Abstract ID: 4009

Primary Topic: Mind-body (including meditation and yoga)
Secondary Topic: Patient-Centered Outcomes Research
Tertiary Topic: Integrative Health and the Underserved

Yoga, physical therapy, and education for sleep quality in adults with chronic low back pain: a secondary analysis of a randomized controlled trial

Late Breaker: Yes


Poor sleep quality is common among adults with chronic low back pain (cLBP). While yoga and physical therapy (PT) are effective cLBP treatments, their impact on sleep quality in adults with cLBP is unknown.

Methods/Session Format

To evaluate the effectiveness of yoga and PT for improving sleep quality in adults with cLBP we performed a secondary analysis of a clinical trial which randomized 320 adults with cLBP to 12 weekly group-based yoga classes, 1-on-1 PT sessions, or an educational book. Sleep quality was measured using Pittsburgh Sleep Quality Index (PSQI) global score (0-21) at baseline, 12 and 52 weeks. We identified participants with poor baseline sleep quality (PSQI >5). We calculated 12 and 52-week within-group PSQI change scores (from baseline) using regression-based multiple imputation to impute missing data. To compare Yoga and PT to education, we calculated 12 and 52-week between-group PSQI change scores. Lastly, we compared the proportion of ‘responders’ (i.e., 3-point PSQI improvement at 12-weeks) in each group.


Participants (mean age=46.0 [SD=10.7]; mean PSQI score=10.2 [SD=3.9]) were mostly female (64%), black (57%), and had poor sleep quality (93%). Sleep quality improved at 12 and 52 weeks among yoga (PSQI mean difference [MD]=-1.2 [95%CI: -1.9, -0.6] and -2.4 [95%CI: -3.2, -1.5], respectively), PT (MD=-1.1 [95%CI: -1.8, -0.3] and MD=-1.7 [95%CI: -2.7, -0.7], respectively), and education (MD=-0.5, [95%CI: -1.4, 0.5] and MD=-1.0 [95%CI: -2.1, 0.1], respectively) participants. Small non-significant (p>0.05) between-group differences favored Yoga and PT to education at twelve (MD=-0.7 and -0.5, respectively) and fifty-two (MD=-1.1 and -0.5, respectively) weeks. More yoga (35%) and PT (35%) than education (25%) participants were responders (p=0.37).


In a sample of adults with cLBP, the vast majority had poor sleep quality at baseline. Small within-group improvements in Yoga and PT groups were similar in magnitude; the education improvement was less in magnitude.