Intended for healthcare professionals

CCBYNC Open access

Rapid response to:

Research

Psychological interventions for chronic, non-specific low back pain: systematic review with network meta-analysis

BMJ 2022; 376 doi: https://doi.org/10.1136/bmj-2021-067718 (Published 30 March 2022) Cite this as: BMJ 2022;376:e067718

Rapid Response:

Re: Psychological interventions for chronic, non-specific low back pain: systematic review with network meta-analysis

Dear Editor,

Moore and colleagues recently published a rapid response to our systematic review,[1] raising concerns regarding our inclusion of trials deemed as untrustworthy, our sample sizes, and our discussion of comparative literature. We reinforce in the points below that the main conclusions of our published review are not erroneous, our analyses were adequately powered, and our selection of comparative literature was appropriate.

1. Untrustworthy trials:
Subsequent to the publication of our network meta-analysis of psychological interventions for chronic, non-specific low back pain,[1] O’Connell et al[2] published a manuscript which raised concerns regarding the untrustworthiness of trials conducted by the Monticone group. Two trials from the Monticone group were included in our network meta-analysis.[3, 4] Hence, subsequent to the publication of O’Connell et al’s findings, we performed additional sensitivity analyses excluding the two trials led by the Monticone group.[3, 4] Re-calculated effect sizes were compared with our published results and are summarised below.[1] Removal of the two Monticone trials precluded sensitivity analysis of fear avoidance at the long-term follow-up, as the original pairwise meta-analysis only included three trials.

The sensitivity analyses revealed:
- no change in the previously observed moderate and clinically important effects of pain education delivered with physiotherapy care for physical function at post-intervention (standardised mean difference 0.55, 95% confidence interval 0.18 to 0.91);
- no change in the observed large and clinically important effects of behavioural therapy delivered with physiotherapy care for pain intensity at post-intervention (0.93, 0.18 to 1.68);
- no change in the observed large and clinically important effects of pain education delivered with physiotherapy care for pain intensity at post-intervention (0.86, 0.40 to 1.33);
- no change in the observed large effects of pain education delivered alone (0.92, 0.25 to 1.59) or with physiotherapy care (0.82, 0.49 to 1.15) for fear avoidance at post-intervention (our conclusions were drawn from the sensitivity analysis excluding patients with leg pain). There is, however, inconsistency in the sensitivity analysis results for fear avoidance at post-intervention, which could not be resolved without changes to the network structure.
- no change in the small and insignificant effect of cognitive behavioural therapy delivered with physiotherapy care for fear avoidance at post-intervention (0.39, -0.04 to 0.82) (our conclusions were drawn from the sensitivity analysis excluding patients with leg pain). There is, however, inconsistency in the sensitivity analysis results for fear avoidance at post-intervention, which could not be resolved without changes to the network structure.
- a reduction from a large to moderate effect of cognitive behavioural therapy delivered with physiotherapy care on physical function (0.54, 0.23 to 0.85) and pain intensity (0.52, 0.05 to 0.99) at post-intervention;
- the large effect of counselling delivered with physiotherapy care changed from insignificant to significant for physical function at post-intervention (0.86, 0.20 to 1.52);
- the large effect of counselling delivered alone changed from insignificant to significant for pain intensity at post-intervention (-1.75, -3.25 to -0.26).

2. Sample size:
In network meta-analysis, data from all patients are analysed simultaneously through a connected network of treatment nodes. In our primary analysis, the total sample size, number of treatment nodes, and number of trials analysed simultaneously at post-intervention were n=6986 (17 treatment nodes across 49 trials) for physical function (Figure 4) and n=6963 (17 treatment nodes across 56 trials) for pain intensity (Figure 5).[1] In these figures, ‘No of patients’ refers to the number of patients allocated to a particular intervention (treatment node) across the original trials forming the network. Hence, our analyses were sufficiently powered. Furthermore, the 240 patients from the Monticone group trials represented only 3% of all patients analysed for both primary outcomes at post-intervention. After exclusion of these trials, the total sample size at post-intervention was n=6746 for physical function and n=6723 for pain intensity.

3. Comparisons with previous literature:
In our review,[1](p.18) we meticulously compared our findings with Henschke et al’s review (2010),[5] not an older 1999 review,[6] as it is the most recent and relevant Cochrane review specifically including people with chronic low back pain. In contrast, only 16 of 75 studies included in William et al’s 2020 Cochrane review[7] consisted of people with chronic low back pain alone. The remaining studies included people with fibromyalgia, mixed persistent pain, and other chronic pain conditions (e.g., temporomandibular disorder).[7] As sub-group analyses for low back pain were not performed,[7] comparison of findings with the William et al review are not appropriate.

Conclusion
The conclusions of our review remain substantially unchanged: compared with physiotherapy care, pain education and behavioural therapy, delivered with physiotherapy care, provide the most sustainable treatment effects in people with chronic, non-specific low back pain.[1](p.21) The effect of cognitive behavioural therapy delivered with physiotherapy care attenuated from large to moderate for physical function and pain intensity at post-intervention, when trials of concern were excluded; however, the effect remained statistically and clinically important.

References:
[1] Ho EK-Y, Chen L, Simic M, et al. Psychological interventions for chronic, non-specific low back pain: systematic review with network meta-analysis. BMJ 2022;376:e067718.
[2] O'Connell NE, Moore RA, Stewart G, et al. Investigating the veracity of a sample of divergent published trial data in spinal pain. Pain 2022. [Online ahead of print]. doi:10.1097/j.pain.0000000000002659
[3] Monticone M, Ferrante S, Rocca B, et al. Effect of a Long-lasting Multidisciplinary Program on Disability and Fear-Avoidance Behaviors in Patients With Chronic Low Back Pain: Results of a Randomized Controlled Trial. Clin J Pain 2013;29:929-38
[4] Monticone M, Ambrosini E, Rocca B, et al. Group-based task-oriented exercises aimed at managing kinesiophobia improved disability in chronic low back pain. Eur J Pain 2016;20:541-51.
[5] Henschke N, Ostelo RW, van Tulder MW, et al. Behavioural treatment for chronic low‐back pain. Cochrane Db Syst Rev 2010;(7):CD002014
[6] Morley S, Eccleston C, Williams A. Systematic review and meta-analysis of randomized controlled trials of cognitive behaviour therapy and behaviour therapy for chronic pain in adults, excluding headache. Pain 1999;80:1-13
[7] Williams AC, Fisher E, Hearn L, et al. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Db Syst Rev 2020;8(8):CD007407

Competing interests: No competing interests

13 May 2022
Emma K Ho
PhD candidate
Dr Lingxiao Chen: University of Sydney, School of Health Sciences, Sydney Musculoskeletal Health, Sydney, Australia; A/Prof Claire E Ashton-James: University of Sydney, Sydney Medical School and Kolling Institute, Sydney, Australia; A/Prof Jill A Hayden: University of Dalhousie, Faculty of Medicine, Nova Scotia, Canada; Prof Manuela L Ferreira: University of Sydney, School of Health Sciences, Sydney Musculoskeletal Health, Sydney, Australia; Prof Paulo H Ferreira: University of Sydney, School of Health Sciences, Sydney Musculoskeletal Health, Sydney, Australia.
University of Sydney, School of Health Sciences, Sydney Musculoskeletal Health, Sydney, Australia
Faculty of Medicine and Health, Susan Wakil Health Building, University of Sydney, NSW 2006, Australia