Article Text

Download PDFPDF

Effectiveness of extrinsic feedback for management of non-specific low back pain: a systematic review protocol
  1. Daniel Cury Ribeiro1,
  2. Aleksandra Katarzyna Mącznik1,
  3. Stephan Milosavljevic2,
  4. J Haxby Abbott3
  1. 1 School of Physiotherapy, Centre for Health, Activity, and Rehabilitation Research, University of Otago, Dunedin, New Zealand
  2. 2 School of Rehabilitation Science, College of Medicine, University of Saskatchewan, Saskatoon, Saskatchewan, Canada
  3. 3 Dunedin School of Medicine, University of Otago, Dunedin, New Zealand
  1. Correspondence to Dr Daniel Cury Ribeiro; daniel.ribeiro{at}


Introduction Low back pain is the greatest cause of years lived with disability worldwide and is linked with high societal and economic burden. Neuromuscular control impairments are a common clinical presentation in patients with non-specific low back pain. Musculoskeletal physiotherapists commonly use feedback as a part of the management of low back disorders. This systematic review will aim to assess the effectiveness of extrinsic biofeedback for reducing pain, disability and recurrence of pain in patients with non-specific low back pain.

Methods and analysis Systematic searches will be performed in CINAHL, Embase, Medline, PsycInfo, Scopus and Web of Science. We will include randomised controlled trial studies, if the study recruited patients with non-specific low back pain; compared extrinsic feedback versus either placebo or control; another intervention; or in addition to an intervention versus that intervention alone; and have used pain, disability scores or low back pain recurrence as outcome measures. We will exclude studies with designs other than randomised controlled trials. We will assess the risk of bias within included studies using the PEDro scale, and the strength of evidence using the Grades of Recommendation, Assessment, Development and Evaluation approach.

Ethics and dissemination Ethical approval and patient consent are not required since this is a systematic review based on published studies. The results of this study will be published in an international peer-reviewed journal.

PROSPERO registration number CRD42017077888

  • low back pain
  • systematic review
  • protocol
  • feedback
  • musculoskeletal disorders
  • rehabilitation medicine

This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Strengths and limitations of this study

  • Comprehensive and exhaustive search for relevant studies from several databases.

  • This review is limited to evidence from randomised controlled trials.

  • No language restrictions will be imposed.

  • The number of articles identified and included in this review may prevent us from conducting meta-analysis which might limit the conclusions that can be drawn from our findings.


Low back pain (LBP) is the most common musculoskeletal disorder, and the leading cause of years lived with disability.1 2 The accepted clinical subcategory of non-specific LBP is a multifactorial disorder and defined as ‘pain and discomfort, localised below the costal margin and above the inferior gluteal folds, with or without referred leg pain’.3 The 1-month prevalence of LBP is 23.2%.2 Patients with non-specific LBP commonly present with slow recovery and may not fully recover within 12 months since first episode of LBP.4

Non-specific LBP is influenced by biological, psychological and social factors.5 These three factors interact and impact on symptoms onset, recovery and clinical outcomes of patients with LBP.6 Biological factors include impaired motor control,7 delayed muscle activity of lumbopelvic muscles8 and reduced postural strategies for maintaining balance.9 Psychological factors (eg, fear avoidance, negative beliefs and depressive symptoms) may hinder recovery10 and may be associated with higher pain levels.11 Social factors such as family environment, socioeconomic and educational status and religion may modulate pain reported by patients with LBP.12

Management of non-specific LBP commonly focuses on education, pain control13 and exercise therapy targeting neuromuscular impairments.14 When targeting neuromuscular impairments, musculoskeletal physiotherapists may use extrinsic feedback to help patients improve movement control and awareness of lumbopelvic muscles.15 Extrinsic feedback can be defined as any form of information provided to the patient that originates from an external source (eg, mirror, pressure biofeedback, tactile or verbal input by the clinician).16 17 The way extrinsic feedback is provided to patients can hinder or improve motor control.18 Our previous review assessed how extrinsic feedback was provided to patients in trials and found that the majority of the studies did not adopt ideal forms of feedback provision.17

There are different ways of providing extrinsic feedback, and these can be grouped into two categories in relation to their content and timing characteristics.16 17 19 Content characteristics refer to the focus of the intervention.18 20 For example, how feedback is presented to patient (eg, absolute scores, average scores, general feedback about how movement is being and how it should be performed).18 20 Timing characteristics refer to when the feedback is provided to patients (eg, simultaneously to exercise execution, after the exercise has been executed, or provided when the patient requests this information).18 20 21 According to previous research on ‘motor control and learning’, there are some content and timing characteristics that will enhance/optimise motor control.18 20 21 We present a description for content and timing characteristics of extrinsic feedback in table 1.

Table 1

Description of content and timing characteristics for extrinsic feedback provision

A recent review reported that extrinsic feedback combined with physiotherapy compared with physiotherapy intervention alone was superior for pain reduction, but was no different to physiotherapy intervention alone for disability levels in patients with neck pain.19 Currently, the effectiveness of extrinsic feedback for the management of patients with non-specific LBP is unknown. The aim of the proposed systematic review is to assess the effectiveness of extrinsic biofeedback for reducing pain, disability and recurrence of pain in patients with non-specific LBP.


Patient and public involvement

Patients and/or public were not involved in the development of this research project.


This study will be a systematic review with meta-analysis, if data allow, and will be based on the Preferred Reporting items for systematic Reviews and Meta-Analyses Protocols checklist.22

Types of studies

We will include randomised controlled trials, and exclude narrative reviews, systematic reviews, non-randomised clinical trials, cross-sectional studies and observational studies. We will not impose language restrictions.

Types of participants

Studies must have recruited symptomatic individuals with non-specific LBP (acute, subacute or chronic), aged between 18 and 65 years. We will exclude studies involving patients with specific or systematic diseases (eg, tumours, cauda equina syndrome, fracture, inflammatory arthropathy).

Types of interventions

Any interventions that include the use of any form of extrinsic biofeedback (eg, verbal, tactile, pressure biofeedback, electromyography feedback, body positional biofeedback or other types of feedback) will be included in this review. We will exclude studies focusing on behavioural feedback or ergonomic training. Studies must include at least one of the following comparators:

  • Extrinsic feedback versus placebo or control.

  • Extrinsic feedback versus another form of intervention.

  • Extrinsic feedback+intervention versus intervention alone.

Types of outcome measures

We will consider the following primary outcome measures: (1) pain must be measured with Visual Analogue Scale, Numeric Pain Scale or any other validated instrument; (2) disability levels must be measured with validated instruments (eg, Oswestry, Quebec Back Pain Disability Scale, the Roland-Morris Disability Questionnaire); (3) recurrence of LBP (as reported by patients or assessed by a clinical researcher). We will consider motor performance tests as secondary outcomes. Motor performance must be assessed using a form of biofeedback or electromyography pre-intervention and postintervention.


Systematic searches will be performed in the following databases from the inception: CINAHL, Embase, Medline, PsycInfo, Scopus and Web of Science. Search strategy is presented in table 2. This search strategy was developed in consultation with a health sciences librarian, tested and used for a review previously conducted by our research team.17

Table 2

Search strategy

After the exclusion of duplicates, two independent reviewers will conduct title, abstract and full-text screening against inclusion criteria. If any disagreement persists, a third reviewer will adjudicate. The reference lists of included studies will be screened for additional relevant studies.

Data extraction and management

Data will be extracted by two reviewers independently and compared. If any disagreements persist and cannot be resolved by consensus, a third reviewer will be consulted. We will extract the following data: authors, year of publication, country of origin, study design, study purpose, experimental and comparison interventions and their characteristics, number of participants in each group and their characteristics, frequency of the interventions (if applicable), follow-up intervals (if applicable), drop-out rates, outcomes measures, main findings and authors’ conflict of interest.

Risk of bias within included studies and quality of evidence assessment

The risk of bias within included studies will be assessed using the PEDro scale. Reporting of the following aspects will be assessed: eligibility criteria, random allocation, concealed allocation, similarity between groups at baseline, blinding of subjects, therapists and assessors, attrition rate <15%, analysis by intention to treat, between-group comparison, and both point estimate and variability measures.

Data synthesis and analysis

If possible, we will use RevMan statistical software V.5.3 (The Nordic Cochrane Centre) for conducting the meta-analysis using a random-effects model. For the purpose of this review, outcome measures will be categorised into the following based on time points of assessment: immediate (within 2 weeks of the intervention delivery), short-term (2–13 weeks after intervention delivery), medium-term (14–50 weeks after intervention delivery) and long-term effects (51 or more weeks after intervention delivery). 

For continuous data, we will calculate the mean difference (MD), and 95% confidence intervals (CI) if outcome measure scales are the same. In the case of different outcome measure scales, we will calculate the standardized mean difference (SMD) and 95% CI. For the purpose of this review, the effect size will be categorised as: small=SMD ranging from 0 to 0.2; medium=SMD ranging from >0.2 to <0.5; and large=SMD ranging from >0.5 to 0.8.23 24

We will assess heterogeneity using I2 statistics25 26 and will consider heterogeneity to be substantial if I2 ranges from 50% to 90%.27 Sensitivity analyses will be conducted by the quality of studies and by the length of follow-up. If possible, we will conduct subgroup analysis based on the type of the disorder (ie, acute, subacute or chronic non-specific LBP).

The strength of evidence will be assessed using the Grades of Recommendation, Assessment, Development and Evaluation (GRADE) approach.28 The GRADE approach uses four quality levels: high, moderate, low and very low. The strength of evidence will be downgraded by one level according to the following criteria: (1) limitations in the design and implementation of available studies suggesting risk of bias, (2) indirectness of evidence, (3) unexplained heterogeneity or inconsistency, (4) imprecision of results and (5) high probability of publication bias.

If meta-analysis is not possible, we will present a narrative synthesis of the findings. In this case, quantitative findings for each study will be descriptively reported and summarised.

Ethics and dissemination

This systematic review has been prospectively registered at the PROSPERO (CRD42017077888). The results of this review will be published in an international peer-reviewed journal.


The authors thank Mr Richard German (Faculty Librarian) in assisting with the electronic search strategy for this review, and the financial support from the Centrefor Health, Activity and Rehabilitation Research (CHARR), School ofPhysiotherapy – University of Otago.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.


  • Contributors DCR is the leading researcher and was responsible for conceiving this study, designing the protocol, preparing the search strategy and coauthoring the first draft of this manuscript. He has accepted the final version of the manuscript for submission. AKM contributed to the design of the review protocol, preparing the search strategy and coauthoring the first draft of this manuscript. JHA has contributed to the design of the review protocol and to the manuscript. SM contributed to the design of the review protocol and to the manuscript. DCR (principal investigator), JHA (coinvestigator) and SM (coinvestigator) secured the Health Research Council Emerging Researcher First Grant (15/527). All authors have contributed to the conception and design of the study protocol, development of the search strategy, the establishment of the inclusion and exclusion criteria, data extraction criteria, analyses and interpretation, and accepted the final version of the manuscript for submission.

  • Funding DCR (principal investigator), JHA (coinvestigator) and SM (coinvestigator) are supported by the Health Research Council Emerging Researcher First Grant (15/527).

  • Competing interests None declared.

  • Patient consent Not required.

  • Ethics approval Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.