Article Text

Download PDFPDF

It's time for change with the management of non-specific chronic low back pain
  1. Peter O'Sullivan
  1. Correspondence toProfessor Peter O'Sullivan, Professor/Specialist Musculoskeletal Physiotherapist, Curtin University, GPO Box U1987, Perth, WA 6845, Australia; p.osullivan{at}

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.

Low back pain (LBP) is the second greatest cause of disability in the USA.1 USA data supports that in spite of an enormous increase in the health resources spent on LBP disorders, the disability relating to them continues to increase.2 The management of LBP is underpinned by the exponential increase in the use of physical therapies, opiod medications, spinal injections as well as disc replacement and fusion surgery.2 This is maintained by the underlying belief that LBP is fundamentally a patho-anatomical disorder and should be treated within a biomedical model.1 This is in spite of calls over a number of years to adopt a bio-psycho-social approach, and evidence that only 8–15% of patients with LBP have an identified patho-anatomical diagnosis, resulting in the majority being diagnosed as having non-specific LBP.3 Of this population, a small but significant group becomes chronic and disabled, labelled non-specific chronic low back pain (NSCLBP), consuming a disproportionate amount of healthcare resources.4

  1. Over the past decade, the traditional biomedical view of LBP has been greatly challenged. This is a result of: the failure of simplistic single-dimensional therapies to show large effects in patients with NSCLBP58;

  2. the results of clinical trials testing commonly prescribed interventions demonstrating that no management approaches are clearly superior57 9;

  3. the stories of NSCLBP patients relating their own ongoing pain experiences of multiple failed treatments, conflicting diagnoses, lost hope and ongoing suffering10;

  4. the indisputable evidence supporting the multidimensional nature of NSCLBP as a disorder, where disability levels are more closely associated with cognitive and behavioural aspects of pain rather than sensory and biomedical ones11 12;

  5. positive outcomes in randomised controlled trials (RCTs) are best predicted by changes in psychological distress, fear avoidance beliefs, self-efficacy in controlling pain and coping strategies13 14;

  6. the evidence supporting the broad subgrouping of NSCLBP disorders on the basis of neuro-physiological,15 16 cognitive,17 physical factors18 and lifestyle behaviours.19 20

Underlying primary healthcare clinical practice has been simplistic biomechanical and structural models of LBP and pelvic girdle pain (PGP), which focus on structural diagnoses such as spinal and pelvic ‘instability’.21 22 These have been based on a belief that LBP and PGP is a result of structural (ie, degenerative), biomechanical and motor control deficits resulting in segmental or regional ‘instability’ of the lumbo-pelvic region.2124 It is now clear that there is little evidence (basic science and outcome studies) to support the view that ‘instability’ underpins the basis of disabling NSCLBP. There are no studies that demonstrate a clear relationship between spinal or pelvic mobility, degenerative processes, pain and disability.25 26 Similarly, common patho-anatomical findings such as degenerative disc disease, annular tears, fissures, facet joint arthrosis and disc bulges have been found not to be predictive of future LBP.26 This highlights the limitation of radiological imaging and spinal structure to provide clear meaning to people's experience of pain. Rather, factors such as depression,26 27 stress, cognitive and physical behaviours and lifestyle factors are more predictive of future LBP episodes.11 20

Yet in spite of this evidence, patients with disabling NSCLBP disorders continue to be provided with biomedical diagnoses and on the basis of these beliefs, people are prescribed with stabilising exercises, pelvic belts, supportive vests, spinal injections or even stabilisation surgery.1 2 23 28 These ‘magic bullet’ approaches, for some, may in fact have the potential to drive fear, abnormal body focus and reinforce pain-related movement and avoidance behaviours, hypervigilance, catastrophising, pain and disability fuelling the vicious cycle of pain.29

Diagnostic labels such as ‘instability’ should be reserved solely for ‘unstable fractures’ and ‘unstable spondylolisthesis’.30 The application of this diagnostic label to NSCLBP and PGP disorders is both inaccurate and potentially detrimental.21 29 31 This ‘belief system’, which I once advocated, has resulted in the development of an educational and management industry aimed at enhancing spinal stability for the prevention and treatment of NSCLBP, influencing the practice of physiotherapy, allied health as well as sports rehabilitation and training industries across the world.28 32 This approach commonly instructs patients to contract their ‘stabilising’ muscles (pelvic floor and transverse abdominal wall and lumbar multifidus) prior to spinal loading and during movement.6 9 23 33 In sports and gym rehabilitation settings, patients are frequently instructed to brace their abdominal wall and back muscles, to create more spinal ‘stability’ with the belief that ‘more stability is better’.

Yet these management approaches have not arrested the growing disability associated with NSCLBP.2 Although there is evidence for the efficacy of stabilising exercise for NSCLBP,34 a number of high-quality RCTs have demonstrated that specific spinal stabilising exercises for NSCLBP are not superior to other conservative approaches6 9 35 36, they have small effect sizes37 and they are only marginally better than placebo treatment consisting of detuned shortwave and ultrasound.33 Yet the benefits of this approach continue to be exclusively promoted, in spite of mounting scientific evidence that questions the underlying basis of this clinical belief system. This includes evidence that the motor control characteristics of non-specific low back pain (NSLBP) and NSCLBP commonly lie in findings of:

  1. increased co-contraction (stability) of trunk muscles38 39 and guarded spinal movement40;

  2. hyperactivity of trunk muscles (including muscles such as erector spinae, lumbar multifidus, pelvic floor and transverse abdominal wall) in NSCLBP and PGP disorders18 41 42;

  3. an inability of the back muscles to relax18 40;

  4. a tendency for earlier onsets of the antero-lateral abdominal wall muscles during rapid arm movements rather than timing delays43;

  5. in some cases, trunk muscle hypertrophy in muscles such as lumbar multifidus44 and quadratus lumborum45 have been documented in LBP sport populations.

Further to this, there is evidence for:

  1. a lack of association between muscle density (degeneration) of lumbar multifudus and LBP in a recent large population study46;

  2. a lack of association between changes in transversus abdominus muscle timing47 and lumbar multifidus cross-sectional area13 as a predictor of positive outcomes (disability levels) in RCTs.

This body of research challenges current practice and beliefs and is a counter view to previous literature. Some of the apparent conflict within the literature appears to have arisen where the results of studies with small sample sizes, investigating subjects with recurrent LBP48 49, have been extrapolated to the broad NSCLBP population without the results of these studies being reproduced in these populations or in larger groups.

The physiotherapy, manual therapy and medical professions have long focused on trying to find the magic ‘technique’, ‘muscle’, ‘injection’ or ‘surgical technique’ required to solve the problem of NSCLBP and PGP disorders. This reductionist approach to dealing with complex disorders in a simplistic manner clearly hasn't delivered for our patients50 and contradicts current knowledge that NSCLBP should be considered within a multidimensional bio-psycho-social framework. In fact, it has been proposed that single-dimensional approaches may in fact exacerbate chronic disorders reinforcing a cumulative feedback loop.29

In response to the calls to manage NSCLBP from a bio-psycho-social perspective, a number of RCTs have tested cognitive behavioural approaches to more effectively manage the disorder. Yet systematic reviews of these approaches have failed to demonstrate greater efficacy than other active conservative approaches in managing NSCLBP.51 Possible reasons for this failure may relate to the lack of patient-centred and targeted management52 as well as a failure to address other dimensions such as neuro-physiological factors and maladaptive lifestyle and movement behaviours known to be associated with NSCLBP disorders.18 20 30

There is strong evidence that NSCLBP disorders are associated with a complex combination of physical behavioural, lifestyle, neuro-physiological (peripheral and central nervous system changes), psychological/cognitive and social factors.12 20 30 These factors together have the potential to promote maladaptive cognitive behaviours (negative beliefs, fear, avoidance, catastrophising, hypervigilance),53 pain behaviours (pain communicative and avoidant behaviours)54 and movement behaviours,30 setting up a vicious cycle of pain sensitisation and reinforcing disability. Changes in immune and neuro-endocrine function linked to altered stress responsiveness coupled with activation of the pain neuro-matrix in the brain may result in tissue hyperalgesia and altered neuro-muscular responses.11 It is thought that these processes are mediated by environmental/genetic interactions.55

The balance and contribution of these different factors will likely vary for each individual. For example, it is known that not all NSCLBP disorders are associated with significant psychosocial factors.17 56 However, there is strong evidence that disability and factors such as sick leave are best predicted by factors such as negative back pain beliefs, fear and distress.17 57 Futhermore, psychological factors such as fear and catastrophising commonly associated with disabling pain have lifestyle, physical, neuro-muscular40 as well as neuro-biological consequences, highlighting that the mind and the body are inextricably linked.11

There is also growing evidence that NSCLBP disorders can be broadly categorised or subgrouped based on different psychosocial/coping behaviours,17 58 59 neuro-physiological characteristics,15 56 pain behaviours54 and movement behaviours,18 30 providing greater potential for targeting of multidimensional interventions.60 These broad subgroups, rather than being rigid entities which are characterised by prediction rules,61 may provide a framework for the clinician to tailor management to patients in a more targeted person-centred multidimensional manner.30 58 59

There is emerging evidence to support this view that patient-centred multidimensional targeted behavioural approaches have greater efficacy than current practice for the management of NSCLBP disorders in primary care settings. Asenlöf et al62 63 compared individually tailored treatment targeting activity levels, motor behaviour and cognitions, demonstrating superior outcomes to exercise therapy. A patient-centred multidimensional behavioural approach called ‘classification-based cognitive functional therapy’ that targets maladaptive cognitive, lifestyle, pain and movement behaviours was more effective (greater effect sizes) than manual therapy and exercise for localised NSCLBP.64 Hill et al65 employed a patient-centred stratification approach to target physiotherapy treatment based on psychosocial risk profile, demonstrating superior outcomes and cost saving over standard physiotherapy care. Further research into this patient-centred multidimensional approach is clearly required but recent evidence is encouraging for improved outcomes.

Other behavioural therapies such as mindfulness meditation,66 acceptance and commitment therapy,67 brain-directed therapies68 69 and targeted medical management70 hold hope for the multidisciplinary management of some of the highly complex and disabling central nervous system pain disorders.

In spite of this emerging evidence, recent research highlights that health professionals dealing with LBP disorders have difficulty accurately identifying psycho-social risk in their patients, limiting their capacity to target management.71 It appears that specific training in behavioural aspects of a patients presentation is required to enable health professionals to identify psycho-social risk factors and maladaptive movement behaviours from a clinical examination.72 There is also growing evidence to support the critical role that the quality of the therapeutic relationship plays in the management of pain disorders.73 Practitioner-related factors such as communication skills, empathy, level of confidence and beliefs have an important influence on patient outcomes and compliance to treatment.74 Conversely, patient beliefs and expectations also have a profound influence on health disorder outcomes.75

With all this in mind, the challenges for the future in more effectively dealing with NSCLBP disorders are likely to involve primary healthcare providers shifting rigidly held biomedical beliefs and developing greater skills and knowledge across a number of domains. These skills are likely to include:

  1. Greater understanding of the complex multidimensional nature of NSCLBP.

  2. Developing diagnostic skills to clearly differentiate specific pathology as a driver of pain from NSLBP disorders.

  3. Develop more effective communication skills utilising empathy, reflective questioning and motivational interviewing techniques in order to listen to the patients' story and explore their pain beliefs, fears, coping strategies, life stresses, psycho-social factors, pain behaviour, impairments and goals. This allows for the development of an effective therapeutic relationship and the accurate interpretation of clinical information within a bio-psycho-social framework in order to identify the primary drivers of pain and disability. This in turn provides the capacity to clearly outline the vicious cycle of the disorder in a patient-centred way.

  4. Identification of maladaptive cognitive behaviours (negative beliefs, stress responsiveness, provocative coping strategies, hypervigilance, fear, catastrophising, anxiety, depression etc).

  5. Identifying neuro-physiological processes such as central and peripheral sensitisation.

  6. The analysis and interpretation of pain communicative and avoidant behaviours54 and movement and postural behaviours30 in order to determine adaptive (protective) from maladaptive (provocative) behaviours.30

  7. Synthesising and interpreting clinical information across multiple domains.

  8. Developing multidimensional and flexible interventions that target maladaptive cognitive, lifestyle, pain and movement behaviours in an integrated manner.

  9. Facilitation of behavioural change in patients by enhancing clinical skills such as empathy, motivation, support, creativity, goal setting, flexible person-centred functional rehabilitation programmes and clear feedback.

  10. Developing clear multidisciplinary approaches to management where indicated.

  11. Developing a broad framework for subgrouping of NSCLBP patients from a multidimensional perspective.17 30 52 58 This will allow the broad categorisation of LBP disorders based on the presence of dominant psycho-social, neuro-physiological, lifestyle and movement behaviours that act as drivers for the disorder.

  12. Adopting the routine use of screening tools in clinical practice in order to identify risk and targets for change to better direct management.65 76

This approach will likely focus less on treating the structure or signs and symptoms of a disorder in NSCLBP disorders and more on targeting the different combinations of beliefs, cognitive, pain, lifestyle and movement behaviours that underlie and drive disorders. Implementation of this approach will require a paradigm shift in ‘beliefs’ of health professionals in terms of how we understand and deal with NSCLBP disorders. This will involve abandoning ineffective practices, learning new skills, adopting and integrating new approaches. This new knowledge and skill needs to be trained at undergraduate and graduate levels and promoted actively within the professions that deal with these disorders.77 There is also a mandate to educate the public in order to reinforce more positive back pain beliefs to reduce the burden for both individuals and society.78 This will invariably lead to health insurers abandoning the ongoing funding of non-efficacious treatment approaches.

Further research is clearly needed to better identify the underlying mechanisms associated with disabling NSCLBP disorders and their development across the lifespan. This will likely involve a greater understanding of genetic/environmental interactions associated with the development of the nervous system, tissue sensitisation and associated maladaptive behaviours, tracking from early life to adolescence and into adulthood. Developing a greater understanding of those people resilient to these disorders may also be illuminating. Early screening and targeted management of risk groups, based on the identification of the mechanisms that drive them, may aid in the prevention of pain chronicity and disability. Innovative multidimensional, patient-centred and targeted approaches to management for these complex disorders need to be further developed and adequately tested.

Characteristics such as hope, positive help seeking and adaptability are traits of resilience that we need to equip our patients with, who suffer with disabling NSCLBP.12 Adopting a positive multidimensional perspective of health that is person focused may allow us to view NSCLBP in a new light, providing hope for our patients and an environment for innovation, discovery and change.


The author would like to acknowledge the support of Joao Paulo Caneiro in the preparation of this manuscript.


View Abstract


  • Competing interests None.

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