Patient-friendly summary
If you read nothing else
Bottom line
For chronic low back pain, targeted posterior-chain (back, hip and trunk) resistance training over 12–16 weeks may improve pain, disability and strength more than general exercise or walking, with no apparent increase in adverse events—though evidence rests on a small number of varied trials.
Moderate evidencePublished
Evidence hierarchy
Study participants
Recreationally active and sedentary adults with nonspecific chronic low back pain lasting at least 12 weeks; elite athletes and those with known spinal pathology excluded
Study Summary
This systematic review and meta-analysis pooled 8 randomized controlled trials (408 participants: 203 in posterior-chain resistance training [PCRT], 205 in general exercise [GE]) to compare PCRT—exercise targeting thoracic, lumbar and hip-extensor muscles for at least 6 weeks—against general exercise or walking programmes for chronic low back pain (CLBP) in recreationally active and sedentary adults. Both approaches improved CLBP outcomes, but PCRT produced significantly greater reductions in pain and disability and greater gains in muscle strength, with the largest differences seen over 12–16 weeks rather than 6–8 weeks. There was no significant difference in adverse events between the two approaches. The authors conclude clinicians should strongly consider 12–16 weeks of PCRT, while noting heterogeneity, varied protocols, and limited adverse-event reporting.
Key Findings
| Finding | Detail | Impact |
|---|---|---|
| PCRT reduced pain more than general exercise, especially over 12–16 weeks | Overall pain favoured PCRT (SMD = −0.41, 95% CI −0.72 to −0.10, p = 0.009, I² = 51%). Over 12–16 weeks the effect was moderate (SMD = −0.61, 95% CI −1.21 to 0.00, p = 0.05, I² = 74%), whereas the 6–8-week subgroup showed no statistical difference (SMD = −0.26, p = 0.10, I² = 0%). | High |
| PCRT reduced disability more than general exercise, with the larger effect over 12–16 weeks | Pooled disability favoured PCRT (SMD = −0.31, 95% CI −0.56 to −0.06, p = 0.02, I² = 29%). Over 12–16 weeks the effect was moderate (SMD = −0.53, 95% CI −0.97 to −0.09, p = 0.02, I² = 52%); the 6–8-week subgroup was not significant (SMD = −0.15, p = 0.54, I² = 0%). | High |
| PCRT increased muscle strength more than general exercise at both durations | Pooled strength favoured PCRT (SMD = 0.45, 95% CI 0.18 to 0.72, p = 0.001, I² = 0%), with a significant benefit at 6–8 weeks (SMD = 0.34, 95% CI 0.00 to 0.67, p = 0.05) and a larger, moderate effect at 12–16 weeks (SMD = 0.67, 95% CI 0.21 to 1.13, p = 0.004). All strength subgroups had I² = 0%. | High |
| No significant difference in adverse events between PCRT and general exercise | Risk difference was not significant (RD = −0.02, 95% CI −0.10 to 0.05, p = 0.57, I² = 72%). PCRT had 2 reported adverse events and GE had 14, but all 14 GE events (10 back pain, 2 knee, 2 ankle) came from a single trial; 2 of 8 studies reported no adverse-event data and one reported events without numbers, so no level of evidence could be established for this outcome. | Medium |
| Benefits of PCRT appear time-dependent and may relate to progressive overload | Differences favouring PCRT grew with longer programmes (strong evidence for moderate effects on pain, disability and strength at 12–16 weeks vs small/conflicting effects at 6–8 weeks). Excluding one outlier multisite trial with minimal added resistance lowered heterogeneity (I² = 0–12%) and strengthened results for PCRT, suggesting progressive overload may be a key feature. | Medium |
Overall pain favoured PCRT (SMD = −0.41, 95% CI −0.72 to −0.10, p = 0.009, I² = 51%). Over 12–16 weeks the effect was moderate (SMD = −0.61, 95% CI −1.21 to 0.00, p = 0.05, I² = 74%), whereas the 6–8-week subgroup showed no statistical difference (SMD = −0.26, p = 0.10, I² = 0%).
Pooled disability favoured PCRT (SMD = −0.31, 95% CI −0.56 to −0.06, p = 0.02, I² = 29%). Over 12–16 weeks the effect was moderate (SMD = −0.53, 95% CI −0.97 to −0.09, p = 0.02, I² = 52%); the 6–8-week subgroup was not significant (SMD = −0.15, p = 0.54, I² = 0%).
Pooled strength favoured PCRT (SMD = 0.45, 95% CI 0.18 to 0.72, p = 0.001, I² = 0%), with a significant benefit at 6–8 weeks (SMD = 0.34, 95% CI 0.00 to 0.67, p = 0.05) and a larger, moderate effect at 12–16 weeks (SMD = 0.67, 95% CI 0.21 to 1.13, p = 0.004). All strength subgroups had I² = 0%.
Risk difference was not significant (RD = −0.02, 95% CI −0.10 to 0.05, p = 0.57, I² = 72%). PCRT had 2 reported adverse events and GE had 14, but all 14 GE events (10 back pain, 2 knee, 2 ankle) came from a single trial; 2 of 8 studies reported no adverse-event data and one reported events without numbers, so no level of evidence could be established for this outcome.
Differences favouring PCRT grew with longer programmes (strong evidence for moderate effects on pain, disability and strength at 12–16 weeks vs small/conflicting effects at 6–8 weeks). Excluding one outlier multisite trial with minimal added resistance lowered heterogeneity (I² = 0–12%) and strengthened results for PCRT, suggesting progressive overload may be a key feature.
Strengths
- First systematic review and meta-analysis to directly compare posterior-chain resistance training against general exercise/walking for chronic low back pain
- Prospectively registered (PROSPERO) and conducted per PRISMA guidelines with a four-database search
- Independent dual reviewers with high inter-rater agreement (Cohen's kappa = 0.896); 7 of 8 included trials rated 'good' methodological quality on the JBI tool
- Examined multiple clinically relevant outcomes (pain, disability, strength, adverse events) rather than pain alone
- Conservative imputation of missing p values likely underestimated rather than overestimated effects in favour of PCRT
- Low heterogeneity for strength outcomes (I² = 0%) and funnel-plot inspection suggesting low publication-bias risk
Limitations
- Small total sample (8 trials, 408 participants), limiting confidence in subgroup analyses
- Substantial heterogeneity for some outcomes (pain I² up to 74%; adverse events I² = 72%), partly driven by a single outlier multisite trial
- Wide variation in PCRT protocols (exercise selection, load, sets, reps, equipment) and in GE comparators (other resistance training, aerobic/walking, activities of daily living), making the optimal protocol unclear
- Adverse events were poorly and inconsistently reported—2 of 8 studies gave no data and one reported events without numbers—so no level of evidence could be established for safety
- Some data points (p values, variance) were estimated/imputed because they were not published in the original trials
- Participants' prior activity levels were often not clearly specified, limiting interpretation of who benefits most
Key Takeaways for Patients
What This Means for You
- 01For long-lasting (chronic) low back pain, exercise that specifically strengthens the back, hip and trunk extensor muscles (the 'posterior chain') may reduce pain and disability and build strength more than general exercise or walking.
- 02Programmes lasting about 12–16 weeks appear to give bigger benefits than shorter 6–8-week programmes, so sticking with it for several months may matter.
- 03In these studies, this kind of resistance training did not cause more injuries or flare-ups than general exercise or walking—contrary to the common belief that lifting is risky for the back.
- 04Both resistance training and general exercise helped, so being active in some form is valuable; the studies suggest targeted, progressive strengthening adds extra benefit.
- 05This is general research evidence, not a personalized plan—anyone with chronic back pain should have exercises tailored and progressed with a clinician, especially if starting from a deconditioned or painful baseline.
Read the Full Paper
Access the complete peer-reviewed study from Sports Medicine - Open
View Full Study