Background:Yoga has been shown useful in reducing chronic low back pain (CLBP) through largely unknown mechanisms. The aim of this pilot study is to investigate the feasibility of providing yoga intervention to a predominantly underserved population and explore the potential mechanisms underlying yoga intervention in improving CLBP pain.Methods:The quasi-experimental within-subject wait-listed crossover design targeted the recruitment of low-income participants who received twice-weekly group yoga for 12 weeks, following 6–12 weeks of no intervention. Outcome measures were taken at baseline, preintervention (6–12 weeks following baseline), and then postintervention. Outcome measures included pain, disability, core strength, flexibility, and plasma tumor necrosis factor (TNF)-α protein levels. Outcomes measures were analyzed by one-way ANOVA and paired one-tailed t-tests.Results:Eight patients completed the intervention. Significant improvements in pain scores measured over time were supported by the significant improvement in pre- and post-yoga session pain scores. Significant improvements were also seen in the Oswestry Disability Questionnaire scores, spinal and hip flexor flexibility, and strength of core muscles following yoga. Six participants saw a 28.6%–100% reduction of TNF-α plasma protein levels after yoga, while one showed an 82.4% increase. Two participants had no detectable levels to begin with. Brain imaging analysis shows interesting increases in N-acetylaspartate in the dorsolateral prefrontal cortex and thalamus.Conclusion:Yoga appears effective in reducing pain and disability in a low-income CLBP population and in part works by increasing flexibility and core strength. Changes in TNF-α protein levels should be further investigated for its influence on pain pathways.