventions but longer-term effectiveness is uncertain. Using the USPSTF system, walking can be suggested as an effective form of exercise or activity for individuals with chronic musculoskeletal pain but should be augmented with strategies aimed at maintaining participation. Further work is also required examining effects on important health related outcomes in this population in robustly designed studies.