Patient-friendly summary
If you read nothing else
Bottom line
CBT-based interventions provide reliable but modest benefits for chronic musculoskeletal pain, with the largest effect on pain catastrophizing, and are best used as part of multimodal care rather than a stand-alone treatment.
Moderate evidencePublished
Evidence hierarchy
Study participants
Adults with chronic musculoskeletal pain (>=3 months), mainly chronic low back pain, osteoarthritis, rheumatoid arthritis, and fibromyalgia
Full research — for clinicians and curious readers
Study Summary
This systematic review and meta-analysis pooled 14 high-quality randomized controlled trials of cognitive behavioral therapy (CBT) and related CBT-based interventions for adults with chronic musculoskeletal pain. Across the trials, CBT-based interventions significantly reduced pain catastrophizing (moderate-to-large effect, SMD = -0.77), pain intensity (small-to-moderate effect, SMD = -0.41), and functional disability (small effect, SMD = -0.20) immediately after treatment. The authors conclude that CBT offers reliable but modest benefits and should be used as part of multimodal care rather than as a stand-alone treatment, with the largest effects on the cognitive and emotional targets it directly addresses. They caution that results carry substantial heterogeneity, possible publication bias for the pain-intensity outcome, and limited information about long-term durability.
Key Findings
| Finding | Detail | Impact |
|---|---|---|
| Largest effect was on pain catastrophizing | Across six studies (1,178 participants) measured with the Pain Catastrophizing Scale, CBT-based interventions produced a moderate-to-large reduction in catastrophizing (SMD = -0.77, 95% CI: -1.10 to -0.43, p < 0.05), though heterogeneity was high (I-squared = 85%). | High |
| Small-to-moderate reduction in pain intensity | Pooled across all 14 trials, CBT-based interventions reduced post-treatment pain intensity (SMD = -0.41, 95% CI: -0.62 to -0.20, p < 0.05), but with considerable heterogeneity (I-squared = 86.2%). | High |
| Small but significant reduction in functional disability | Across 13 trials (3,137 participants), CBT had a small effect on functional disability (SMD = -0.20, 95% CI: -0.36 to -0.03, p < 0.05). Effects were significant and homogeneous for the RMDQ subgroup (SMD = -0.21, I-squared = 0.0%) but nonsignificant and highly heterogeneous for the ODI subgroup (SMD = -0.09, p = 0.140). | Medium |
| Both traditional CBT and pain-coping skills training reduced catastrophizing | In subgroup analysis, traditional CBT (SMD = -0.74) and pain coping skills training (SMD = -0.89) each significantly reduced catastrophizing, with no statistically significant difference between the two approaches (p = 0.227). | Medium |
| Possible publication bias for the pain-intensity outcome | Egger's test (bias = -4.55, p = 0.002) and Begg's test (z = -3.01, p = 0.003) were both significant for pain intensity, suggesting funnel-plot asymmetry; smaller studies tended to report larger effects. Trim-and-fill identified no studies to adjust and the pooled estimate was unchanged. | Medium |
Strengths
- Prospectively registered protocol (PROSPERO) and PRISMA-guided methodology
- Comprehensive search across six databases (PubMed, Web of Science, Scopus, Embase, PsycINFO, Cochrane Library)
- Formal risk-of-bias assessment with Cochrane ROB 2.0 and inclusion of only RCTs
- Sensitivity (leave-one-out) and subgroup analyses to probe robustness and heterogeneity
- Authors clearly hedge conclusions and acknowledge heterogeneity and publication-bias concerns
Limitations
- Only 14 RCTs, with relatively small individual sample sizes and considerable clinical and methodological heterogeneity (I-squared up to 86.2%)
- Significant Egger's and Begg's tests suggested possible publication bias for the pain-intensity outcome
- Restricted to English-language publications, raising risk of language and publication bias
- Quantitative synthesis limited to post-treatment outcomes, providing only limited information on durability of effects
- GRADE was not used to formally rate certainty of evidence (acknowledged by authors)
- Pooled estimates combined traditional CBT with related interventions (PCST, ACT, CFT), limiting generalizability to strictly defined CBT
- Internal inconsistency in the reported total sample size (abstract states 2,677; results state 3,286)
Key Takeaways for Patients
What This Means for You
Talk therapy approaches like cognitive behavioral therapy (CBT) may help reduce chronic muscle and joint pain, but the benefit for pain itself is generally small-to-moderate.
CBT seems to help most with the worry, helplessness, and fear that can magnify pain (called pain catastrophizing), where the effect was the strongest.
Improvements in everyday physical function were smaller, suggesting CBT works best alongside other treatments like exercise and physical therapy rather than on its own.
The researchers describe CBT as a useful part of a broader pain-management plan, not a cure that fully eliminates pain or restores function.
Related approaches such as pain coping skills training showed similar benefits, so several forms of psychological pain treatment may be reasonable options.
Read the Full Paper
Access the complete peer-reviewed study from Frontiers in Psychology