Best Practice & Research Clinical Rheumatology 27 (2013) 699–708

Contents lists available at ScienceDirect

Best Practice & Research Clinical Rheumatology journal homepage: www.elsevierhealth.com/berh

11

Low back pain research – Future directions Danielle A. van der Windt, PhD, Professor of Primary Care Epidemiology *, Kate M. Dunn, PhD, Reader in Epidemiology 1 Arthritis Research UK Primary Care Centre, Research Institute for Primary Care & Health Sciences, Keele University, Staffordshire ST5 5BG, UK

a b s t r a c t Keywords: Back pain Public health Prognosis Intervention studies Research priorities

Low back pain is a challenge for clinicians and researchers, due to the large variability in clinical presentation, lack of consensus regarding diagnostic criteria or clinical classification; wide variation in course and prognosis, and limited success in identifying effective treatments. However, increasing research efforts has generated an expanding body of evidence on the epidemiology, prognosis and treatment of back pain. This paper presents four key developments in research and clinical practice, and describes how these can influence the future direction of back pain research: (1) the increasing awareness of the impact of low back pain on population health; (2) new approaches to describing and investigating course and prognosis of back pain; (3) the need to better understand the bio-psycho-social mechanisms or pathways that explain impact and long-term outcomes in order to inform intervention research; and (4) the potential for stratified models of care to improve patient outcomes and efficiency of care for back pain. Ó 2013 Elsevier Ltd. All rights reserved.

Introduction Research into the epidemiology and treatment of low back pain has exponentially increased over the past few decades, and at least matches the increase in output in other research fields. Fig. 1 shows the increase in the number of randomised controlled trials in back pain patients (cited in Medline) over the past 30 years, which increased from less than 30 in the period from 1983 to 1987 to more than 650

* Corresponding author. Tel.: þ44 01782 734830; fax: þ44 01782 733911. E-mail addresses: [email protected] (D.A. van der Windt), [email protected] (K.M. Dunn). 1 Tel.: þ44 01782 734703; fax: þ44 01782 733911. 1521-6942/$ – see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.berh.2013.11.001

700

D.A. van der Windt, K.M. Dunn / Best Practice & Research Clinical Rheumatology 27 (2013) 699–708

Primary Care / Community 700 600 500 400 300 200 100 0

Fig. 1. Randomised controlled trials in low back pain (publications cited in Medline); trials in primary care, workplace or community settings are highlighted.

over the past 5 years. The graph also highlights the increasing proportion of trials conducted in primary care, workplace settings and the community, emphasising the rising awareness that low-back pain has wide impact in the community, and is mostly managed within primary and occupational healthcare. Low back pain is a challenge to both clinicians and researchers, due to the wide variability in clinical presentation; lack of consensus regarding diagnostic criteria or clinical classification; large variation in course and prognosis; and limited success in identifying effective treatments. However, increasing research efforts has generated an expanding body of evidence on the epidemiology, prognosis and treatment of back pain, which has led to new insights and changes in thinking regarding the classification and management of back pain. Four themes emerge from these developments that cut across several topics: (1) the increasing awareness of the impact of low back pain on population health; (2) the realisation that the current classification of low back pain (in terms of acute/subacute or chronic pain) is not adequate, leading to the proposition of new approaches to describing and investigating the course and prognosis of back pain; (3) the need to better understand the mechanisms or pathways that explain impact and long-term outcomes in patients with back pain; and (4) the potential of individualised or stratified models of care to improve patient outcomes and efficiency of care for back pain. In this chapter we will discuss the developments across these four themes, and how these may influence future direction in back pain research. Increasing awareness of the population impact of low-back pain The large personal, economic and societal burden of low-back pain has been emphasised in numerous publications, but until recently good quality and up-to-date systematic reviews of the prevalence and impact of low back pain were lacking. The recently published meta-analysis by Hoy et al. [1] summarised evidence from 165 studies in 54 countries, presenting a pooled estimate of the mean (SD) point prevalence of activity-limiting low back pain of 11.9 (2.0)%, and 1-month prevalence 23.2 (2.9)% after adjusting for methodological variation between studies. The authors emphasise the heterogeneity between studies in terms of variation in the definition of back pain, risk of bias (representativeness of the sample and response rates), and study setting. They highlight the fact that the evidence is still very much weighted towards studies conducted in developed countries and adults of working age, with available studies showing a lower prevalence of back pain in developing countries and in older people. The lower prevalence of back pain in low- and middle-income countries could be the result of higher levels of physical activity, higher pain tolerance, or shorter height in these populations [2], but differences in health priorities, limited access to healthcare, or lack of health insurance or disability compensation systems may also lead to a lower reporting of back pain in these populations. Other studies have also reported a decreasing prevalence of back pain in older populations [3],

D.A. van der Windt, K.M. Dunn / Best Practice & Research Clinical Rheumatology 27 (2013) 699–708

701

but as discussed by Dionne et al. [4] this may be explained by a range of factors, including decreased pain perception or increased tolerance or acceptance of pain in older age, the accumulation of other health problems which may take priority, increasing influence of mental health problems such as depression or cognitive impairment, and exclusion from studies of individuals living in nursing homes. Studies investigating pain in older people indicate that the incidence of more severe, disabling pain continues to increase in older age [5,6], and that back pain has larger impact in older age [7]. Awareness of the public health impact of back pain has been boosted by the recent publication of the results of the Global Burden of Disease (GBD) projects, which featured back pain as the condition with the highest impact in terms of years lived with disability [8]. Buchbinder et al. [9] have discussed the difficulties of defining impact and constructing disability weights for low back pain in the GBD projects, which were related to inconsistent definitions of back pain, limited data from developing countries, lack of data on incidence, duration and recurrences of back pain, and limited data on the distribution of severity of back pain in populations. This led to wide uncertainty of estimates, and only allowed the use of a definition of back pain in terms of duration (more or less than 3 months) and presence of leg pain. Buchbinder et al. [9] highlight the importance of more sophisticated descriptors of back pain for population-based studies, and in previous work have identified dimensions of back pain impact that are rarely measured in population-based research, such as loss of independence, worry about the future, and negative or discriminatory actions by others [10]. Re-defining back pain The classification of back pain has posed many challenges, given the fact that a specific pathoanatomical cause of the back pain can only be established in a minority of patients. The large majority of patients with ‘non-specific’ low back pain form a heterogeneous group with widely varying presentations of the condition. A classification based on duration of the current episode (acute: less than 6 weeks; sub-acute: 6–12 weeks; chronic: more than 12 weeks) has been used for many years to distinguish between patients with favourable or poor prognosis, to present treatment recommendations in guidelines, to define eligibility for trials, or to classify back pain studies in systematic reviews. However, realisation has grown that a classification based on episode duration alone does not provide the information required to describe the impact of back pain, accurately estimate prognosis [11] or to make optimal treatment decisions [12]. Dunn et al. [13] demonstrate that evidence from prospective cohort studies does not support a model of back pain characterised by a series of individual, unrelated episodes, but more clearly points towards back pain as a long-term condition. A recent systematic review of the natural history of back pain, which included seven population-based studies with up to 28 years follow-up concluded that the status of low-back pain in individuals is relatively stable over time, with the majority reporting persistent or recurrent symptoms over long-periods of time [14]. These findings are confirmed by recent studies investigating trajectories of back pain based on monthly or weekly assessments of pain and disability, which identified distinct back pain trajectories of differing severity and impact, but most of which are stable over long periods of time with only mild fluctuations [15–19]. Although the course of back pain has mainly been studied in adults, back pain may have its onset early in life. Based on 59 studies Calvo-Muñoz et al. [20] estimated the mean point prevalence of back pain in children and adolescents at 12% (95% CI: 90. to 15.9%), with increasingly higher prevalence rates being reported in more recent and better quality studies. The early onset of the condition and its long-term course confirm the need for a life-time perspective when investigating and managing back pain, and re-defining back pain as a long-term condition rather than managing individual episodes of back pain. Back pain can be presented along a continuum from incidental, mild episodes of pain to a chronic health condition strongly interfering with daily life. People who experience multiple episodes of back pain seem to accumulate risk of chronic disabling pain, and appear more susceptible to the influence of a range of prognostic factors. For example, in a population-based cohort study with 12 years follow-up, Brage et al. [21] found that emotional distress was a predictor of long-term low back disability, but only in individuals with previous back pain. Accumulation over the life course of pain episodes, but also of exposures to physical, psychological, social or environmental stressors (allostatic load [22]) may all contribute to the risk of developing chronic disabling back pain [13,23–25]. A challenge for research is

702

D.A. van der Windt, K.M. Dunn / Best Practice & Research Clinical Rheumatology 27 (2013) 699–708

to investigate this accumulating load over the life-course, and how this may be balanced by individuals’ resources (resilience) to deal with such adversities [26–28]. Such knowledge would provide better insight into the reasons why some individuals are more likely than others to experience more chronic or severe trajectories of back pain. Back pain is often more than just pain in the back. Hartvigsen et al. [29] summarise the findings from numerous studies which indicate that the majority of people with back pain have pain elsewhere, and that the extent of pain is strongly associated with reduced physical functioning, feelings of anxiety and depression, work absence, and reduced quality of life. For example, a Norwegian survey showed that only 16.8% of responders (n ¼ 3179) reported localised musculoskeletal pain, whereas 53% reported pain in more than one site. [30] Using data from a national comorbidity survey (n ¼ 5692) Von Korff et al. [31] found that the prevalence of chronic back pain was more than three times as high among persons who reported other chronic pain conditions as those without these conditions: 34.1 vs 9.6%. If multisite pain is that common, the location of pain, whether in the back or elsewhere may be of little importance when assessing the consequences of pain. Using data from a population-based study in the Netherlands [32] Fig. 2 presents the proportion of individuals with increased anxiety and depressive symptoms among those with either back pain (n ¼ 256), headache (n ¼ 287), abdominal pain (n ¼ 96), or hand pain (n ¼ 129). These were mutually exclusive groups, although they could have other co-occurring symptoms. The results show similar levels of anxiety and in particular depression regardless of the location of pain. Distress was more strongly associated with the extent of pain (widespread pain) than with pain location. Other studies have also shown similarities across regional pain problems in terms of the course of pain and the type of prognostic factors, identifying generic predictors of outcome across different regional musculoskeletal pain conditions [33,34]. A prognostic score for estimating risk of chronic disabling pain developed originally in back pain patients [35] has been shown to accurately predict outcome in people with knee pain [36] and across a range of musculoskeletal pain sites [37]. Hartvigsen et al. [29] therefore propose a stepwise approach to the diagnostic classification and management in patients with musculoskeletal pain, regardless of pain location, of which clinical and cost-effectiveness could be investigated in future research. Understanding the complexity of back pain There has been increasing interest in the role of psychological and social factors in the onset and progression of low-back pain over the past 30 years, and this has resulted in an increasing number of trials investigating interventions that assess or address psychosocial factors in back pain patients (see Fig. 3). However, the results of these trials have often been disappointing, showing only small to moderate effects of a wide range of treatments for back pain [e.g. Refs. [38–40]]. anxiety

depression

45 40 35 % 30 25 20 15 10 5 0

Fig. 2. Increased anxiety and depression scores (HADS > 7) in people (population-based sample in the Netherlands) reporting no pain (n ¼ 346), back pain (n ¼ 256), headache (n ¼ 287), abdominal pain (n ¼ 96) or hand pain (n ¼ 129) or chronic widespread pain (n ¼ 322).

D.A. van der Windt, K.M. Dunn / Best Practice & Research Clinical Rheumatology 27 (2013) 699–708

703

psychosocial factors 700 600 500 400 300 200 100

0

Fig. 3. Randomised controlled trials in low back pain (publications listed in Medline); trials addressing or assessing psychosocial factors have been highlighted.

Several authors have argued that we should move beyond the ‘black box’ approach of pragmatic trials and increase our understanding of theory and the mechanisms underlying proposed interventions. Pincus et al. [41] argue that sub-optimal results of intervention studies may be the results of low competency or fidelity in delivery of interventions, or inadequate matching of interventions to specific patient problems. They feel that strong theoretical guidance is essential in the design, selection, and delivery of treatment methods that specifically target known processes of pathology. When delivering an intervention study, additional data collection before, during and after the treatment period will allow a more thorough evaluation of the processes that may explain response to treatment in trial participants (mediation analysis). The methods to support treatment mediation analysis are still being developed, but offer strong opportunities to better understand the treatment processes and mechanisms and have the potential to further improve the design or delivery of interventions for back pain [42–45]. A strong theoretical or empirical basis for intervention studies requires sufficient knowledge of the role of biological, psychological and social factors in the onset and progression of low-back pain. Research into social factors is limited compared with the large body of evidence on biological and psychological aspects of back pain. With the exception of the psychosocial and physical work environment which has been extensively investigated in occupational settings [e.g. Refs. [46–49]], there is less evidence on the importance of social support and influences of partner, family and friends, social roles, and social factors in communication with health professionals [50]. The literature shows inconsistency regarding concept definitions, but social support from family, friends and social groups in general appears to facilitate a more rapid recovery from back pain [50,51]. The evidence, however, is not consistent and the associations between social factors, emotional responses, and clinical aspects of the pain condition are complex. For example, in a prospective study among primary care patients with acute or subacute back pain, solicitous partner responses (e.g. getting the person in pain to rest, taking over their jobs and duties) were associated with a favourable course of back pain [52], whereas in another, cross-sectional study solicitous partner responses were associated with higher levels of disability, although only in those with low levels of depression. Campbell et al. [53] suggest that solicitous behaviours from partners may be a barrier to recovery by increasing disability but secondly facilitate recovery by reducing depression. Further research is clearly needed to understand the longitudinal relationships between social influences, quality of relationships, emotional well-being and pain or disability. The influence of social factors may vary across the life course [13]. For example, associations have been found between back pain reported by children and lifetime back pain prevalence in their parents. Genetic factors, shared environmental factors or social learning of illness behaviour during childhood may explain these associations [54]. The increasing interest in psychological and social factors should not distract from the potential importance of clinical or pathological aspects of back pain, for example in defining appropriate

704

D.A. van der Windt, K.M. Dunn / Best Practice & Research Clinical Rheumatology 27 (2013) 699–708

indications for surgery [55]. Understanding how biological, psychological and social dimensions interact to determine onset and progression of back pain is essential to generate better options for the management of back pain. The need for the development and validation of methods for a better clinical and multidimensional assessment in back pain patients has been highlighted by several authors in this issue of Best Practice and Research Clinical Rheumatology [29,41,50,55]. This may vary from a brief screening tool to be used for prognostic stratification in primary care to more extensive and detailed assessment in physiotherapy practice, secondary care services or pain rehabilitation clinics.

Moving towards stratified care for back pain The strong move towards developing individualised or stratified models of care, initiated in cancer research [e.g. Refs. [56,57]], has also gained a strong foothold in back pain research. Given the heterogeneous presentation, difficulties in terms of diagnostic classification, highly variable prognosis, and large variability in treatment response, back pain seems highly suitable for stratified care, and this has become a clear research priority [58]. The overall aim of stratified care is to optimise treatment response, increase efficiency of healthcare and reduce unnecessary harm [59]. Foster et al. [60] highlight different approaches to stratified care: (1) targeting treatment based on prognostic stratification where patients at high risk of poor outcome are referred for more extensive treatment while those at low risk can be reassured and offered minimal treatment; (2) targeting treatment based on specific patient or disease characteristics, e.g. offer exposure or graded activity programmes to patients with high levels of fear avoidance; (3) targeting specific treatment to those most likely to respond well to treatment or to those less likely to experience harm, e.g. offering surgical interventions based on clear evidence based indications [55] or avoiding chronic opioid therapy in patients at increased risk of adverse events or opioid dependency [61]. A programme of research is needed to support the development and testing of stratified care, which may include the following phases [59,62]: 1. Identify potential prognostic factors (to support prognostic stratification) or treatment moderators (to target specific interventions – based on clear hypotheses regarding effect modification), and validate these in different settings and populations; 2. Develop or identify matched treatments (may also be phase 1); 3. Test effect modification to determine if subgroups at high risk or scoring high on the treatment moderator indeed show larger effects (or less harm) from the matched treatment compared with subgroups at low risk or scoring low on the moderator; 4. Investigate the impact of stratified care (stratification & matched treatment) on patient outcomes and costs, compared to usual or current best care; 5. Evaluate the effects of implementing stratified care in routine clinical practice on decision making, patient outcomes and costs. The research requires a clear plan regarding the type of stratification (prognostic or based on specific patient or disease characteristics), where needed supported by theory or empirical evidence regarding the mechanisms underlying proposed treatment effects. Prospective high quality cohort studies are needed to derive and validate prognostic factors or multi-dimensional prognostic models, while randomised clinical trials of sufficient size are needed to test effect modification, and to investigate the clinical and cost-effectiveness of the stratified care approach (impact analysis study). Implementation studies and/or health economic modelling will allow an assessment of the (long-term) effects and costs when stratified care is more broadly adopted in clinical practice. Stratified care has great promise and potential, but is not a panacea. There are obstacles related to investigating or implementing stratified care, including the variable predictive performance of prognostic tools across populations and settings, which may lead to reduced benefits of stratified approaches. Furthermore, targeting only those who respond well to a specific treatment may only help a minority of patients, and could mean that large numbers of patients have to be screened to identify

D.A. van der Windt, K.M. Dunn / Best Practice & Research Clinical Rheumatology 27 (2013) 699–708

705

eligible cases. Implementing and delivering a more complex model of stratified care may require extended skills and additional healthcare resources. New developments There are promising avenues which may address the wider population of back pain and also have potential to improve patient outcomes. Following a successful multimedia campaign in Australia, which resulted in improvements in population beliefs regarding back pain and sustained effects on physician’s beliefs and back pain management [63], similar public health campaigns have been developed and tested in other countries although not all with the same effect [64–67]. Implementing simple evidence-based messages to support self-management while making use of the Internet or smartphone technology may also lead to shifts in population levels of pain and disability [68–71]. Although randomised controlled trials of Internet-based interventions have shown mixed effects [50,72] such interventions may be well-suited to address prognostic factors that are relevant across musculoskeletal pain problems or support health and wellbeing in general (e.g. increasing physical activity, weight control, problem solving skills) and thereby build people’s resources to manage painful conditions [29]. Summary We have described key areas of development in research on the epidemiology, prognosis and management of low-back pain, which are likely to have clear implications by providing evidence to support treatment decisions and inform patients about the likely course of their condition. These developments include an increasing awareness of the impact of low-back pain on public health, a redefinition of back pain as a long-term condition; an increasing understanding of the biological, psychological and social dimensions of back pain, and the potential for developing and testing stratified care for back pain. The box below summarises the opportunities for research identified across the 10 chapters of this Low Back Pain issue, which are likely to support further improvements in the care for patients with back pain.

Opportunities for research and further development - Investigate the impact of back pain in developed as well as developing countries and across the lifespan (children, working age adults, and older people), using consistent and comprehensive measures to define personal and social impact - Redefine and investigate back pain in terms of pain trajectory rather episode duration - Investigate public health or population-based interventions which may address or mitigate the impact of back pain on public health, and potentially make use of novel information technologies - Investigate the influence of prognostic factors across the lifespan, and investigate how pain episodes and prognostic factors accumulate over the life course to determine the risk of developing chronic disabling pain - Develop, further test or implement evidence-based and practically useful methods for the biopsychosocial assessment of back and other musculoskeletal pain (from brief screens that are applicable in primary care to more extensive assessments to be used in referred populations) - Investigate the mechanisms or processes explaining pathways to outcome in back pain, including the interactions between physiological or pathological, psychological and social factors

706

D.A. van der Windt, K.M. Dunn / Best Practice & Research Clinical Rheumatology 27 (2013) 699–708

- Investigate the clinical and cost-effectiveness of theory-informed or mechanism-focussed clinical and workplace interventions - Incorporate mediation analysis in intervention studies to investigate why and how interventions work in order to optimise design and delivery of interventions - Investigate the full spectrum of health risks relative to benefits of interventions, including drug treatments and surgical interventions, and establish clear indications for these treatments - In the design and evaluation of stratified care, make clear decisions regarding the type of stratification and plan the different phases of research

References *[1] Hoy D, Bain C, Williams G, March L, Brooks P, Blyth F, et al. A systematic review of the global prevalence of low back pain. Arthritis Rheum 2012;64:2028–37. [2] Volinn E. The epidemiology of low back pain in the rest of the world: a review of surveys in low- and middle-income countries. Spine 1997;22:1747–54. [3] Fejer R, Leboeuf-Yde C. Does back and neck pain become more common as you get older? A systematic literature review. Chiropr Man Ther 2012;20:24. [4] Dionne CE, Dunn KM, Croft PR. Does back pain prevalence really decrease with increasing age? A systematic review. Age Ageing 2006;35:229–34. [5] Thomas E, Mottram S, Peat G, Wilkie R, Croft P. The effect of age on the onset of pain interference in a general population of older adults: prospective findings from the North Staffordshire Osteoarthritis Project (NorStOP). Pain 2007;129:21–7. [6] Macfarlane GJ, Beasley M, Jones EA, Prescott GJ, Docking R, Keeley P, et al., MUSICIAN Study Team. The prevalence and management of low back pain across adulthood: results from a population-based cross-sectional study (the MUSICIAN study). Pain 2012;153:27–32. [7] Scheele J, Enthoven WT, Bierma-Zeinstra SM, Peul WC, van Tulder MW, Bohnen AM, et al. Characteristics of older patients with back pain in general practice: BACE cohort study. Eur J Pain 2013 Jul 19. http://dx.doi.org/10.1002/j.1532-2149.2013. 00363.x. *[8] Vos T, Flaxman AD, Naghavi M, Lozano R, Michaud C, Ezzati M, et al. Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380: 2163–96. [9] Buchbinder R, Brooks P, Woolf A, Hoy D. Placing the global burden of low back pain in context. Best Pract Res Clin Rheumatol. [10] Buchbinder R, Batterham R, Elsworth G, Dionne CE, Irvin E, Osborne RH. A validity-driven approach to the understanding of the personal and societal burden of low back pain: development of a conceptual and measurement model. Arthritis Res Ther 2011;13:R152. [11] Von Korff M, Dunn KM. Chronic pain reconsidered. Pain 2008;138:267–76. [12] Pransky G, Buchbinder R, Hayden J. Contemporary low back pain research – and implications for practice. Best Pract Res Clin Rheumatol 2010;24:291–8. [13] Dunn KM, Hestbaek L, Cassidy JD. Low back pain across the life course. Best Pract Res Clin Rheumatol, [Chapter 2]. [14] Lemeunier N, Leboeuf-Yde C, Gagey O. The natural course of low back pain: a systematic critical literature review. Chiropr Man Ther 2012;20:33. [15] Axén I, Leboeuf-Yde C. Trajectories of low back pain. Best Pract Res Clin Rheumatol, [Chapter 3]. [16] Macedo LG, Maher CG, Latimer J, McAuley JH, Hodges PW, Rogers WT. Nature and determinants of the course of chronic low back pain over a 12-month period: a cluster analysis. Phys Ther 2013 Sep 26 [Epub ahead of print], PMID: 24072729. [17] Dunn KM, Jordan K, Croft PR. Characterizing the course of low back pain: a latent class analysis. Am J Epidemiol 2006;163: 754–61. [18] Tamcan O, Mannion AF, Eisenring C, Horisberger B, Elfering A, Muller U. The course of chronic and recurrent low back pain in the general population. Pain 2010;150:451–7. [19] Leboeuf-Yde C, Jensen RK, Axen I. Absence of low back pain in patients followed weekly over one year with automated text messages. Chiropr Man Ther 2012;20:9. *[20] Calvo-Muñoz I, Gómez-Conesa A, Sánchez-Meca J. Prevalence of low back pain in children and adolescents: a metaanalysis. BMC Pediatr 2013;13:14. [21] Brage S, Sandanger I, Nygård JF. Emotional distress as a predictor for low back disability: a prospective 12-year population-based study. Spine 2007;32:269–74. [22] McEwen BS, Stellar E. Stress and the individual: mechanisms leading to disease. Arch Intern Med 1993;153:2093–101. *[23] Dunn KM. Extending conceptual frameworks: life course epidemiology for the study of back pain. BMC Musculoskelet Disord 2010;11:23. [24] Von Korff M, Alonso J, Ormel J, Angermeyer M, Bruffaerts R, Fleiz C, et al. Childhood psychosocial stressors and adult onset arthritis: broad spectrum risk factors and allostatic load. Pain 2009;143:76–83. *[25] Dominick CH, Blyth FM, Nicholas MK. Unpacking the burden: understanding the relationships between chronic pain and comorbidity in the general population. Pain 2012;153:293–304. [26] Seery MD, Leo RJ, Holman EA, Silver RC. Lifetime exposure to adversity predicts functional impairment and healthcare utilization among individuals with chronic back pain. Pain 2010;150:507–15.

D.A. van der Windt, K.M. Dunn / Best Practice & Research Clinical Rheumatology 27 (2013) 699–708

707

[27] Ramírez-Maestre C, Esteve R, López AE. The path to capacity: resilience and spinal chronic pain. Spine 2012;37:E251–8. [28] Viniol A, Jegan N, Hirsch O, Leonhardt C, Brugger M, Strauch K, et al. Chronic low back pain patient groups in primary care – a cross sectional cluster analysis. BMC Musculoskelet Disord 2013;14:294 [Epub ahead of print]. [29] Hartvigsen J, Natvig B, Ferreita M. Is it all about pain in the back? Best Pract Res Clin Rheumatol, [Chapter 4]. [30] Kamaleri Y, Natvig B, Ihlebaek CM, Benth JS, Bruusgaard D. Number of pain sites is associated with demographic, lifestyle, and health-related factors in the general population. Eur J Pain 2008;12:742–8. [31] Von Korff M, Crane P, Lane M, Miglioretti DL, Simon G, Saunders K, et al. Chronic spinal pain and physical-mental comorbidity in the United States: results from the national comorbidity survey replication. Pain 2005;113:331–9. [32] Van der Windt DA, Dunn LM, Spies MN, Mallen CD, Blankenstein AH, Stalman WAB. Impact of physical symptoms on perceived health in the community. Psychosom Res 2008;64:265–74. [33] Henschke N, Ostelo RW, Terwee CB, van der Windt DA. Identifying generic predictors of outcome in patients presenting to primary care with non-spinal musculoskeletal pain. Arthritis Care Res 2012;64:1217–24. [34] Mallen CD, Peat G, Thomas E, Dunn KM, Croft PR. Prognostic factors for musculoskeletal pain in primary care: a systematic review. Br J Gen Pract 2007;57:655–61. *[35] Von Korff M, Miglioretti DL. A prognostic approach to defining chronic pain. Pain 2005;117:304–13. [36] Thomas E, Dunn KM, Mallen CD, Peat GM. A prognostic approach to chronic pain: application to knee pain in older adults. Pain 2008;139:389–97. [37] Muller S, Thomas E, Dunn KM, Mallen CD. A prognostic approach to defining chronic pain across a range of musculoskeletal pain sites. Clin J Pain 2013;29:411–6. [38] Keller A, Hayden J, Bombardier C, van Tulder M. Effect sizes of non-surgical treatments of non-specific low-back pain. Eur Spine J 2007;16:1776–88. [39] Machado LA, Kamper SJ, Herbert RD, Maher CG, McAuley JH. Analgesic effects of treatments for non-specific low back pain: a meta-analysis of placebo-controlled randomized trials. Rheumatology (Oxford) 2009;48:520–7. [40] Henschke N, Ostelo RW, van Tulder MW, Vlaeyen JW, Morley S, Assendelft WJ, et al. Behavioural treatment for chronic low-back pain. Cochrane Database Syst Rev 2010;7:CD002014. [41] Pincus T, McCracken LM. Psychological factors and treatment opportunities in low back pain. Best Pract Res Clin Rheumatol, [Chapter 5]. [42] Mansell G, Kamper SJ, Kent P, Why and how back pain interventions work: what can we do to find out? Best Pract Res Clin Rheumatol. [43] Kraemer HC, Wilson GT, Fairburn CG, Agras WS. Mediators and moderators of treatment effects in randomized clinical trials. Arch Gen Psychiatry 2002;59:877–83. [44] Kazdin AE. Mediators and mechanisms of change in psychotherapy research. Ann Rev Clin Psychol 2007;3:1–27. *[45] Emsley R, Dunn G, White IR. Mediation and moderation of treatment effects in randomised controlled trials of complex interventions. Stat Methods Med Res 2010;19:237–70. [46] Shaw WS, van der Windt DA, Main CJ, Loisel P, Linton SJ“Decade of the Flags” Working Group. Early patient screening and intervention to address individual-level occupational factors (“blue flags”) in back disability. J Occup Rehabil 2009; 19:64–80. [47] Campbell P, Wynne-Jones G, Muller S, Dunn KM. The influence of employment social support for risk and prognosis in nonspecific back pain: a systematic review and critical synthesis. Int Arch Occup Environ Health 2013;86:119–37. [48] Soklaridis S, Ammendolia C, Cassidy D. Looking upstream to understand low back pain and return to work: psychosocial factors as the product of system issues. Soc Sci Med 2010;71:1557–66. [49] Steenstra IA, Verbeek JH, Heymans MW, Bongers PM. Prognostic factors for duration of sick leave in patients sick listed with acute low back pain: a systematic review of the literature. Occup Environ Med 2005;62:851–60. [50] Shaw WS, Campbell P, Nelson CJ, Main CJ, Linton SJ. Best Pract Res Clin Rheumatol, [Chapter 6]. *[51] Campbell P, Wynne-Jones G, Dunn KM. The influence of informal social support on risk and prognosis in spinal pain: a systematic review. Eur J Pain 2011;15:444.e1–444.e14. [52] Jellema P, Van der Windt DAWM, Van der Horst HE, Stalman WAB, Bouter LM. Predictors of an unfavourable course of low back pain in general practice. Comparison of four instruments. Br J Gen Pract 2007;57:15–22. [53] Campbell P, Jordan KP, Dunn KM. The role of relationship quality and perceived partner responses with pain and disability in those with back pain. Pain Med 2012;13:204–14. [54] Shraim M, Mallen CD, Dunn KM. GP consultations for medically unexplained physical symptoms in parents and their children: a systematic review. Br J Gen Pract 2013;63:e318–25. [55] Jacobs WCH, Rubinstein SM, Koes BW, Van Tulder MW, Peul WC. Evidence for surgery in degenerative lumbar spine disorders. Best Pract Res Clin Rheumatol, [Chapter 9]. [56] Williams C, Brunskill S, Altman D, Briggs A, Campbell H, Clarke M, et al. Cost-effectiveness of using prognostic information to select women with breast cancer for adjuvant systemic therapy. Health Technol Assess 2006;10:1–204. iii–iv, ix–xi. [57] Mok TS, Wu YL, Thongprasert S, Yang CH, Chu DT, Saijo N, et al. Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma. N Engl J Med 2009;361:947–57. [58] Costa Lda C, Koes BW, Pransky G, Borkan J, Maher CG, Smeets RJ. Primary care research priorities in low back pain: an update. Spine 2013;38:148–56. *[59] Hingorani AD, van der Windt DA, Riley RD, Abrams K, Moons KG, Steyerberg EW, et alPROGRESS Group. Prognosis research strategy (PROGRESS) 4: stratified medicine research. BMJ 2013;346:e5793. [60] Foster NE, Hill JC, O’Sullivan P, Hancock P. Stratified models of care. Best Pract Res Clin Rheumatol, [Chapter 7]. [61] Von Korff M. Long-term use of opioids for complex chronic pain. Best Pract Res Clin Rheumatol, [Chapter 8]. [62] Reilly BM, Evans AT. Translating clinical research into clinical practice: impact of using prediction rules to make decisions. Ann Intern Med 2006;144:201–9. [63] Buchbinder R, Jolley D, Wyatt M. 2001 Volvo Award Winner in clinical studies: effects of a media campaign on back pain beliefs and its potential influence on management of low back pain in general practice. Spine 2001; 26:2535–42.

708

D.A. van der Windt, K.M. Dunn / Best Practice & Research Clinical Rheumatology 27 (2013) 699–708

*[64] Gross DP, Deshpande S, Werner EL, Reneman MF, Miciak MA, Buchbinder R. Fostering change in back pain beliefs and behaviors: when public education is not enough. Spine J 2012;12:979–88. [65] Waddell G, O’Connor M, Boorman S, Torsney B. Working Backs Scotland: a public and professional health education campaign for back pain. Spine 2007;32:2139–43. [66] Gross DP, Russell AS, Ferrari R, Battie MC, Schopflocher D, Hu R, et al. Evaluation of a Canadian back pain mass media campaign. Spine 2010;35:906–13. [67] Werner EL, Ihlebaek C, Laerum E, Wormgoor ME, Indahl A. Low back pain media campaign: no effect on sickness behaviour. Patient Educ Couns 2008;71:198–203. [68] Macea D, Gajos K, Daglia Calil YA, Fregni F. The efficacy of web-based cognitive behavioral interventions for chronic pain: a systematic review and meta-analysis. J Pain 2010;11:917–29. [69] Kristjánsdóttir OB, Fors EA, Eide E, Finset A, Stensrud TL, van Dulmen S, et al. A smartphone-based intervention with diaries and therapist-feedback to reduce catastrophizing and increase functioning in women with chronic widespread pain: randomized controlled trial. J Med Internet Res 2013;15:e5. [70] Krein SL, Kadri R, Hughes M, Kerr EA, Piette JD, Holleman R, et al. Pedometer-based internet-mediated intervention for adults with chronic low back pain: randomized controlled trial. J Med Internet Res 2013;15:e181. [71] Del Pozo-Cruz B, Adsuar JC, Parraca J, Del Pozo-Cruz J, Moreno A, Gusi N. A web-based intervention to improve and prevent low back pain among office workers: a randomized controlled trial. J Orthop Sports Phys Ther 2012;42:831–41. [72] Bender JL, Radhakrishnan A, Diorio C, Englesakis M, Jadad AR. Can pain be managed through the Internet? A systematic review of randomized controlled trials. Pain 2011;152:1740–50.