Incidence and prognosis of mid-back pain in the general population: A systematic review

Transcription

1 SYSTEMATIC REVIEW Incidence and prognosis of mid-back pain in the general population: A systematic review M.S. Johansson 1, M. Jensen Stochkendahl 2, J. Hartvigsen 1,2, E. Boyle 1,3, J.D. Cassidy 1,3,4 1 Department of Sports Science and Clinical Biomechanics, Faculty of Health, University of Southern Denmark, Odense, Denmark 2 Nordic Institute of Chiropractic and Clinical Biomechanics, Odense, Denmark 3 Division of Epidemiology, Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada 4 Division of Health Care and Outcomes Research, Krembil Research Institute, University Health Network, Toronto, ON, Canada Correspondence Melker S. Johansson melker.johansson@gmail.com Funding sources This systematic review was funded by the Danish Chiropractors Research Foundation. The funder was not involved in the development of the protocol, management of the review, analysis or interpretation of data, preparation of the report or decision to submit the review for publication. Conflicts of interest All authors declare that they do not have any conflicts of interest. Systematic review registration number CRD (PROSPERO). Accepted for publication 15 March 2016 doi: /ejp.884 Abstract Background and objective: Despite being common early in life and affecting individuals quality of life to the same degree as neck and low back pain, research into epidemiological aspects of mid-back pain (MBP) has been scarce. The purpose of our systematic review was therefore to describe the incidence and prognosis of MBP in the general population. The PRISMA Statement guided the study process. Databases: A systematic search was conducted in CINAHL, PEDro, PsycINFO and Scopus. Results: Of 3194 unique records identified, seven were included in our qualitative synthesis. The 3-month and 2-year incidence proportions of MBP in children and adolescents were approximately 4% and 50%, respectively. In adults, the 1-month incidence proportion was less than 1%. The persistence or recurrence of MBP over a 1- to 4-year period was between 13% and 45% in children and adolescents; a change in spinal pain location over time was common. Individuals reporting MBP have an increased risk of future care seeking compared with people without musculoskeletal complaints. No studies assessing adult MBP recovery trajectories or prognostic factors were identified. Conclusions: Knowledge about the incidence and prognosis of MBP in the general population is limited. The incidence of MBP in children and adolescents seems to be similar to the incidence of neck and low back pain; in adults, it is lower than that of neck and low back pain. Studies investigating recovery trajectories of MBP in adults and prognostic factors for MBP are lacking. What does this study add?: The incidence of mid-back pain (MBP) in young individuals is similar to that of neck and low back pain, and 50% report persistent pain; however, the evidence base is limited. Knowledge about adult trajectories and prognostic factors for MBP is lacking. 1. Introduction Mid-back pain (MBP) can be defined as pain experienced in the body region between the 1st and 12th thoracic vertebrae and the corresponding posterior aspect of the trunk (Leboeuf-Yde et al., 2009), and it can occasionally cause radiating pain into the anterior chest wall (Kellgren, 1939). Pain in the mid back is relatively common early in life, with a 1-month prevalence of 13% and 35% in children and adolescents, respectively, compared 2016 European Pain Federation - EFIC â Eur J Pain (2016) 1

2 with approximately 5% and 15% for neck pain, and 4% and 36% for low back pain (Wedderkopp et al., 2001; Kjaer et al., 2011). In adults, however, MBP is less common than both neck and low back pain, with an estimated 1-year prevalence of about 15% compared with approximately 32% and 43% for neck and low back pain, respectively (Niemelainen et al., 2006; Leboeuf-Yde et al., 2009). Epidemiological and clinical aspects of MBP have been investigated to a lesser extent than neck and low back pain. In 2009, Briggs et al. reviewed the literature related to the prevalence, incidence and associated factors of thoracic spine pain in the general population (Briggs et al., 2009). They found that being female, having co-occurring musculoskeletal pain and self-reported headache during the previous month were associated with MBP. However, their results pertained mainly to the prevalence and associated risk factors, since only two studies that reported risk factors, five incidence studies, and no prognostic studies were identified. Consequently, now more than 6 years later, clinicians are still challenged when seeking information about the prognosis of MBP. Thus, we believe that a systematic review summarizing the current literature to assist clinicians informing their patients is needed. The purpose of our systematic review was to describe the incidence and prognosis of MBP in terms of trajectories, factors predicting prognosis and health-related outcomes in the general population. 2. Methods 2.1 Protocol and registration The protocol was registered in the International Prospective Register of Systematic Reviews (PROS- PERO) (CRD ). The Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) Statement (Moher et al., 2009) guided the development of the study protocol and the conduct and reporting of this systematic review. 2.2 Eligibility criteria Inclusion criteria We considered population-based longitudinal cohort studies, randomized controlled trials and case control studies with a self-reported measure of MBP with or without radiating chest pain for inclusion. Incidence studies where the first onset of or a new episode of MBP was described as incidence rates or proportions were included. The following outcomes were considered relevant for prognostic studies (Deyo et al., 1998; Bombardier et al., 2001): trajectories (e.g. persistence of pain and change in pain location), recovery (e.g. self-reported and globally perceived recovery), health (e.g. general health, physical and/or mental health, and health-related quality of life), pain (e.g. change in pain intensity), back-related function (e.g. back disability, limitations of activities of daily living and level of physical activity) and social disability (e.g. work disability, return-to-work, disability pension and care seeking). Trajectories should be described in terms of proportions with persistent or recurrent pain or change in pain location, intensity or pain-related disability. The remaining outcomes should be described in terms of occurrence of event or time to event, reported as proportions, odds ratios, relative risks or hazard rate ratios. Published peer-reviewed research articles written in English, Swedish, Danish or Norwegian were considered eligible Exclusion criteria We did not consider systematic reviews, cross-sectional studies and case reports or case series for inclusion. We excluded studies of working populations and studies including participants with MBP of specific pathological origin, such as inflammatory disorders (i.e. spondyloarthropathies), structural changes (e.g. fractures, osteoporosis, scoliosis, Scheuermann s disease, hyperkyphosis, Diffuse Idiopathic Skeletal Hyperostosis, intervertebral disc herniation and others), infections (i.e. spondylodiscitis), cancer, cardiovascular disease or any non-musculoskeletal thoracic pain including referred pain from viscera. 2.3 Information sources and literature search We applied the search strategy, developed with assistance from a research librarian, to the following databases: CINAHL, PEDro, PsycINFO and Scopus, and we included studies published at any time before January 12, 2015 (i.e. no limitations for year of publication; see Methods S1 for search strategy, including search terms). In addition, reference lists of included records were screened to identify relevant studies not captured by our search. 2 Eur J Pain (2016) 2016 European Pain Federation - EFIC â

3 2.4 Study selection process Studies were selected through three steps: screening of titles (1) and abstracts (2), followed by a full-text analysis (3) of the retrieved records. After each step in the selection process, the reviewers (MSJ and MJS) met to reach consensus regarding eligibility. The inter-rater agreement (j statistics) for study selection was assessed (Kirkwood and Sterne, 2003). 2.5 Data extraction process and risk of bias assessment The reviewers (MSJ and MJS) independently extracted a range of data items (Results S1 and S2) from the included studies using a standardized form and assessed each study for the risk of bias, using the Quality In Prognosis Studies (QUIPS) tool. The QUIPS tool focuses on the risk of bias in six methodological domains (see Table 1 and Methods S2 for details) and has shown acceptable inter-rater reliability (Hayden et al., 2013). Consensus throughout the study selection, data extraction and risk of bias assessment was reached through meetings between the reviewers. In case of disagreement, a third reviewer s (JDC) opinion was sought. 2.6 Synthesis of results and analysis The results from the data extraction and the risk of bias assessment were summarized in tables. With regard to the outcomes, incidence was summarized as incidence proportions (i.e. proportions of participants developing Table 1 Risk of bias assessment of studies investigating the incidence and or prognosis of mid-back pain in the general population. Domains assessed for risk of bias using the Quality In Prognosis Studies (QUIPS) tool First author, year, country Study participation Study attrition Prognostic factor measurement Outcome measurement Study confounding Statistical analysis and reporting Comments Mikkelsson M., 1997, Finland El-Metwally A., 2004, Finland Hagen K., 2006, Norway El-Metwally A., 2007, Finland Kjaer P., 2011, Denmark Kristman V., 2012, Denmark Aartun E., 2014, Denmark Hartvigsen J., 2014, Denmark Low Moderate N/A Low N/A High Reasons for and potential impact of subjects lost to follow-up are unclear. Inadequate description of analytical strategy. Inconsistent reporting of numbers of MBP cases in text and tables Low Moderate N/A Low N/A Moderate Moderate participation rate. It is not clear how proportions of persistent/ recurrent pain were calculated. Low Low N/A Low N/A Low Low Low N/A Low N/A Low Moderate Moderate N/A Low N/A Low Exclusion criteria are unclear. Reasons for non-participation not described. Moderate study participation rate Low Low N/A Low N/A Low Low Moderate N/A Moderate N/A Low Cases lost to follow-up were significantly different from analysed cases with regards to sex and previous pain status. Risk of recall bias related to the incidence measurements Low Low N/A Low N/A Low N/A, not applicable; MBP, mid-back pain. Green, orange, and red are illustrating a low, moderate and high risk of bias, respectively European Pain Federation - EFIC â Eur J Pain (2016) 3

4 MBP); trajectories as proportions of persistent or recurrent pain and proportions of change in pain sites; and for other binary outcomes, odds ratios (OR), relative risks (RR) or hazard rate ratios (HRR) were used. Studies considered to have a high risk of bias were rejected. 3. Results 3.1 Search results and study selection Of the 3194 unique records identified, 169 records were assessed for eligibility in full-text. We included eight of these in the review (Mikkelsson et al., 1997; El-Metwally et al., 2004, 2007; Hagen et al., 2006; Kjaer et al., 2011; Kristman et al., 2012; Aartun et al., 2014; Hartvigsen et al., 2014) (Fig. 1). Cohen s j for the title and abstract screening and full-text analysis were 0.75, 0.77 and 0.94, respectively. 3.2 Risk of bias assessment Results of the risk of bias assessments are presented in Table 1. In general, the studies had a low risk of bias; however, four studies had a moderate risk of bias across different domains of the QUIPS tool (Mikkelsson et al., 1997; El-Metwally et al., 2004; Kjaer Figure 1 Flow chart illustrating the study processes identification, screening, eligibility and inclusion of studies investigating the incidence and/or prognosis of mid-back pain in the general population. MBP is mid-back pain. 4 Eur J Pain (2016) 2016 European Pain Federation - EFIC â

5 et al., 2011; Aartun et al., 2014). The study by Mikkelsson et al. (1997) had a high risk of bias related to the statistical analysis and reporting of results and was rejected from the qualitative synthesis of our results. No studies investigating prognostic factors for MBP or a causal relationship between a predictor and outcome were identified; hence, the risk of prognostic factor measurement bias and the risk of study confounding were not assessed. 3.3 Study characteristics Characteristics of the included studies are presented in Results S1 and S2. These studies were published between 2004 and 2014 and were undertaken in three Nordic countries. Cohort sizes varied from 564 (El-Metwally et al., 2004) to 25,922 participants (Hagen et al., 2006), and follow-up periods varied from about 1 month to 20 years. Participation rates ranged from 62% to 96% and follow-up rates ranged from 72% to 100%. The incidence of MBP was investigated in three studies (Hagen et al., 2006; El-Metwally et al., 2007; Aartun et al., 2014). In five studies, the prognosis of MBP was described in terms of persistence or recurrence over time (El-Metwally et al., 2004; Kjaer et al., 2011), change in pain sites (Aartun et al., 2014), health-related benefits (Kristman et al., 2012) and future care seeking (Hartvigsen et al., 2014). The terminology used to describe the body region as well as the construct of MBP and outcome measurement varied across the studies. Four studies focused on children and or adolescents in primary school settings (El-Metwally et al., 2004, 2007; Kjaer et al., 2011; Aartun et al., 2014); two of these used different subcohorts of a Finnish primary school children cohort (El-Metwally et al., 2004, 2007). Finally, three studies included adults (Hagen et al., 2006; Kristman et al., 2012; Hartvigsen et al., 2014). 3.4 Results of individual studies Incidence of MBP In Finnish school children aged 9 13, the 3-month incidence proportion of MBP was found to be % (El-Metwally et al., 2007). In Danish adolescents aged 11 at baseline, the 2-year incidence proportion was found to be 49.8% [95% confidence interval (CI): ] (Aartun et al., 2014). In adults, the 1-month incidence proportion was 0.7% and 0.4% in Norwegian females and males, respectively (Hagen et al., 2006) Prognosis: MBP trajectories In Finnish school children aged 9 13, MBP was found to persist or reoccur in 30 35% of children both 1 and 4 years later (El-Metwally et al., 2004). In Danish primary school children aged 9 at baseline, Kjaer et al. (2011) found a fluctuating pattern of MBP over time, with a 1-month period prevalence of 20%, 13% and 28% at age 9, 12 and 15, respectively. Eighteen per cent of 9-year-old children with MBP reported persistent pain in the mid back at age 12, while 47% of 12-year-old adolescents with MBP reported persistent pain in the same region at age 15 (Kjaer et al., 2011). Aartun et al. (2014) observed that a change in spinal pain location or pain in additional spinal regions was common over a 2-year period in Danish adolescents; of those with isolated MBP at baseline, 29.6% reported a combination of MBP and co-occurring neck pain, 20.4% reported pain in all three spinal regions and 14.8% reported isolated neck pain 2 years later, while 13% reported persistent isolated pain in the mid back at follow-up. No spinal pain was reported by 7.4% (Aartun et al., 2014) Prognosis: health-related benefits and future care seeking The incidence rate of sickness absence (i.e. all causes, work and non-work related) was 51.6 (95% CI: ) and 95.5 (95% CI: ) per 1000 person-years for individuals reporting non-radiating MBP lasting 1 30 days and >30 days (i.e. short and long term), respectively. The corresponding incidence rates for radiating MBP were 55.3 (95% CI: ) and 63.4 (95% CI: ) (Kristman et al., 2012). In general, these incidence rates were not different from those found in individuals with other or without spinal pain. However, both shortand long-term radiating MBP had a statistically significant lower incidence rate of sickness absence compared with radiating pain in other spinal sites. MBP was not associated with the occurrence of, or with the time-to-first sickness benefit period, regardless of pain duration and radiating pain status in adult Danish twins. However, long-term non-radiating MBP was associated with a decreased number of sickness benefit periods, relative to no spinal pain (Kristman et al., 2012). Compared with those without musculoskeletal pain, persons with MBP as their primary pain complaint had a statistically significant increased risk of consulting a general practitioner (RR: 1.54, 95% CI: ), visiting an outpatient hospital 2016 European Pain Federation - EFIC â Eur J Pain (2016) 5

6 department (RR: 1.54, 95% CI: ) and being admitted to a hospital (RR: 1.76, 95% CI: ) (Hartvigsen et al., 2014). This was generally found for all primary musculoskeletal pain sites. People with a primary complaint of pain in the mid back did not however, have an increased risk of consulting a physical therapist or a chiropractor. Finally, compared with individuals with a primary complaint of low back pain, persons reporting MBP as their primary complaint were found to have an elevated risk of outpatient hospital consultations (RR: 1.54, 95% CI: ) (Hartvigsen et al., 2014). 4. Discussion We have systematically searched and reviewed the literature related to the incidence and prognosis of MBP in the general population. Overall, the results of this systematic review reflect the limited knowledge about these issues and are based on relatively few studies using different definitions of MBP as well as different outcome measures. However, the studies had generally a low risk of bias. In summary, the 3-month and 2-year incidence proportions of MBP in children and adolescents were approximately 4% and 50%, respectively. In adults, the 1-month incidence proportion was less than 1%. Due to the limited number of studies, these estimates should be cautiously interpreted. Over a 1- to 4-year period, persistence or recurrence of MBP was between 13% and 45% in children and adolescents and a change in spinal pain location appears to be common. We did not find any population-based studies assessing MBP trajectories in adults or studies investigating prognostic factors for MBP. With regard to other health-related outcomes, MBP is not associated with the occurrence of or with the time-to-first sickness benefit period. However, relative to those without a musculoskeletal complaint and similar to individuals with other musculoskeletal pain, individuals reporting MBP have an increased risk of future care seeking. The incidence proportions of MBP reported by El-Metwally et al. (2007) and Aartun et al. (2014) are not in concordance. This could be explained by their different definitions of MBP and outcome constructs (i.e. different recall periods for incidence measures). Similarly, in the systematic review by Briggs et al. (2009), a marked variability in the 1-year incidence data (i.e %) was found, likely due to differences in age groups and in pain definitions between the studies. We included two (Hagen et al., 2006; El-Metwally et al., 2007) of the five studies related to the incidence of MBP that Briggs et al. (2009) included, as well as one additional study (Aartun et al., 2014); the three studies excluded in our review were assessed not to study the general population. The incidence of MBP appears to be similar to the incidence of low back and neck pain in children and adolescents. However, in adults, the incidence of MBP is between three and seven times lower than that of neck and low back pain. For example, the pooled estimates of the incidence proportion of first-time low back pain and low back pain in pain-free individuals of the general population were 26% and 27%, respectively. However, it should be emphasized that the estimate related to the incidence of low back pain in pain-free individuals is based on heterogeneous studies (i.e. I 2 = 99%) (Taylor et al., 2014). In adolescents, pain in one spinal region (i.e. neck, mid back or low back) seems to spread to additional spinal regions over a relatively short period of time (Aartun et al., 2014). It is possible that this increase in number of pain sites occurs in mid-adolescence more often than in younger children. For example, a larger proportion of 13-year-olds without MBP reported pain in the mid back at age 15, compared to the proportion of 9-year-olds without MBP that reported pain in the same region at age 13 (Kjaer et al., 2011). We lack knowledge about the trajectory of MBP in adults, as well as knowledge about prognostic factors for MBP. Pain in the three spinal regions is similar in several aspects; individuals with spinal pain report similar pain duration as well as similar types and number of consequences (Leboeuf-Yde et al., 2011, 2012), they have an increased risk of future care seeking (Hartvigsen et al., 2014) and a similar pattern of co-occurring spinal pain (Hartvigsen et al., 2013), regardless of their age and where it hurts. This suggests that spinal pain could be a general disorder and that the anatomical region is less significant in terms of these aspects. In continuation of this, it is possible that the trajectory of MBP is similar to that of neck and low back pain. The trajectory of neck pain in both care-seeking and non-care-seeking populations has been described as poor (Walton et al., 2013) since 50 75% report neck pain 1 5 years after the onset (Carroll et al., 2009). The trajectory of low back pain in a primary care setting is often described as a substantial improvement in pain intensity within the first 3 months after onset, followed by a more slow improvement over time (Itz et al., 2013; Artus et al., 2014). However, this is based on 6 Eur J Pain (2016) 2016 European Pain Federation - EFIC â

7 population averages, and studies exploring different low back pain trajectories have identified distinct subgroups with highly varying profiles (Dunn et al., 2006; Tamcan et al., 2010; Kongsted et al., 2015). In a comprehensive systematic review focussing on the natural course of low back pain in the general population, the authors conclude that the low back pain status at baseline is predictive of future pain. First, individuals without low back pain at baseline often continue to be pain free over many years. Second, subgroups characterized by either intermittent or more persistent pain were noted among those with low back pain at baseline, indicating a more heterogeneous course among individuals with pain. Finally, improvement to a pain-free state was rarely reported as a common finding in the included studies (Lemeunier et al., 2012). Whether a unique MBP trajectory or subgroups similar to those found in low back pain populations exist and whether the prognosis of MBP is influenced by factors similar to other spinal pain conditions are yet not known. 4.1 Strengths and limitations We believe that our search strategy and thorough study selection process identified the relevant studies. Based on our a priori knowledge about the limited number of studies investigating the nature of MBP and a varying terminology being used, a broad search strategy was designed that resulted in exclusion of a large number of records. An excellent inter-rater agreement (Fleiss, 1981) was found for the study selection process; however, we are aware that this does not necessarily indicate a good accuracy. The methodological assessment tool used in our systematic review has been specifically developed to assess the risk of bias in prognostic cohort studies, and its construct is in line with the recommendations of the PRISMA Statement (Moher et al., 2009). The primary limitation of this systematic review is the sparse literature related to our objectives. Additionally, the studies identified have different definitions of MBP and used different outcome constructs. For example, one of the two studies investigating the incidence of MBP in children and adolescents used a 2-year recall period, which can have resulted in biased estimates. We decided not to calculate standardized incidence proportions because of study heterogeneity. However, the included studies had generally a low risk of bias and only one study was rejected due to a high risk of bias. 4.2 Clinical and research implications Many of the studies that we screened investigated pain in a range of body regions, such as the neck, shoulder and low back; studied back pain as a whole (i.e. not differentiating between the low and the mid back); or focused on only one spinal region; however, the mid back was often not included. This should change in future studies to increase our knowledge of MBP in the general and clinical populations. Spinal pain often manifests in more than one region (Kamaleri et al., 2008; Leboeuf-Yde et al., 2012; Hartvigsen et al., 2013). If MBP actually is similar to pain in the neck or the low back, we believe that an approach where neck, mid-back and low back pain is further investigated as a general disorder could help us to improve our understanding of the nature of spinal pain. However, before such an approach can be implemented with confidence, further well-conducted studies focussing on epidemiological and clinical aspects of MBP are needed. We believe our results can inform researchers designing future studies. A consensus process to reach a standard definition of MBP, similar to what has been made for low back pain (Dionne et al., 2008), and the use of shorter recall periods when measuring incidence would be important initial steps towards more homogenous and comparable studies. The investigation of the prognosis of MBP should ideally be conducted within previously recommended frameworks of prognostic research (Hemingway et al., 2013). By studying its course, exploring and investigating prognostic markers and factors, prognostic models can be developed and ultimately translated and implemented in clinical practice. 5. Conclusions Knowledge about the incidence and prognosis of MBP in the general population is limited. The incidence of MBP in children and adolescents seems to be similar to the incidence of neck and low back pain. In adults, the incidence of MBP is lower than that of neck and low back pain. Up to nearly half of children and adolescents with MBP experience persistent or recurrent pain. Individuals with MBP appear to have an increased risk of future care seeking compared to people without pain. Trajectories of MBP in adults and prognostic factors for MBP in general are unknown European Pain Federation - EFIC â Eur J Pain (2016) 7

8 Acknowledgements Acknowledgements are directed to research librarian Majbritt Ursula Johansen at the University Library of Southern Denmark, for the assistance in developing the search strategy used in our systematic review. Author contributions M.S.J., M.J.S., J.H., E.B. and J.D.C. contributed to the conception and design of the study. M.S.J. developed the protocol and M.J.S., J.H., E.B. and J.D.C. contributed with critical revising. M.S.J. developed the search strategy with assistance from a research librarian. The study selection process, conduct and summary of the data extraction and risk of bias assessment were done by M.S.J. and M.J.S. M.S.J. developed the manuscript. M.J.S., J.H., E.B. and J.D.C. contributed with critical revising during the development of the manuscript. All authors have discussed the results and have given approval to the publishing of the final version of the article. References Aartun, E., Hartvigsen, J., Wedderkopp, N., Hestbaek, L. (2014). Spinal pain in adolescents: Prevalence, incidence, and course: A schoolbased two-year prospective cohort study in 1,300 Danes aged BMC Musculoskelet Disord 15, 187. Artus, M., van der Windt, D., Jordan, K.P., Croft, P.R. (2014). The clinical course of low back pain: A meta-analysis comparing outcomes in randomised clinical trials (RCTs) and observational studies. BMC Musculoskelet Disord 15, 68. Bombardier, C., Hayden, J., Beaton, D.E. (2001). Minimal clinically important difference. Low back pain: Outcome measures. J Rheumatol 28, Briggs, A.M., Smith, A.J., Straker, L.M., Bragge, P. (2009). Thoracic spine pain in the general population: Prevalence, incidence and associated factors in children, adolescents and adults. A systematic review. BMC Musculoskelet Disord 10, 77. Carroll, L.J., Hogg-Johnson, S., van der Velde, G., Haldeman, S., Holm, L.W., Carragee, E.J., Hurwitz, E.L., Cote, P., Nordin, M., Peloso, P.M., Guzman, J., Cassidy, J.D. (2009). Course and prognostic factors for neck pain in the general population: Results of the Bone and Joint Decade Task Force on Neck Pain and Its Associated Disorders. J Manipulative Physiol Ther 32, S87 S96. Deyo, R.A., Battie, M., Beurskens, A.J., Bombardier, C., Croft, P., Koes, B., Malmivaara, A., Roland, M., Von Korff, M. and Waddell, G.(1998). Outcome measures for low back pain research. A proposal for standardized use. Spine (Phila Pa 1976) 23, Dionne, C.E., Dunn, K.M., Croft, P.R., Nachemson, A.L., Buchbinder, R., Walker, B.F., Wyatt, M., Cassidy, J.D., Rossignol, M., Leboeuf- Yde, C., Hartvigsen, J., Leino-Arjas, P., Latza, U., Reis, S., Gil Del Real, M.T., Kovacs, F.M., Oberg, B., Cedraschi, C., Bouter, L.M., Koes, B.W., Picavet, H.S., vantulder, M.W., Burton, K., Foster, N.E., Macfarlane, G.J., Thomas, E., Underwood, M., Waddell, G., Shekelle, P., Volinn, E. and Von Korff, M.(2008). A consensus approach toward the standardization of back pain definitions for use in prevalence studies. Spine (Phila Pa 1976) 33, Dunn, K.M., Jordan, K., Croft, P.R. (2006). Characterizing the course of low back pain: A latent class analysis. Am J Epidemiol 163, El-Metwally, A., Salminen, J.J., Auvinen, A., Kautiainen, H., Mikkelsson, M. (2004). Prognosis of non-specific musculoskeletal pain in preadolescents: a prospective 4-year follow-up study till adolescence. Pain 110, El-Metwally, A., Salminen, J.J., Auvinen, A., MacFarlane, G. and Mikkelsson, M. (2007). Risk factors for development of non-specific musculoskeletal pain in preteens and early adolescents: A prospective 1-year follow-up study. BMC Musculoskelet Disord 8, 46. Fleiss, J.L. (1981). Statistical methods for rates and proportions, 2nd edition. (New York: John Wiley & Sons). Hagen, K., Svebak, S., Zwart, J.A. (2006). Incidence of musculoskeletal complaints in a large adult Norwegian county population. The HUNT study. Spine 31, Hartvigsen, J., Davidsen, M., Hestbaek, L., Sogaard, K., Roos, E.M. (2013). Patterns of musculoskeletal pain in the population: A latent class analysis using a nationally representative interviewer-based survey of 4817 Danes. Eur J Pain 17, Hartvigsen, J., Davidsen, M., Sogaard, K., Roos, E.M., Hestbaek, L. (2014). Self-reported musculoskeletal pain predicts long-term increase in general health care use: A population-based cohort study with 20-year follow-up. Scand J Public Health 42, Hayden, J.A., van der Windt, D.A., Cartwright, J.L., Cote, P., Bombardier, C. (2013). Assessing bias in studies of prognostic factors. Ann Intern Med 158, Hemingway, H., Croft, P., Perel, P., Hayden, J.A., Abrams, K., Timmis, A., Briggs, A., Udumyan, R., Moons, K.G.M., Steyerberg, E.W., Roberts, I., Schroter, S., Altman, D.G., Riley, R.D. (2013). Prognosis research strategy (PROGRESS) 1: A framework for researching clinical outcomes. BMJ 346, e5595 e5595. Itz, C.J., Geurts, J.W., van Kleef, M., Nelemans, P. (2013). Clinical course of non-specific low back pain: A systematic review of prospective cohort studies set in primary care. Eur J Pain 17, Kamaleri, Y., Natvig, B., Ihlebaek, C.M., Bruusgaard, D. (2008). Localized or widespread musculoskeletal pain: Does it matter? Pain 138, Kellgren, J.H. (1939). On the distribution of pain arising from deep somatic structures with charts of segmental pain areas. Clin Sci, 4, Kirkwood, B.R. and Sterne, J.A.C. (2003). Assessing reproducibility of measurements. In Essential Medical Statistics (United Kingdom: Blackwell Science Ltd), pp Kjaer, P., Wedderkopp, N., Korsholm, L. and Leboeuf-Yde, C. (2011). Prevalence and tracking of back pain from childhood to adolescence. BMC Musculoskelet Disord 12, 98. Kongsted, A., Kent, P., Hestbaek, L. and Vach, W.(2015). Patients with low back pain had distinct clinical course patterns that were typically neither complete recovery nor constant pain. A Latent Class Analysis of longitudinal data. Spine J 15, Kristman, V.L., Hartvigsen, J., Leboeuf-Yde, C., Kyvik, K.O., Cassidy, J.D. (2012). Does radiating spinal pain determine future work disability? A retrospective cohort study of 22,952 Danish Twins. Spine 37, Leboeuf-Yde, C., Nielsen, J., Kyvik, K.O., Fejer, R., Hartvigsen, J. (2009). Pain in the lumbar, thoracic or cervical regions: Do age and gender matter? A population-based study of 34,902 Danish twins years of age. BMC Musculoskelet Disord 10, 39. Leboeuf-Yde, C., Fejer, R., Nielsen, J., Kyvik, K.O., Hartvigsen, J. (2011). Consequences of spinal pain: Do age and gender matter? A Danish cross-sectional population-based study of 34,902 individuals years of age. BMC Musculoskelet Disord 12, 39. Leboeuf-Yde, C., Fejer, R., Nielsen, J., Kyvik, K.O., Hartvigsen, J. (2012). Pain in the three spinal regions: The same disorder? Data from a population-based sample of 34,902 Danish adults. Chiropr Man Therap 20, 11. Lemeunier, N., Leboeuf-Yde, C., Gagey, O. (2012). The natural course of low back pain: A systematic critical literature review. Chiropr Man Therap 20, 33. Mikkelsson, M., Salminen, J.J., Kautiainen, H. (1997). Non-specific musculoskeletal pain in preadolescents. Prevalence and 1-year persistence. Pain 73, Moher, D., Liberati, A., Tetzlaff, J., Altman, D.G. and Group, P. (2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. BMJ 339, b2535. Niemelainen, R., Videman, T. and Battie, M.C. (2006). Prevalence and characteristics of upper or mid-back pain in Finnish men. Spine (Phila Pa 1976) 31, Eur J Pain (2016) 2016 European Pain Federation - EFIC â

9 Tamcan, O., Mannion, A.F., Eisenring, C., Horisberger, B., Elfering, A., Muller, U. (2010). The course of chronic and recurrent low back pain in the general population. Pain 150, Taylor, J.B., Goode, A.P., George, S.Z., Cook, C.E. (2014). Incidence and risk factors for first-time incident low back pain: A systematic review and meta-analysis. Spine J 14, Walton, D.M., Carroll, L.J., Kasch, H., Sterling, M., Verhagen, A.P., Macdermid, J.C., Gross, A., Santaguida, P.L., Carlesso, L. (2013). An overview of systematic reviews on prognostic factors in neck pain: Results from the International Collaboration on Neck Pain (ICON) Project. Open Orthop J 7, Wedderkopp, N., Leboeuf-Yde, C., Andersen, L.B., Froberg, K. and Hansen, H.S. (2001). Back pain reporting pattern in a Danish population-based sample of children and adolescents. Spine (Phila Pa 1976) 26, Supporting Information Additional Supporting Information may be found online in the supporting information tab for this article: Method S1. Search strategy. Method S2. Explanation of the QUIPS tool. Results S1. Description of studies investigating the incidence of mid-back pain in the general population. Results S2. Description of studies investigating the prognosis of mid-back pain in the general population European Pain Federation - EFIC â Eur J Pain (2016) 9