Patient-friendly summary
If you read nothing else
Bottom line
CBT showed a small short-term benefit for social participation in chronic low back pain but no significant advantage over other treatments for depression, anxiety, or sleep; the evidence is limited and needs confirmation by larger, higher-quality trials.
Preliminary evidencePublished
Evidence hierarchy
Study participants
Adults over 18 with chronic nonspecific low back pain lasting more than 6 months
Study Summary
This systematic review and meta-analysis pooled 16 randomized controlled trials (2527 patients with chronic nonspecific low back pain lasting more than 6 months) to test whether cognitive behavioral therapy (CBT) improves social engagement and reduces pain susceptibility compared with waiting lists or other active treatments. CBT showed a small but statistically significant advantage for improving social participation after treatment (SMD = -0.30, 95% CI -0.60 to -0.01, P = .04), but this benefit did not persist at 6-month follow-up. CBT was not statistically superior to other treatments for reducing depression (SMD = -0.07, P = .27) or anxiety (SMD = -0.07, P = .57); sleep findings could not be pooled because too few studies (3) reported them. The authors caution that the limited number and low methodological quality of the included trials mean these conclusions need confirmation by larger, high-quality RCTs.
Key Findings
| Finding | Detail | Impact |
|---|---|---|
| CBT modestly improved social participation after treatment | Across 9 studies, CBT was superior to other treatments for short-term social engagement (SMD = -0.30, 95% CI -0.60 to -0.01, Z = 2.02, P = .04), analyzed with a random-effects model due to high heterogeneity. The pooled estimate should be interpreted with caution given inconsistent outcome measures. | High |
| The social-engagement benefit did not persist at 6 months | In a subgroup analysis of 3 studies reporting 6-month follow-up data (fixed-effects model, low heterogeneity), CBT did not show statistically better long-term effects on social participation compared with other treatments. | Medium |
| CBT showed no significant advantage for depression | Across 9 studies, CBT was not statistically better than other treatments for depression after treatment (SMD = -0.07, 95% CI -0.19 to 0.05, Z = 1.11, P = .27; fixed-effects, P = .07, I2 = 44%). At 6-month follow-up (3 studies, random-effects, I2 = 75%) it also showed no significant long-term benefit. | High |
| CBT showed no significant advantage for anxiety | Across 5 studies (random-effects model, high heterogeneity), CBT was not statistically better than other treatments for improving anxiety (SMD = -0.07, 95% CI -0.30 to 0.16, Z = 0.52, P = .57). | High |
| Sleep findings could not be pooled | Only 3 studies assessed sleep quality with inconsistent outcome measures, so data were analyzed descriptively. Some high-quality studies favored CBT for short-term sleep and pain self-efficacy, but only one showed a significant long-term sleep advantage for CBT. | Low |
Across 9 studies, CBT was superior to other treatments for short-term social engagement (SMD = -0.30, 95% CI -0.60 to -0.01, Z = 2.02, P = .04), analyzed with a random-effects model due to high heterogeneity. The pooled estimate should be interpreted with caution given inconsistent outcome measures.
In a subgroup analysis of 3 studies reporting 6-month follow-up data (fixed-effects model, low heterogeneity), CBT did not show statistically better long-term effects on social participation compared with other treatments.
Across 9 studies, CBT was not statistically better than other treatments for depression after treatment (SMD = -0.07, 95% CI -0.19 to 0.05, Z = 1.11, P = .27; fixed-effects, P = .07, I2 = 44%). At 6-month follow-up (3 studies, random-effects, I2 = 75%) it also showed no significant long-term benefit.
Across 5 studies (random-effects model, high heterogeneity), CBT was not statistically better than other treatments for improving anxiety (SMD = -0.07, 95% CI -0.30 to 0.16, Z = 0.52, P = .57).
Only 3 studies assessed sleep quality with inconsistent outcome measures, so data were analyzed descriptively. Some high-quality studies favored CBT for short-term sleep and pain self-efficacy, but only one showed a significant long-term sleep advantage for CBT.
Strengths
- First meta-analysis to focus specifically on pain susceptibility and social engagement (within an ICF framework) in chronic low back pain rather than only symptoms or activity levels
- Strict inclusion criteria requiring pain duration of at least 6 months, making results more relevant to typical chronic low back pain patients
- Adhered to Cochrane methodology and PRISMA reporting, with screening, data extraction, and risk-of-bias assessment performed independently by two researchers
- Pre-registered protocol (PROSPERO CRD42023385655)
- Relatively large total pooled sample (2527 patients)
Limitations
- Limited number of included studies and, for some outcomes, few trials per analysis (e.g., 3-5 studies)
- Low methodological quality of included trials, including unclear or absent randomization details in several studies and one trial that allocated patients by depression level rather than randomly
- Most studies could not be blinded due to the nature of the treatment
- High statistical heterogeneity and inconsistent outcome measures, so several pooled estimates should be interpreted with caution
- Sleep outcomes could not be pooled due to too few studies and inconsistent metrics
- One unusually large study influenced the depression pooled effect (though removing it did not change the statistical conclusion)
Key Takeaways for Patients
What This Means for You
- 01For long-standing low back pain (lasting more than 6 months), cognitive behavioral therapy (CBT) may help you re-engage with work, hobbies, and relationships in the short term, though the benefit was modest and faded by 6 months.
- 02In this pooled analysis, CBT was not clearly better than other treatments at easing depression or anxiety, so it should not be assumed to be the best option for managing mood.
- 03Evidence on sleep was too limited to draw firm conclusions.
- 04CBT is considered relatively safe compared with some other treatments, and remote or self-help formats may make it easier to access.
- 05Because the underlying studies were few and of low quality, these findings are not definitive and may change as larger, better trials are done. Discuss your options with your clinician.