Interventions: Relaxation/mindfulness (music therapy, hypnosis, progressive muscle relaxation with biofeedback), multimodal therapy of guided imagery and music, cognitive/behavioral (CBT-based programs, and a postoperative management program comprising motivational interviewing to improve self-efficacy and goal attainment), and a combination of relaxation/mindfulness and cognitive (pain coping skills program) intervention

Comparator: active treatment or control treatment (e.g., standard care, placebo, no treatment).

Clinical outcomes: Postoperative knee pain severity after TKR, measured using Pain VAS, NRS, WOMAC Pain Scale, and the Short-form MPQ(primary)

Function, HRQoL, and psychological well-being/status (secondary)

SAE

Follow-up: no time limit placed on assessment duration/follow-up

Clinical outcomes: Pain: VAS, modified von Korff scale

Disability: RMDQ, “Recovery” defined as resolution of pain (DDS) and restoration of function (SIP),

HRQoL: Health status (QWB), EQ-5D, SF-12, SF-36, life satisfaction (CLS)

Follow-up: 18 weeks – 24 months

Intervention: Cognitive therapy, mindfulness, and ACT (CBT and CBT-integrated with physical therapy) delivered pre, peri, or postoperatively.

Duration ranged from 3 – 8 sessions, 30 minutes – 3 hours/session, and delivered in person or by phone, in individual or group format, and were led by a nurse, physiotherapist, or psychologist.

Clinical outcomes: Pain: Pain interference and/or severity measured by VAS, NRS, BPI, SF-36 BP scale, and LBP-RS, and consumption of rescue analgesics

Functional outcomes: disability measured by ODI, ambulation using CAS, and mobility using the five-chair-stand test, timed up-and-go test, and 10 min walk test

Follow-up: First postoperative week, 2 months to 2–3 years

Intervention: Physiotherapist delivered psychological interventions (CBT, stress management, relaxation training, hypnosis, mindfulness or acceptance-based interventions, coping skills training, problem-solving, systematic desensitization, motivational interview, and anxiety and depression interventions) combined with physiotherapy

Comparator: physiotherapy alone or usual care or no treatment

Clinical outcomes: Pain: VAS, NRS, modified von Korff scale

Disability: RMDQ, ODQ/ODI, PDI, NDI, WOMAC, QBPDS

Follow-up: short-term (4 to 16 weeks), long-term (26 to 52 weeks).

Intervention: CBT with or without another intervention (e.g. physiotherapy)

Comparator: Placebo, no treatment (information booklet), or waiting list controls, and other types of interventions (didactic discussion, manual therapy)

Clinical outcomes: Pain: NRS

Disability: NDI

Follow-up: Short-term follow-up

Mas et al. 2019¹⁷

Spain

Intervention: Usual clinical care + a biopsychosocial multidisciplinary intervention, consisting of physiotherapy, CBT and medication

Comparator: Guideline-based pharmacological treatment

Clinical outcomes:

Disability: RMDQ (primary)

Pain intensity: MPQ and

Quality of Life: SF-12 (secondary)

Follow-up: 12 months

Intended users: all clinicians

Target population: adults (≥18 years) with acute (< 4 weeks), subacute (4 to 12 weeks), and chronic (> 12 weeks) nonradicular low back pain, radicular low back pain, or symptomatic spinal stenosis

Magnitude of effect determined as follows:

For pain:

• Mean post-treatment between-group difference on a VAS of 0 to 100 or equivalent: 5 – 10 points (small), > 10 – ≤ 20 points (moderate)
• Mean between-group difference on a numerical rating scale of 0 to 10: 0.5 – 1.0 point (small), > 1.0 – ≤ 2.0 points (moderate)
• Standardized mean difference of 0.2 – 0.5 (small effect), > 0.5 – ≤ 0.8 (moderate)

For function:

• Mean between-group difference on the ODI: 5 – 10 points (small), > 10 – ≤ 20 points (moderate)
• Mean between-group difference on the RDQ: 1 – 2 points (small), > 2 – ≤ 5 points (moderate)
• Standardized mean difference of 0.2 – 0.5 (small effect), > 0.5 – ≤ 0.8 (moderate)

No large effects were found with any intervention.

• Overall, the review was conducted following Cochrane guidance to ensure the methodology was robust and systematic.
• The scope of the systematic review was clear, with predefined inclusion/exclusion criteria for patient population, interventions, comparators, outcomes and study design.
• A systematic and comprehensive literature search was conducted, with predefined search strategy and no language restrictions applied and in multiple databases. In addition, a manual search from reference list of retrieved papers and review articles was also performed. The reporting of the search strategy followed the requirements of PRISMA statement.
• Study selection was done by one reviewer, which was validated by another investigator, disagreements were resolved by discussion or consensus involving a third investigator.
• Data extraction and validation were done independently by separate researchers. Investigators of the primary studies were contacted if disaggregated data were not reported; and studies were excluded if contact was unsuccessful.
• The Cochrane risk of tool for RCTs was used to assess the risk of bias and quality of evidence, with information extracted and validated independently by separate researchers. However, no information was provided on how disagreements were resolved.
• Study and patient characteristics of each included trial were provided with adequate details.
• Opportunities for pooling data for meta-analysis were explored but limited because of heterogeneity in the content, duration, and intensity of the interventions, and conclusions were therefore based on narrative synthesis. Between-study heterogeneity was not assessed quantitatively as meta-analysis could not be conducted.
• All authors and reviewers of the report declared no conflicts of interest. The source of funding for the report was provided.

• All included trials had high or unclear risk of bias for at least two domains in the Cochrane risk of bias tool. Overall, the included studies performed poorly in the domains of performance, detection, attrition, and reporting bias.
• An assessment of publication bias was not done, and not justified.

• Overall, study methodology appeared to be appropriate and robust.
• The scope of the systematic review was clear, with predefined inclusion/exclusion criteria for patient population, interventions, comparators, outcomes and study design.
• A systematic and comprehensive literature search was conducted, with predefined search strategy and in multiple databases. In addition, a manual search from reference list of retrieved papers and review articles was also performed.
• Study selection was done independently by two reviewers, disagreements were resolved by discussion or consensus involving a third investigator.
• Study and patient characteristics of each included trial were provided with adequate details.
• Opportunities for pooling data for meta-analysis were explored but limited due to too few studies included, as well as heterogeneity in almost every key aspect—most notably in how the definition of condition, nature and frequency of the CBT intervention, length of follow-up, and the outcome measures. Conclusions were therefore based on narrative synthesis. Between-study heterogeneity was not assessed quantitatively as meta-analysis could not be conducted.
• All authors and reviewers of the report declared no conflicts of interest. The source of funding for the report was provided.

• There was no information on whether data extraction and validation were done independently by separate researchers or not.
• An assessment of methodological quality of the included studies was not made, although the authors stated an absence of widely accepted rating systems for prospective studies as the reason for this. However, a qualitative analysis and summary was made for each study.
• An assessment of publication bias was not done, and not justified.

• Overall, study methodology appeared to be appropriate and robust.
• The scope of the systematic review was clear, with predefined inclusion/exclusion criteria for patient population, interventions, comparators, outcomes and study design.
• A systematic and comprehensive literature search was conducted, with predefined search strategy and in multiple databases. In addition, additional articles were identified through a hand search of relevant bibliographies. The reporting of the search strategy followed the requirements of PRISMA statement.
• Study selection was done independently by two reviewers, disagreements were resolved by consensus.
• The Cochrane risk of tool for RCTs was used to assess the risk of bias and quality of evidence, with information extracted and validated independently by separate researchers. Any discrepancies in bias-risk ratings were resolved by an independent third researcher. Most trials had generally low risk of bias across the six domains, with some reporting performance and detection bias to be high.
• Three authors independently abstracted data from each article, and any discrepancies in data capture were resolved via consensus.
• Study and patient characteristics of each included trial were provided with adequate details.
• Pooling data for meta-analysis was not conducted due to heterogeneity in interventions, statistical analyses, follow-up periods, and outcomes assessed. Conclusions were therefore based on narrative synthesis. Between-study heterogeneity was not assessed quantitatively as meta-analysis could not be conducted.
• All authors and reviewers of the report declared no conflicts of interest. The source of funding for the report was provided.

• An assessment of publication bias was not done, and not justified.

• Overall, study methodology appeared to be appropriate and robust.
• The scope of the systematic review was clear, with predefined inclusion/exclusion criteria for patient population, interventions, comparators, outcomes and study design.
• A systematic and comprehensive literature search was conducted, with predefined search strategy and no language restrictions applied and in multiple databases. In addition, a manual search from reference list of retrieved papers and review articles was also performed. The reporting of the search strategy followed the requirements of PRISMA statement.
• Study selection was done in duplicate, disagreements were resolved by discussion or consensus involving a third reviewer.
• Data extraction and validation were done independently by separate researchers. Disagreements were resolved by consensus and otherwise by consultations with a third reviewer.
• Two reviewers independently assessed the risk of bias and quality of evidence using the Cochrane risk of bias tool for RCTs, with disagreements were resolved by discussion and consensus. All of the reviewed studies showed a high risk of performance bias, whereas up to a quarter of studies showed a high risk of attrition bias and other biases.
• Study and patient characteristics of each included trial were provided with adequate details.
• Statistical methods for data analysis appeared to be appropriate, between-study heterogeneity was assessed, subgroup analyses were done that were aligned with the objective of the systematic review.
• Funnel plots were used to identify publication bias and assess reporting bias. Asymmetry of the funnel plots was assessed visually and using the Egger test when a minimum of 10 studies were included in the meta-analysis because the test power of fewer than 10 studies is usually too low to distinguish chance from real asymmetry.
• The overall quality of the evidence for the main analyses was assessed using the GRADE approach.
• All authors and reviewers of the report declared no conflicts of interest. The source of funding for the report was provided.

• Overall, study methodology appeared to be appropriate and robust.
• The scope of the systematic review was clear, with predefined inclusion/exclusion criteria for patient population, interventions, comparators, outcomes and study design.
• A systematic and comprehensive literature search was conducted, with predefined search strategy and no language restrictions applied and in multiple databases. In addition, a manual search from reference list of retrieved papers and review articles was also performed. The reporting of the search strategy followed the requirements of PRISMA statement.
• Study selection was done by 5 teams in duplicate, disagreements were resolved by discussion or consensus involving another reviewer.
• Data extraction were done independently by separate researchers. Disagreements were resolved by consensus and in consultations with a third reviewer if disagreement persisted.
• Two reviewers independently assessed the risk of bias and quality of evidence for each included RCT using the 12 criteria recommended by the Cochrane Back Review Group, which is an expansion of the criteria described in the Cochrane Handbook for Systematic Reviews of Interventions. Most of the included studies reported low or unclear risk of bias in almost all domains except for the domains associated with blinding.
• Study and patient characteristics of each included trial were provided with adequate details.
• Statistical methods for data analysis appeared to be appropriate, between-study heterogeneity was assessed, Subgroup analyses were planned but not conducted due to insufficient numbers of studies in each pairwise comparison.
• The overall quality of the evidence for the main analyses was assessed using the GRADE approach.
• Funnel plots were used to assess reporting bias when at least 10 studies were included in a meta-analysis and studies were not of similar size. Asymmetry of the funnel plots was assessed visually and using the Egger test Harbord method.
• All authors and reviewers of the report declared no conflicts of interest. The source of funding for the report was provided.

Reporting

• The objective of the study, main outcomes, inclusion and exclusion criteria, interventions being compared, potential confounders, and main findings were clearly described.
• The estimates of random variability (standard deviation/error or 95% confidence intervals) and exact P values were reported for the main outcomes.

External validity

• Participants in the trial were generally representative of the population from which they were recruited. Patients were excluded if they had more than 12 weeks of pain, history of prior LBP within 6 months, LBP that coexisted with cognitive impairment or psychiatric disorders, other causes of disability which impeded responding to the questionnaires, pregnancy and breastfeeding, physical problems in the preceding 3 months; and a diagnosis of fibromyalgia. These exclusion criteria do not appear to seriously affect the generalizability.

Internal validity

• Randomization method was reported, a reasonable justification was provided why a cluster randomization method was chosen i.e. to minimize contamination.
• This was a single-blinded trial, with unblinded physician and blinded patients. Additionally, the analysts were blinded to the treatment allocation, adding to the strength of the findings.
• The interventions were administered in a standardized manner, by allowing only one qualified psychologist and one physiotherapist with expertise in group interventions to implement the interventions in all healthcare centres. Further, data collection was done in a robust manner.
• The outcomes measured were valid and reliable, or were widely accepted in the area of pain research.
• The statistical tests were appropriate; in accordance with CONSORT guidelines. The effect of cluster was taken into account using appropriate analytical method. Multiple imputation method was used to address potential biases due to incomplete follow-up. The final models were adjusted for significant confounders and significant interaction variables.
• There were no major imbalances between treatment arms in terms of baseline demographic and clinical characteristics, indicating randomization was preserved across clusters.

Internal validity

• Control for multiple comparisons was not in place with a priori statistical hierarchy for secondary outcomes.
• Some confounding factors were not accounted for in the model and may limit the conclusions of the study. For example, the differences in the profile of patients, since they were allocated by primary healthcare centres and socioeconomic status was not considered.
• There were more healthcare centres allocated to the intervention group than the control group; however, the impact of this observation on the results is unclear.

Sample size/power

• Even though the authors conducted a power calculation, it is unclear if the trial had sufficient power to detect a clinically important effect for the primary endpoint. The authors estimated that 348 participants per intervention arm would allow detecting a mean difference of 2.5 points (SD 5.7) between the groups after accounting for 20% dropout. However, the number of patients recruited in the intervention and control group (n = 262 and 239, respectively) was well short of the target sample size. the number was even smaller at 3 and 12 months of follow-up.

Findings of each relevant trial up to month 3 are described below.

Music:

• 5 RCTs with 256 participants evaluated the effectiveness of music therapy for reducing acute postoperative pain following TKR.
• One 2-arm study reported lower mean VAS pain scores at 3 hours (1.5 (SD 1.4) vs 3.9 (SD 3.4); p=0.01) and 24 hours (2.4 (SD 1.7) vs 4.1 (SD 2.9); p=0.04) post-surgery with patient-selected music played on headphones during surgery compared to white noise.
• The remaining studies found no benefits of 20-minute ‘easy listening’ music, 30-minute soothing piano and Chinese violin music, 12 to 15-minute instrument only music with varying degrees of harmonicity and rhythmicity, and 5-minute music therapist-delivered live music compared to similarly timed quiet rest period, usual care, wearing headphones with no input, and no music, respectively; with respect to VAS or NRS pain score at various timepoints (minutes to days).

Guided imagery and music:

• One 2-arm RCT evaluated the effectiveness of guided imagery on outcomes post-surgery with 82 participants.
• The study compared 19–21 minutes of audio-recorded guided imagery combined with soothing instrumental background music daily for 2 weeks before surgery and 3 weeks after surgery to a control group who received 17–21 min of commercially available spoken word audio recordings (eg, poetry, short stories, essays) at the same timepoints. However, no comparisons were made between trial arms even though pain was measured using the WOMAC pain score and VAS pain score up to 6 months post-surgery.

Hypnosis:

• One 3-arm RCT with 24 patients evaluated the effectiveness of a prerecorded hypnotic audio recording on pain outcome post-surgery.
• There was a small difference in mean NRS ratings between patients who received 35-minute prerecorded hypnosis audio (listened to at least once pre-surgery and at least once 24 hours post-surgery) compared to those who received minimal treatment effect (psychoeducation, diaphragmatic breathing, relaxing music), and treatment as usual, no statistical comparison reported – mean NRS score at 72 hours: 1.77 vs 2.23 vs 2.59.

Progressive muscle relaxation with biofeedback:

• One 2-arm RCT with 66 participants compared a multimodal combination of 3—minute training on biofeedback-assisted progressive muscle relaxation skills pre-surgery and 30-minute sessions of continuous passive motion therapy twice a day for 5 days post-surgery to standard continuous passive motion therapy.
• The former intervention showed a statistically significant lower NRS pain scores compared with the control group P < 0.001).

Pain coping skills program:

• One 3-arm RCT with 402 patients evaluated the effectiveness of a pain coping skills program to arthritis education and to usual care. The program comprised of eight 50-minute sessions over a 2-month period beginning 2 weeks pre-surgery and ending 6 weeks after surgery and included sessions on cognitive restructuring, thought identification and challenging, self-calming and relaxation techniques, and activity management; delivered in-person and via telephone.
• At baseline and up to 12 months post-surgery, no differences were found in mean WOMAC pain treatment scores or group-by-time interaction.

Enhanced postoperative recovery using motivational interviewing:

• One 2-arm RCT with 308 participants evaluated the effectiveness of an enhanced postoperative recovery program (comprising 10 telephone calls with a navigator over a 6-month postoperative period aimed at identifying postsurgical objectives and improving self-efficacy using motivational interviewing) compared to usual care (including inpatient physiotherapy and outpatient physiotherapy after discharge) to improve postoperative functional status.
• At baseline and up to 6 months post-surgery, no differences were found in mean WOMAC pain treatment scores between interventions.

CBT programs:

• 2 RCTs with 150 participants evaluated the effectiveness of CBT-based programs.
• One 2-arm RCT with 50 participants compared CBT-based program (up to ten 1-hour sessions for reducing anxiety and depression) to standard care. At 4 months post-surgery, no differences were found in mean WOMAC pain treatment scores between the interventions.

• 1 RCT with 93 patients with subacute LBP compared CBP (individual education, coping strategies, reassurance (minimum 2 sessions) with exercise and control. Results showed improved disability in CBT group vs. control (P = 0.02). However, no indication which outcome favored CBT (VAS, RMDQ), SF-36, Cantril’s Ladder Scale.
• 1 RCT with 64 patients with subacute LBP compared the clinical benefits of CBP (9 individual biopsychosocial treatment 3×/week and 12 relaxation sessions, 4×/week) and control. Results showed reduced pain intensity (VAS) for CBT group vs. control at 6 months (all P ≤ 0.005).
• 1 RCT with 67 patients with subacute LBP compared the clinical benefits of CBP (6 sessions of Individual, custom program including problem solving, coping skills, relaxation) and control. Results showed CBT group had greater “recovery” (resolution of pain [DDS] and restoration of function [SIP] at 6 months) if completed 4 (P = 0.02) or 6 (P = 0.002) sessions. However, no results provided for health status (QWB).
• 1 RCT with 38 patients with subacute LBP compared the clinical benefits of CBP (Individual psychotherapy, minimum 7 sessions, 1×/week) and control. However, none of the reported outcomes were relevant for this report.
• 1 RCT with 701 patients with a mix of subacute and chronic LBP compared the clinical benefits of CBP (6 sessions of group therapy addressing “behaviors and beliefs about physical activity and avoidance of activity”) and control. Results showed CBT group had significantly improved RMDQ, von Korff score, SF-12 physical component subscale at 3, 6, and 12 months (all P < 0.032); unknown how much of this effect was driven by chronic LBP participants.
• 1 RCT with 216 patients with a mix of subacute and chronic LBP compared the clinical benefits of CBP (4 sessions of “Cognitive educational program”) and control. Results showed CBT group was not significantly different than control with respect to RMDQ, VAS, and EQ-5D.

“Five of the six showed significant improvements associated with CBT, but the heterogeneity of the studies prevented quantitative comparisons” pg 1

“CBT has not been adequately studied as a potential early intervention treatment for sALBP patients. None of the six identified papers studied US civilians or leveraged innovations such as teletherapy—able to reach patients in remote or underserved areas—underscoring critical gaps in current back pain treatment. Given the severity of the US opioid epidemic, nonpharmacologic options such as CBT should be rigorously explored in the sALBP population” pg 1

Findings of each relevant trial up to month 3 are described below.

Trial 1:

• In a Swedish trial consisting of 107 patients undergoing lumbar fusion surgery, the clinical effectiveness of three 90-minute in-person PMT sessions (SOC+ CBT-informed physiotherapy intervention including behavioral self-management, cognitive restructuring, relaxation training, and motivational interviewing) and usual care (physiotherapist-instructed inpatient exercise therapy program and 20-minute exercise instruction at discharge) were compared.
• At 3 months, Disability (ODI): −9.7 (−15.8 to −3.6, P=0.002), d=1.2. Back pain (VAS): −11.7 (−19 to −4.3, P=0.002), d=1.45.

Trial 2:

• In an American trial consisting of 86 patients undergoing laminectomy, the clinical effectiveness of CBPT (6 weekly 30-minute telephone-based CBPT sessions consisted of behavioral self-management, problem-solving, cognitive restructuring, and relaxation training + education program) and education (education program consisting of physical therapy, stress management, sleep hygiene, energy management, communication with health providers, injury prevention) were compared at 3 months.
• BPI back pain: −0.88 (−1.5 to −0.25, P=0.007), d=0.62
• BPI leg pain: −1.2 (−2.1 to 0.34, P=0.007), d=0.62
• BPI interference: −1.5 (−2.4 to −0.57, P=0.002), d=0.72
• ODI: −9.8 (−15.3 to −4.4, P<0.001), d=0.79
• Five-chair-stand test: −7 (−13.7 to −0.37, P=0.04), d=0.49
• TUG test: −1.6 (−3.3 to 0.19, P=0.08); d=0.41
• 10 m walk test: 0.1 (−0.14 to 0.21, P=0.08), d=0.41–0.49

Between-group differences in score reduction

• BPI back pain: −0.85 (−1.4 to −0.25, P=0.006, R²=0.64)
• BPI leg pain: −1.1 (−1.9 to −0.2, P=0.009, R²=0.447)
• BPI pain interference: −1.3 (−2.1 to −1.4, P=0.005, R²=0.49)
• ODI: −9.4 (−14.9 to −4, P=0.001, R²=0.59)
• Five-chair-stand test: −4.3 seconds (−7.7 to −0.82, P=0.02, R²=0.52)
• TUG test: −1.8 seconds (−3.2 to −0.16, P=0.02, R²=0.62)
• 10 m walk test: m/s (0.008 to 0.18, P=0.07, R²=0.33)

Trial 3:

• In an Italian trial consisting of patients undergoing lumbar spinal fusion, the clinical effectiveness of CBT + exercise (two 1-hour psychologist-delivered individual CBT sessions per week over 4 weeks targeting catastrophizing and kinesophobia, in addition to exercise program) and exercise (five 90-minute physiotherapist-led exercise sessions per week for 4 weeks) were compared at 4 weeks.
• At 4 weeks post-surgery, all group effects and time effects significant at P<0.001 level.

Group × time effects

Effect size corresponds to difference between groups

• ODI: F=20.37, P<0.001, d=0.8
• NRS back: F=40.87, P<0.001, d=1.13
• NRS leg: F=12.32, P<0.001, d=0.62
• SF-36-BP: F=12.25, P<0.001, d=0.62

Trial 4:

• In an Danish trial consisting of patients undergoing lumbar spinal fusion patients, the clinical effectiveness of CBT (four 3-hour group CBT sessions in addition to usual care) and usual care (information on operation, anesthesia, medication, PO rehab, physical restrictions after surgery) were compared at 7 days post-surgery.
• PO mobility improved on day 3: Walk, P=0.02, Rise and sit from chair, P=0.0017, Get in and out of bed, P=0.0017). However, pain and rescue analgesic use were not statistically significant.

Trial 5:

• In the same Danish trial described above, a longer 3-month follow-up showed an improvement in disability (ODI), P=003; however, back pain and leg pain were not.

Low to high quality evidence demonstrated small to medium effects for some psychological outcomes at short-term and long-term follow-ups. Only short-term pain and disabilityrelated outcomes are reported below.

Pain outcomes:

• There was moderate quality evidence that physiotherapist delivered psychological intervention combined with physiotherapy decreased pain in the short-term (26 studies, n = 3024, MD= −0.37; 95% CI: −0.65 to −0.09; I²= 63%; P=0.008), although the effect size was small.
• No statistically significant difference was found for pain in the subgroup of patients with LBP only at the short-term (15 studies, n=1921; MD=−0.33; 95% CI: −069 to 0.03; I²=67%; P=0.07).
• No statistically significant difference was found for pain in the subgroup of patients with WAD and neck pain only at the short-term (7 studies, n=683; MD=−0.27; 95% CI, −0.81 to 0.26; I²=55%; P=0.31).
• No statistical difference was found for self-efficacy in the subgroup of patients with OA-related pain at short-term follow-up (4 studies, n=420; MD=−0.73; 95% CI: −1.61 to 0.15; I²= 70%; P= 0.10)

Disability Outcomes:

• Low-quality evidence showed that psychological interventions combined with physiotherapy was associated with a statistically significant but small improvement in disability compared with physiotherapy alone or usual care in the short term (29 studies, n=4249; SMD=−0.14; 95% CI: −0.26 to −0.01; I²=72%; P=0.03).
• No statistically significant difference was found for pain in the subgroup of patients with LBP only at the short-term (17 studies, n= 2298; SMD=−0.12; 95% CI, −0.31 to 0.07; I²= 80%; P= 0.21).
• No statistically significant difference was found for pain in the subgroup of patients with WAD and neck pain only at the short-term (9 studies, n= 1605; SMD=−0.14; 95% CI, −0.32 to 0.04; I²=60%; P= 0.12).
• No statistical difference was found for self-efficacy in the subgroup of patients with OA-related pain at short-term follow-up (3 studies, n=346; SMD=−0.20; 95% CI, −0.41 to 0.01; I²=0%; P= 0.06).

“The results indicate that psychological interventions delivered by physiotherapist show promise to improve health outcomes, particularly psychological outcomes, in musculoskeletal pain conditions” pg 1

“The results of the review of these 34 studies indicated that combined physiotherapy and psychological interventions delivered by physiotherapists shows promise to improve health outcomes, particularly psychological outcomes, in musculoskeletal pain conditions” pg18

• Based on meta-analysis of 2 studies comparing the effects of CBT versus other interventions (information booklet, didactic discussion, manual therapy) at short-term (6–13 weeks) and long-term follow-up (52 weeks), there was low quality evidence that CBT was statistically significantly better than other interventions for improving subacute neck pain (SMD −0.24, 95% CI: −0.48 to 0.00, I² = 7%, P = 0.05), while no difference was found in terms of disability (SMD −0.12, 95% CI: −0.36 to 0.12, I² = 0%, P = 0.31)
• One of the two included RCTs showed that the CBT group outperformed the no treatment group in terms of pain and disability (mean pain severity: possible range 0 to 6, MD −0.80, 95% CI: −1.27 to −0.33); Neck Disability Index MD −5.80, 95% CI: −10.52 to −1.08).
• The remaining RCT reported that CBT was better than manual therapy at improving pain and disability (Numerical Rating Scale MD 0.99, 95% CI 0.15 to 1.83; Neck Disability Index MD 2.42, 95% CI 0.52 to 4.32). For other outcomes (such as Tampa Scale of Kinesiophobia or the Pain Coping and Cognition List), there was no significant difference between groups.

Disability:

• The biopsychosocial multidisciplinary intervention group showed a statistically improvement in the adjusted analysis of the RMDQ outcome compared to the control group at 3 months (− 1.33 points, 95% CI: − 2.22 to − 0.45, p = 0.005) and at 12 months (− 1.11 points, 95% CI: − 2.08 to − 0.13, P = 0.027).
• An MCID (2.5 points from baseline as reported by the authors) was achieved in both groups, with a difference over 3.5 points in the intervention group compared with baseline at each time-point (3.8 RMDQ points at 3 months and 5.1 RMDQ points at 12 months).

Pain:

• With respect to the total intensity score of MPQ, a marginal difference was observed in the intervention group compared to the control group (− 0.69 points; 95% CI: − 1.41 to 0.02; p = 0.058) at 12 months, but not at 3 months (−0.49 points; 95% CI: −1.39 to 0.42; p = 0.294).
• The intervention group presented a statistically significant differences compared to the control group at 3 months for current intensity score (− 0.32 points; 95% CI: − 0.63 to − 0.02; p = 0.040) and for VAS score (− 0.77 points; 95% CI: − 1.53 to − 0.01; P = 0.046). However, no statistically significant differences were seen for either endpoints at 12 months (−0.18 points; 95% CI: − 0.43 to 0.08; p = 0.162 for current intensity score and (−0.27 points; 95%CI: − 0.88 to 0.34; P = 0.374 for VAS).

Quality of life:

• The outcome of SF-12 increased in both groups during the follow-up period, but no statistically significant differences between groups were observed on the physical domain (0.55 points, 95% CI: − 1.19 to 2.29, P = 0.520 and 0.53 points, 95% CI: −1.20 to 2.27, P = 0.532 at 3 and 12 months, respectively) and mental health domain (2.56 points, 95%CI: − 0.33 to 5.45, P = 0.082 and 1.48 points, 95% CI: −0.86 to 3.83, P = 0.206 at 3 and 12 months, respectively).

“A multidisciplinary biopsychosocial intervention in a working population with non-specific subacute LBP has a small positive impact on disability, and on the level of pain, mainly at short-term, but no difference on quality of life” pg 2

“The main conclusion of this study is that a multidisciplinary biopsychosocial intervention in a working population with non-specific subacute LBP has a small positive effect on disability and intensity of pain. Although greater in the intervention group, minimal clinically important differences were achieved in both groups. The results did not show any differences on quality of life” pg 8