Prognosis of Acute Idiopathic Neck Pain is Poor: A Systematic Review and Meta-Analysis

824

REVIEW ARTICLE (META-ANALYSIS)

Prognosis of Acute Idiopathic Neck Pain is Poor: A Systematic Review and Meta-Analysis Julia M. Hush, PhD, C. Christine Lin, PhD, Zoe A. Michaleff, BAppSc(Phty) Hons, Arianne Verhagen, PhD, Kathryn M. Refshauge, PhD Listen to this article’s audio podcast at www.archives-pmr.org. ABSTRACT. Hush JM, Lin CC, Michaleff ZA, Verhagen A, Refshauge KM. Prognosis of acute idiopathic neck pain is poor: a systematic review and meta-analysis. Arch Phys Med Rehabil 2011;92:824-9. Objective: To conduct a systematic review and meta-analysis on the prognosis of acute idiopathic neck pain and disability. Data Sources: EMBASE, CINAHL, Medline, AMED, PEDro, and CENTRAL were searched from inception to July 2009, limited to human studies. Reference lists of relevant systematic reviews were searched by hand. Search terms included: neck pain, prognosis, inception, cohort, longitudinal, observational, or prospective study and randomized controlled trial. Study Selection: Eligible studies were longitudinal cohort studies and randomized controlled trials with a no treatment or minimal treatment arm that recruited an inception cohort of acute idiopathic neck pain and reported pain or disability outcomes. Eligibility was determined by 2 authors independently. Seven of 20,085 references were included. Data Extraction: Pain and disability data were extracted independently by 2 authors. Risk of bias was assessed independently by 2 authors. Data Synthesis: Statistical pooling showed a weighted mean pain score (0 –100) of 64 (95% confidence interval [CI], 61– 67) at onset and 35 (95% CI, 32–38) at 6.5 weeks. At 12 months, neck pain severity remained high at 42 (95% CI, 39 – 45). Disability reduced from a pooled weighted mean score (0 –100) at onset of 30 (95% CI, 28 –32) to 17 (95% CI, 15–19) by 6.5 weeks, without further improvement at 12 months. Studies varied in length of follow-up, design, and sample size. Conclusions: This review provides Level I evidence that the prognosis of acute idiopathic neck pain is worse than currently recognized. This evidence can guide primary care clinicians

From the Discipline of Physiotherapy, University of Sydney (Hush, Refshauge) and the Musculoskeletal Division, The George Institute for Global Health (Lin, Michaleff) Sydney, NSW, Australia; Department of General Practice, Erasmus MC, University Medical Centre, Rotterdam, The Netherlands (Verhagen). Presented to the International Association for the Study of Pain World Pain Congress, September 2, 2010, Montreal, QC, Canada. Supported by the National Health and Medical Research Council, Australia; and an Australian Postgraduate Award provided by the Australian Commonwealth Government. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprint requests to Julia M. Hush, PhD, Discipline of Physiotherapy, Faculty of Health Sciences University of Sydney, 75 East St, Lidcombe, NSW 2141, Australia, e-mail: [email protected]. Published online April 1, 2011 at www.archives-pmr.org. 0003-9993/11/9205-00933$36.00/0 doi:10.1016/j.apmr.2010.12.025

Arch Phys Med Rehabil Vol 92, May 2011

when providing prognostic information to patients. Further research to identify prognostic factors and long-term outcomes from inception cohorts would be valuable. Key Words: Meta-analysis [publication type]; Neck pain; Prognosis; Rehabilitation; Review [publication type]. © 2011 by the American Congress of Rehabilitation Medicine ECK PAIN IS ONE OF THE MOST common musculoN skeletal disorders, with an estimated annual incidence in Most the general population between 15% and 18%. 1,2

episodes of neck pain are of unknown origin, usually referred to as nonspecific or idiopathic neck pain,3 and are frequently recurrent.4 Neck pain affects activities of daily living and can have a major impact on quality of life.5 In addition to personal suffering, neck pain results in a substantial societal burden in terms of loss of productivity and continuing financial costs.6 There is a lack of high-quality evidence about the prognosis of acute idiopathic neck pain in clinical practice guidelines.7,8 As a result, primary care clinicians do not currently have high-level evidence to guide clinical decision-making or provide prognostic advice to patients with acute neck pain. Patients with spinal pain want information about the expected course of their condition9 and provision of such advice enhances patient satisfaction.10,11 Quality evidence-based data about prognosis also provides a basis for policy makers to allocate health resources. While systematic reviews have been conducted on the prognosis of whiplash-associated neck disorders and acute low back pain,12-14 we are unaware of any such review of the course of acute idiopathic neck pain. Previous reviews on neck pain prognosis have been limited to prevalent cases and survival cohorts.4,15 Clinically relevant and accurate estimates of prognosis are derived from inception cohort studies of patients from the acute onset of the condition.16 The 2000 to 2010 Neck Pain Task Force identified that high quality evidence about neck pain prognosis is a research priority.17 The aim of this study was to obtain an estimate of acute idiopathic neck pain prognosis by performing a systematic review and meta-analysis of inception cohort studies. METHODS Data Sources and Searches We conducted this systematic review in accordance with Preferred Reporting Items for Systematic Reviews and MetaAnalyses guidelines. Studies were identified through searches of the following databases from their inception to July 2009: EMBASE, CINAHL, Medline, AMED, PEDro, and CENTRAL.

List of Abbreviations CI

confidence interval

825

PROGNOSIS OF ACUTE NECK PAIN, Hush

Records identified through database searches n=20,085

Records after duplicates removed n=18,571

Records screened n=18,571

Records excluded n=18,262

Full-text articles assessed for eligibility n=309

Full-text articles excluded (n=302)

Studies included in quantitative synthesis n=7 articles reported in 6 studies

Reasons: Ineligible study design (n=246) Inappropriate population (n=27) Unable to translate (n=17) Unable to retrieve (n=9) No relevant outcome (n=3)

• • • • •

Fig 1. Retrieval of studies for review.

Searches were limited to human studies. Key search terms used to identify the study population of interest were: neck pain, neck injury, neck muscles, cervical, or cervical vertebrae. Search terms used to identify studies of appropriate design were: inception, cohort, longitudinal, observational or follow-up studies, randomized controlled trial, clinical trial. Search terms used to identify relevant outcomes were: prognosis, recovery, outcome, and/or prediction. Reference lists of relevant systematic reviews were searched by hand. Foreign language articles were included if translation could be conducted. Two authors independently determined the eligibility of selected articles. Disagreements were resolved first by consensus and then by consultation with a third author. Study Selection Articles were eligible for inclusion if they were of prospective design that charted the course of the neck pain, enrolled an inception cohort (⬍6wk after onset of pain) of adult participants with idiopathic neck pain, and reported pain or disability outcomes. It should be noted that there is no commonly agreed time period for the assembly of an inception cohort; however, the principle is that participants are assembled at a similar and early time in the course of their condition and followed prospectively. While the common point of division between acute and chronic pain is 3 months, we felt that this period was too long and decided to include studies of participants with pain onset within 6 weeks. Randomized controlled trials were also included if pain or disability outcomes were reported for a no treatment or minimal treatment arm (eg, sham or placebo treatments). Studies that enrolled participants with mixed conditions (eg, back and neck pain) or duration of symptoms (eg,

acute and chronic) were included if at least 75% of the sample had idiopathic neck pain of less than 6 weeks duration, or if data for the different populations were reported separately. Articles were excluded if they described participants with whiplash conditions or specific pathology of the cervical spine such as fracture, dislocation, infection, or tumor. There was no restriction on the length of follow-up, except that randomized controlled trials, which only collected outcomes before and after a single treatment session, were excluded. Retrieval of studies is illustrated in figure 1. Data Extraction and Quality Assessment Pain and disability data and study characteristics were extracted independently by 2 authors. Risk of bias was assessed with criteria used by Pengel et al12 in their systematic review of low back pain prognosis, with modifications as follow. The criterion of blinded outcomes (assessor unaware of at least 1 prognostic factor used to predict prognostic outcome) was not relevant because prognostic factors were not investigated. Where multiple articles reported on the same cohort, an individual quality item was coded as “yes” if any of the articles satisfied the criteria. The item completeness of follow-up was scored “yes” whenever the studies explicitly reported at least 1 prognostic outcome that was available from at least 80% of the study population. When studies did not specifically report loss to follow-up but performed the analysis based on all included participants, this item was also scored as a “yes”. Two authors independently assessed risk of bias for each included study. We resolved disagreements first by consensus and then by consultation with a third author. Arch Phys Med Rehabil Vol 92, May 2011

826

PROGNOSIS OF ACUTE NECK PAIN, Hush

of their condition. One study of small sample size21 did not report the specific duration of the neck pain for those grouped in the acute (⬍7wk) category.

Data Synthesis and Analysis Data from the minimal or no treatment arm were extracted from randomized controlled trials. Further information or data were requested from authors when required. We converted pain and disability scores to a 0 to 100 scale to facilitate comparison between studies. We pooled these data from neck pain onset to 12 months, using generic inverse variance analysis with the software Review Manager (RevMan) Version 5.0.a One study reported on pain outcomes at 1 week18 and another reported pain outcomes at 4 weeks.19 These data were pooled to provide data for pain outcomes at 2.5 weeks. Similarly, pain data from 3 studies18,20,21 reported at either 6 or 7 weeks were pooled to indicate outcomes at 6.5 weeks. Two studies20,21 reported disability outcomes at either 6 or 7 weeks and these data were pooled for 6.5 weeks follow-up.

Risk of Bias Assessment The initial agreement of the 2 reviewers on risk of bias assessment was 80% (␬⫽0.79). After discussion, no disagreements persisted. All studies met 50% or more of the criteria (table 2). Most studies defined the sample and/or provided information about the representativeness of the study sample. Loss-to-follow-up was 0% in 3 studies18,19,24; 22% in Bale and Newell,21 26% in Vos et al,20 and not reported in Landers et al.23 All but 1 study23 provided outcomes as raw data, percentages, survival rates, or continuous outcomes.

RESULTS The search retrieved 20,085 references, of which 6 studies, published in 7 articles, fulfilled our inclusion criteria (table 1).18-24 Two articles20,22 reported the same participant population and were considered a single study. Three of the included studies used an inception cohort design, following the course of acute neck pain in participants recruited from chiropractic,21 physiotherapy,23 and general practice20,22 clinics. The remaining 3 studies were randomized controlled trials from which we extracted data from the minimal treatment groups; these were: sham ultrasound,19 placebo tetrazepam,24 or neck collar, rest, and analgesic medication.18 The majority of subjects (244/283, 86%) in these studies had experienced an onset of neck pain in the past 2 weeks (see table 1) and are therefore representative of patients at the inception

Prognosis of Acute Neck Pain Pain and disability data extracted from individual studies are shown in tables 3 and 4 and pooled data plotted in figures 2 and 3. Data from Salzmann et al24 could not be included in statistical pooling because categorical scales were used. Five studies reported on neck pain severity.18-21,23 At the onset of acute neck pain, the pooled weighted mean pain score on a 0 to 100 scale was 64 (95% CI, 61– 67). One small study18 of 10 participants reported a low pain severity (mean ⫾ SD, 18⫾23) at 1 week (see table 3), reducing the weighted mean pain outcome at 2.5 weeks to 28 (95% CI, 21–35) that had been pooled from 1-week and 4-week data. At 6.5 weeks, weighted mean pain severity was 35 (95% CI, 32–38). After this initial reduction of neck pain by 45% from onset to 6.5 weeks, pain severity does not appear to resolve further from this stage to 52

Table 1: Characteristics of Included Studies Study/Year

Bale,

21

2005

Participants* ● ● ● ●

Brockow,19 2008

● ● ● ●

Landers,23 2008

● ● ● ●

Nordemar,18 1981

● ● ● ●

Salzmann,24 1993

● ● ● ●

Vos,20,22 2008, 2006

● ● ● ●

N⫽7 (chiropractic clinic) Inception time: ⬍7wk Sex: 43 Age: 36⫾11 N⫽63 (GP clinics) Inception time: ⬍7d Sex: 78 Age: 45⫾14 N⫽23 (physiotherapy clinics) Inception time: ⬍7d Sex‡: 71 Age‡: 50⫾13 N⫽10 Inception time: ⬍3d Sex: 50 Age: 43⫾16 N⫽10 Inception time: NR Sex: 40 Age: 42⫾NR N⫽187 (GP clinics) Inception time: ⬍6wk (79% ⬍2wk) Sex: 64 Age: 38⫾13 F, 43⫾15 M

Design

Cohort study

RCT† comparing subcutaneous carbon dioxide insufflations with sham ultrasound

Outcome Measures

Follow Up (wk)

Pain: Bournemouth questionnaire for neck pain Disability: Bournemouth questionnaire for disability Pain: 0–100mm visual analog scale Disability: NR

4–10

Cohort study

Pain: NR Disability: Neck Disability Index

RCT comparing TENS or manual therapy added to a minimal intervention of neck collar, rest and analgesics† RCT comparing tetrazepam (3mg for 3d) with placebo tetrazepam†

Pain: 0–100mm visual analog scale Disability: NR

Cohort study

Pain: categorical scale: 1 (no pain) to 5 (severe pain) Disability: categorical scale: 1 (no disability) to 5 (severe disability) Pain: 0–10 visual analog scale Disability: Neck Disability Index

3

12

1, 6, 12

1

6, 12, 26, 52

NOTE. Age: mean ⫾ SD. Sex: % women. Abbreviations: F, female; GP, general practitioner; M, male; NR, not reported; RCT, randomized controlled trial; TENS, transcutaneous electrical nerve stimulation. *Data presented for acute subgroup where studies included acute and chronic participants. † Data extracted from sham minimal/no treatment group. ‡ Age and sex not reported separately for acute subgroup.

Arch Phys Med Rehabil Vol 92, May 2011

827

PROGNOSIS OF ACUTE NECK PAIN, Hush Table 2: Risk of Bias Assessment of Included Studies of the Course of Acute Neck Pain* Study/Year

Defined Sample†

Representative Sample‡

Complete Follow Up§

Prognosis储

Bale,21 2005 Brockow,19 2008 Landers,23 2008 Nordemar,18 1981 Salzmann,24 1993 Vos,20,22 2008, 2006

Yes Yes Unclear No Yes Yes

No Yes Yes Yes Unclear Yes

No Yes Yes Yes Yes No

Yes Yes No Yes Yes Yes

*Statistical adjustment and blinded outcome criteria not included as prognostic factors were not investigated. Detailed inclusion and exclusion criteria. Participants selected by random selection or consecutive cases. § At least 1 prognostic outcome available from at least 80% of the study population at any time point after inception. 储 Studies provide raw data, percentages, survival rates, or continuous outcomes. † ‡

weeks. At 1 year, pain severity was 42 (95% CI, 39 – 45) on a 0 to 100 scale. Four studies reported disability data20,21,23,24 (see tables 3 and 4). Statistical pooling of continuous data showed that the course of disability was similar to pain severity (see fig 3). Disability reduced from a pooled weighted mean score of 30 (95% CI, 28 –32) on a 0 to 100 scale at onset, to 23 (95% CI, 21–25) at 1 week. By 6.5 weeks, neck disability had reduced by 42% to 17 (95% CI, 15–19), but no further improvements were evident at 12, 26, or 52 weeks (see fig 3).

quality of life. We reveal that the intensity of persisting neck pain is twice as high than that for low back pain, which remains at a constant level of 15 on a 0 to 100 scale from 3 to 12 months.12 The course of disability for neck pain reported in this review (reducing from 30 on a 0 –100 scale at baseline, to remain at 17 from 6.5wk to 1y) is comparable with low back pain prognosis.12,13 The course of acute idiopathic neck pain beyond 1 year is unknown. It is unclear whether the average pain and disability scores at 12 months are derived from a small subgroup with very high levels of pain and disability. It is also unclear whether ongoing symptoms measured at 12 months are due to persisting symptoms from the initial acute episode or to recurrent episodes. The design and results of our review differ from the 2 previous systematic reviews on the prognosis of neck pain,4,15 neither of which sought to review inception cohort studies. Borghouts et al15 reviewed literature published until 1996, therefore, not including relevant studies published during the past 14 years. Other aspects of the Borghouts’ review15 that differentiate from this review are: the majority of included studies (11 of 12) were not conducted in a primary care setting, only 2 of

DISCUSSION In this study, to our knowledge the first systematic review about the course of acute idiopathic neck pain from its inception, we report that outcomes for acute neck pain are surprisingly poor and resolution is incomplete. There is a rapid decrease in pain by 45% and disability by 43% during the first 6.5 weeks. While this degree of symptom reduction may be clinically worthwhile to some patients,25 the severity and duration of symptom persistence (between 37– 42 on a 0 –100 scale up to 1y) are likely to interfere with daily functioning and

Table 3: Pain Scores Extracted From Included Studies, Converted to a 0 to 100 Scale Study/Year

Baseline

1wk

4wk

6–7wk

12wk

26wk

52wk

Bale,21 2005 Brockow,19 2008 Landers,23 2008 Nordemar,18 1981 Salzmann,24 1993 Vos,20,22 2008, 2006

59⫾20 69⫾16 NR 45⫾13 * 64⫾20

NR NR NR 35⫾45 * NR

NR 31⫾32 NR NR NR NR

11⫾9 NR NR 0⫾0 NR 41⫾23

NR NR NR 0⫾0 NR 37⫾23

NR NR NR NR NR 37⫾22

NR NR NR NR NR 42⫾20

Values are mean ⫾ SD. Abbreviation: NR, not reported. *Categorical data reported of % participants scoring from 0 (no pain) to 5 (severe pain). Baseline: 40% scored 5; 50% scored 4; 10% scored 3. 1 week: 10% scored 5; 20% scored 4; 70% scored 3. Table 4: Disability Scores Extracted From Included Studies, Converted to a 0 to 100 Scale Study/Year 21

Bale, 2005 Brockow,19 2008 Landers,23 2008 Nordemar,18 1981 Salzmann,24 1993 Vos,20,22 2008, 2006

Baseline

1wk

6–7wk

12wk

26ws

52wk

50⫾29 NR 39⫾17 NR * 29⫾14

NR NR NR NR * 23⫾15

9⫾9 NR NR NR NR 18⫾15

NR NR 25⫾20 NR NR 17⫾15

NR NR NR NR NR 17⫾14

NR NR NR NR NR 16⫾14

Values are mean ⫾ SD. Abbreviation: NR, not reported. *Categorical data reported of % participants scoring from 0 (no disability) to 5 (severe disability). Baseline: 50% scored 5; 50% scored 4. 1 week: 10% scored 5; 30% scored 4; 60% scored 3.

Arch Phys Med Rehabil Vol 92, May 2011

828

PROGNOSIS OF ACUTE NECK PAIN, Hush

12 studies reported on acute patients (defined as ⬍3mo), and retrospective studies were included. No results about prognosis in terms of disability were available for acute patients. Pain intensity outcomes (visual analog scale or Numerical Rating Scale) were reported for 1 study of acute patients as “100% decrease”15(p7) from an unspecified initial value and at an unspecified time point. In the Carroll et al4 review, relevant literature published until 2007 was included, although it missed the high quality Vos et al20 study, published in 2008. Of the 5 separate cohorts included in this review, none exclusively investigated acute neck pain. One study26 reported on patients seeking general practitioner care for “new episodes of neck pisodes of neck pain,”26(pS77) but 62% of these had had neck or shoulder pain for many months. In their best-evidence synthesis,27 a meta-analysis of the course of neck pain was not conducted. Rather, outcomes were described in general terms, such as: “Most of the evidence indicates that between half and three quarters of people who experience neck pain at some initial point will report neck pain 1 to 5 years later.”4(pS78) Therefore, the systematic review and meta-analysis on the prognosis of acute neck pain reported in this article, represent a new contribution to this body of literature in that it provides precise quantitative estimates of pain and disability outcomes for the 12 months after onset of acute neck pain. These results can be compared with a recent systematic review on the prognosis of acute whiplash14 that included inception cohort studies of patients with acute neck pain due to a whiplash injury of a similar duration (within 6wk). The results mirror those reported in the present review with respect to the general course of pain and disability: on a 0 to 100 scale, whiplash-associated pain reduced from 41 at baseline to 24 at 6 months, and disability reduced from 36 at baseline to 20 at 6 months.14 A distinction is that for whiplash, Kamper et al14 showed that pain and disability continued to improve for up to 12 weeks after whiplash injury with a pooled 12 month score of 25 and 19, respectively. Our results show that idiopathic neck pain does not resolve further after 6.5 weeks and that pain severity at 12 months is higher (42 on a 0 –100 scale), although disability is similar (17) at 12 months. A further contrast is the scarcity of research in idiopathic neck pain compared with the whiplash review, which identified 38 unique inception cohorts. Given the high prevalence of neck pain in the general population and the individual suffering and societal burden that this musculoskeletal condition causes, we were surprised to find so few large inception cohort studies with long-term follow-up conducted on this topic. This stands in contrast to the

Fig 3. The course of idiopathic neck pain-related disability (pooled weighted means and 95% CIs).

abundance of clinical trials that have investigated treatment efficacy for neck pain. It is also apparent that there is insufficient evidence to understand factors associated with prognosis of acute neck pain. A comprehensive picture of both long-term prognosis and prognostic factors will be attained from further studies that follow inception cohorts of patients with acute idiopathic neck pain for at least 1 year and use large, representative samples. It will be particularly useful to monitor the recurrence of pain episodes in future cohort studies. Study Limitations One limitation is the small number of studies that could be included in the meta-analysis, and only 1 (Vos20) was a purposely-designed inception cohort study. Nonetheless, in addition to the Vos20 study, this review provides stronger evidence about the earlier course of neck pain, by pooling data from 6 different relevant studies. This is particularly useful because the course during the first 12 weeks of acute idiopathic neck pain differs from that of low back pain and neck pain after a whiplash injury (as discussed above with reference to the Kamper14 review of whiplash recovery). It should be noted that data for weeks 26 and 52 are only derived from the Vos20 study. It is also possible that the evidence may be limited by risk of bias, although all studies included in the review met at least 50% of the quality criteria. Four studies had small sample sizes ranging from 7 to 23 participants,18,21,23,24 which is likely to impact on heterogeneity. Nonetheless, the relatively small CIs of pooled data indicate reasonable precision of the pain and disability estimates. CONCLUSIONS To our knowledge, this study provides the first systematic review and meta-analysis of acute idiopathic neck pain prognosis from its inception. The results show that the prognosis of acute idiopathic neck pain is markedly worse than previously recognized and that resolution is incomplete. This level 1 evidence may be useful for the allocation of health resources by policy makers and can be incorporated into future clinical practice guidelines to guide primary care clinicians when providing prognostic information to patients.

Fig 2. The course of idiopathic neck pain severity (pooled weighted means and 95% CIs).

Arch Phys Med Rehabil Vol 92, May 2011

Acknowledgements. We thank research assistant Angie Johnson, BAppSc(Phty), for her excellent technical assistance with this study.

PROGNOSIS OF ACUTE NECK PAIN, Hush

References 1. Croft PR, Lewis M, Papageorgiou AC, et al. Risk factors for neck pain: a longitudinal study in the general population. Pain 2001; 93:317-25. 2. Côté P, Cassidy JD, Carroll LJ, Kristman V. The annual incidence and course of neck pain in the general population: a populationbased cohort study. Pain 2004;112:267-73. 3. Bogduk N. Regional musculoskeletal pain. The neck. Baillieres Best Pract Res Clin Rheumatol 1999;13:261-85. 4. Carroll LJ, Hogg-Johnson S, van der Velde G, et al. Course and prognostic factors for neck pain in the general population: results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and Its Associated Disorders. Spine 2008;33(4 Suppl):S75-82. 5. Ariëns GAM, Borghouts JAJ, Koes BW. Neck pain. In: Crombie IK, editor. The epidemiology of pain. Seattle: IASP Pr; 1999. p 235-55. 6. Borghouts JA, Koes BW, Vondeling H, Bouter LM. Cost-of-illness of neck pain in The Netherlands in 1996. Pain 1999;80:629-36. 7. National Health and Medical Research Council. Evidence-based management of acute musculoskeletal pain. Canberra: National Health and Medical Research Council; 2003. 8. Guzman J, Haldeman S, Carroll LJ, et al. Clinical practice implications of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and its Associated Disorders: from concepts and findings to recommendations. Spine 2008;33(4 Suppl):S199-213. 9. May SJ. Patient satisfaction with management of back pain: part 1: what is satisfaction? Review of satisfaction with medical management. Physiotherapy 2001;87:4-9. 10. May SJ. Patient satisfaction with management of back pain: part 2: an explorative, qualitative study into patients’ satisfaction with physiotherapy. Physiotherapy 2001;87:10-20. 11. Hills R, Kitchen S. Satisfaction with outpatient physiotherapy: focus groups to explore the views of patients with acute and chronic musculoskeletal conditions. Physiother Theory Pract 2007;23:1-20. 12. Pengel LH, Herbert RD, Maher CG, Refshauge KM. Acute low back pain: systematic review of its prognosis. BMJ 2003;327:323. 13. Henschke N, Maher CG, Refshauge KM, et al. Prognosis in patients with recent onset low back pain in Australian primary care: inception cohort study. BMJ 2008;337:a171. 14. Kamper SJ, Rebbeck TJ, Maher CG, McAuley JH, Sterling M. Course and prognostic factors of whiplash: a systematic review and meta-analysis. Pain 2008;138:617-29. 15. Borghouts JA, Koes BW, Bouter LM. The clinical course and prognostic factors of non-specific neck pain: a systematic review. Pain 1998;77:1-13.

829

16. Fletcher RH, Wagner EH. Clinical epidemiology: the essentials. 3rd ed. Baltimore: Williams & Wilkins; 1996. 17. Carroll LJ, Hurwitz EL, Côté P, et al. Research priorities and methodological implications: the Bone and Joint Decade 20002010 Task Force on Neck Pain and Its Associated Disorders. Spine (Phila Pa 1976) 2008;33:S214-20. 18. Nordemar R, Thörner C. Treatement of acute cervical pain- a comparative group study. Pain 1981;10:93-101. 19. Brockow T, Heissner T, Franke A, Resch KL. Evaluation of the efficacy of subcutaneous carbon dioxide insufflations for treating acute non specific neck pain in general practice: a sham controlled randomized trial. Eur J Pain 2008;12:9-16. 20. Vos CJ, Verhagen AP, Passchier J, Koes BW. Clinical course and prognostic factors in acute neck pain: an inception cohort study in general practice. Pain Med 2008;9:572-80. 21. Bale A, Newell D. Chiropractic for neck pain: a pilot study examining whether the duration of the pain affects the clinical outcome. Clinical Chiropractic 2005;8:179-88. 22. Vos C, Verhagen A, Koes B. Reliability and responsiveness of the Dutch version of the Neck Disability Index in patients with acute neck pain in general practice. Eur Spine J 2006;15:1729-36. 23. Landers MR, Creger RV, Baker CV, Stutelberg KS. The use of fear-avoidance beliefs and nonorganic signs in predicting prolonged disability in patients with neck pain. Man Ther 2008;13: 239-48. 24. Salzmann E, Wiedemann O, Loffler L, Sperber H. [Tetrazepam in the treatment of acute cervical syndrome. Randomized doubleblind pilot study comparing tetrazepam and placebo] [German]. Fortschritte der Medizin 1993;111:544-8. 25. Farrer JT, Young JP Jr, LaMoreaux L, Werth JL, Poole RM. Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain 2001;94:149-58. 26. Bot SD, van der Waal JM, Terwee CB, et al. Predictors of outcome in neck and shoulder symptoms: a cohort study in general practice. Spine (Phila Pa 1976) 2005;30:E459-70. 27. Carroll LJ, Cassidy JD, Peloso PM, et al. Methods for the best evidence synthesis on neck pain and its associated disorders: the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and Its Associated Disorders. Spine 2008;33(4 Suppl):S33-8. Supplier a. Cochrane Information Management System (IMS). 2008. Available at: http://ims.cochrane.org/revman/download/5.1_beta_download. Accessed March 11, 2011.

Arch Phys Med Rehabil Vol 92, May 2011