Effects of Exercise on Low Back Pain:a Meta-analysis

Objective:To conduct a meta-analysis on the effects of exercise intervention on low back pain, and the optimal exercise intervention program was proposed accordingly.Methods:A total of 35 articles and 2 224 subjectswere included in this study, the Review Manager 5.3 was used to dosubgroup analysis by setting six adjustment variables: intervention program, exercise time,exercise frequency, intervention duration, study quality and sample size.The overall effect, heterogeneity, and publication bias were evaluated by using random effects model.Results: 1). The effect size of exercise intervention on low back pain was d=-0.86, which achieveda large effect and statistically significant. 2).Exercise frequency(I2=82.5%) and study quality(I2=89.3%) were highly heterogeneous in the association between exercise intervention and low back pain; exercise time(I2=59.2%), intervention duration(I2=60.2%)and sample size(I2=59.8%) reveled a moderate heterogeneity in the association of exercise intervention with low back pain, and the intervention program(I2=44.0%) was close to moderate heterogeneity in the association between exercise intervention and low back pain.Conclusion: 1) Exercise intervention was significantlyin relieving low back pain. 2) Both suspension training and yoga training were effective in relieving low back pain,and the suspension training was better. 3) The optimal exercise frequency was3~4 times per week. 4) Compared with other programs,a 4-week exercise intervention program was most effective in relieving low back pain.