For Peer Review Only

BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from

ACUPUNCTURE FOR ACUTE NON-SPECIFIC LOW BACK PAIN: A PROTOCOL FOR A RANDOMISED, CONTROLLED, MULTICENTER INTERVENTION STUDY IN GENERAL PRACTICE – THE ACUBACK STUDY.

For peer review only

Journal: BMJ Open

Manuscript ID: bmjopen-2012-001164

Article Type: Protocol

Date Submitted by the Author: 21-Mar-2012

Complete List of Authors: Skonnord, Trygve; University of Oslo, Institute of Health and Society Skjeie, Holgeir; University of Oslo, Institute of Health and Society Brekke, Mette; University of Oslo, Institute of Health and Society Grotle, Margreth; Oslo University Hospital, Clinic for surgery and neurology Lund, Iréne; Karolinska Institutet, Department of Physiology and Pharmacology Fetveit, Arne; University of Oslo, Institute of Health and Society

Primary Subject General practice / Family practice Heading:

Secondary Subject Heading: Complementary medicine http://bmjopen.bmj.com/

COMPLEMENTARY MEDICINE, Back pain < ORTHOPAEDIC & TRAUMA Keywords: SURGERY, PAIN MANAGEMENT, PRIMARY CARE, Clinical trials < THERAPEUTICS, Adverse events < THERAPEUTICS

on September 29, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 ACUPUNCTURE FOR ACUTE NON-SPECIFIC LOW BACK 4 5 PAIN: A PROTOCOL FOR A RANDOMISED, CONTROLLED, 6 MULTICENTER INTERVENTION STUDY IN GENERAL 7 8 PRACTICE – THE ACUBACK STUDY. 9 10 Corresponding author: Trygve Skonnord, MD. Department of General Practice, Institute of 11 Health and Society, University of Oslo, Post-box 1130 Blindern N-0318 Oslo, Norway 12 E-mail: [email protected], Cell phone: +47 41323232, Fax: +47 33306899. 13 14 Authors: Trygve Skonnord,1 Holgeir Skjeie,1 Mette Brekke,1 Margreth Grotle,2 Iréne Lund,3 15 Arne Fetveit.For1 peer review only 16 17 18 Affiliation information: 1 19 Department of General Practice, Institute of Health and Society, University of Oslo, PO box 20 1130 Blindern N-0318 Oslo, Norway. 21 2FORMI, Clinic for surgery and neurology Oslo University Hospital, Ullevaal, N-0407 Oslo, 22 Norway 23 3Department of Physiology and Pharmacology, Karolinska Institutet, SE-171 77 Stockholm, 24 Sweden. 25

26 27 Running title: Acupuncture for acute non-specific low back pain. 28 29 Key words: Acupuncture Therapy, Acute Pain, Clinical Research Protocol, General Practice, 30 Low Back Pain 31 32 Word Count: 3841

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 on September 29, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 15 BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 ABSTRACT: 4 Introduction: 5 Some general practitioners (GPs) treat acute low back pain (LBP) with acupuncture, despite 6 lacking evidence of its effectiveness for this condition. 7 8 The aim of this study is to evaluate whether a single treatment-session with acupuncture can 9 reduce time to recovery when applied in addition to standard LBP-treatment according to the 10 Norwegian national guidelines. Analyses of prognostic factors for recovery and cost- 11 effectiveness will also be carried out. 12 Methods and analysis: 13 In this randomised, controlled, multicenter study in general practice in Southern Norway, 270 14 patients will be allocated into one of two treatment-groups, using a web-based application 15 based on blockFor randomisation. peer Outcome reviewassessor will be blinded only for group allocation of the 16 patients. 17 18 The control group will receive standard treatment, while the intervention group will receive 19 standard treatment plus acupuncture treatment. There will be different GPs treating the two 20 groups, and both groups will just have one consultation. Adults who consult their GP because 21 of acute LBP will be included. Patients with nerve root affection, "red flags", pregnancy, 22 previous sick leave more than 14 days and disability pension will be excluded. The primary 23 outcome of the study is the median time to recovery (days). The secondary outcomes are rated 24 global improvement, back-specific functional status, sick leave, medication, GP visits and 25 side effects. A pilot study will be conducted. 26 Ethics and dissemination: 27 28 Participation is based on informed written consent. We will apply for an ethical approval from 29 the Regional Committee for Medical and Health Research Ethics (REC), when the study- 30 protocol is completed. Results from this study, positive or negative, will be disseminated in 31 scientific medical journals. 32 Discussion:

33 This study aims to explore whether acupuncture is a useful clinical tool for treating acute LBP http://bmjopen.bmj.com/ 34 in general practice. 35 Trial Registration: 36 ClinicalTrials.gov Identifier: NCT01439412. 37 38 39 40 41 on September 29, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 4 INTRODUCTION: 5 Low back pain (LBP) is a very common disorder with consequences for the individual patient 6 as well as for the society. Up to 80% of the population experiences back pain at least once in 7 their lifetime, about 50% during the previous year. Point prevalence is 15%, and the condition 8 1 9 relapses frequently, 40% within 6 months. Back pain is the medical condition that ranks highest in terms of Norwegian socio-economic expenses.2 Most people with acute LBP 10 3 11 experience improvements in pain and disability within a month, but the median time to 4 5 12 recovery recorded in studies on back pain varies widely, from 7 days to 58 days. 13 Acute LBP is treated primarily in the primary healthcare by general practitioners (GPs), 14 physiotherapists, manual therapists and chiropractors. Recommended treatment according to 15 clinical guidelinesFor contains peer information aboutreview the condition, advice only to stay active and, if 16 possible, avoid bed rest, early and gradual mobilisation after the acute phase, pain treatment 17 with paracetamol and/or non-steroidal anti-inflammatory drugs (NSAIDs) with time- 18 2 6 19 contingent doses. 20 GPs educated and trained in acupuncture also use acupuncture for the treatment of both acute 7 21 and chronic LBP cases. There is evidence that acupuncture is effective in chronic LBP, and 2 8- 22 such treatment is therefore recommended in the "National guidelines for LBP" in Norway. 23 10 Reviews and meta-analyses conclude that the documentation of acupuncture treatment for 24 acute LBP is limited by few and poorly conducted studies, however.10 11 25 26 In 1997, He presented a Chinese study with 100 patients afflicted with LBP (5 days to 6 27 months duration) randomised to either manual acupuncture with moxibustion plus Chinese 28 12 29 herbal medicine or Chinese herbal medicine alone. A later Cochrane review concludes that 30 this trial was of low methodological quality and showed limited evidence to back the fact that 31 the combined treatment was more effective for a global measure of pain and function in the 10 32 long-term follow-up.

33 http://bmjopen.bmj.com/ 34 Araki et al have published an abstract of a trial where 40 patients with acute LBP (< 3 days) 35 were randomised into two groups where the intervention group got acupuncture treatment in 36 the acupuncture point SI3 bilaterally and then performed back exercises. The control group, 37 however, received sham acupuncture treatment with mimicked needle insertion, after which 38 they were asked to perform back exercises. Araki et al found no difference between the 39 13 40 effects of acupuncture and that of the sham acupuncture. 41 on September 29, 2021 by guest. Protected copyright. 42 One of the studies referred to in the Cochrane review is a Norwegian study conducted by 43 Kittang et al.14 They found that acupuncture was as effective as medical treatment with 44 Naproxen per os in relation to pain and stiffness, but that the “Naproxen-group” had more 45 side effects and greater recurrences in the observation period. While Kittang et al conclude 46 that acupuncture is effective, this result is based on other studies showing the efficacy of 47 NSAIDs for acute LBP.15 48 49 50 Kennedy et al published a pilot study in 2008, which demonstrated the feasibility of a 51 randomised, controlled trial regarding the penetrating needle acupuncture when compared to 52 non-penetrating sham acupuncture for the treatment of acute LBP in primary care. The study 53 did not reveal any conclusions of the effects of acupuncture.16 54 55 Sham acupuncture involves inserting penetrating or non-penetrating needles in points that are 56 not classical acupuncture points and/or are not located in the same segment. Several studies 57 conclude, however, that sham acupuncture is not a valid placebo control, and this is explained 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 15 BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 17 18 3 by neurophysiologic effects of the sham treatment. Trials with sham acupuncture need to 4 be very comprehensive in order to demonstrate differences between the effects of real 5 acupuncture and the sham treatment.19 6 7 In 2006, Vas et al published a study protocol of a four-branch randomised controlled trial, 8 which intended to obtain further evidence on the effectiveness of acupuncture on acute LBP 9 20 10 and isolate the specific and non-specific effects of the treatment. The results have not yet 11 been published. 12 Shin et al are planning to perform a randomised, controlled trial with two arms comparing a 21 13 motion-style acupuncture with an NSAID injection. 14 15 However, weFor have not yet peer found evidence review favouring acupuncture only as a treatment for acute 16 LBP. Our clinical experience is that an acupuncture treatment of the distal and local 17 acupuncture points combined with small mobilising movements means faster recovery of the 18 pain. This treatment is according to textbooks on acupuncture.22 23 None of the mentioned 19 20 protocols though examine this kind of acupuncture. 21 22 Strong stimulation of the distal acupuncture points is thought to act through activation of the 23 extra-segmental pain inhibition, "Diffuse noxious inhibitory Controls” (DNIC) and the release 24 of endorphins and possibly, other endogenous compounds in the central nervous system. 25 Local acupuncture therapy immediately following the mobilisation movements may provide 26 additional benefits through segmental pain-inhibition along with the inhibition of myofascial 27 trigger points.24 28 29 30 Clear prognostic factors for the efficacy of acupuncture for chronic LBP have, however, not 31 been documented. In a large German study, the authors found that lower age, lower baseline 32 spinal function and more than 10 years of education indicated a better response to 25

33 acupuncture. Some researchers still find that optimistic treatment expectations improve the http://bmjopen.bmj.com/ 34 effects of acupuncture,26 27 while others find the best treatment effects in patients who are 35 neutral to acupuncture as a treatment method.28 There are no similar studies for acute LBP. 36 37 In the present study, we aim at exploring whether a single treatment-session with acupuncture 38 can reduce the time to recovery when applied in addition to the standard LBP-treatment in 39 general practice according to Norwegian national guidelines. We will not use sham 40 41 acupuncture as a control treatment. Further, we will not include a pure placebo group or a on September 29, 2021 by guest. Protected copyright. 42 “waiting list group”, because this will mean not giving treatment to patients with severe pain, 43 which we consider unethical. 44 Secondary aims are pain intensity, disability, sick leave and drug use, and in addition we will 45 evaluate the prognostic factors for recovery, and if possible, identify any factors that 46 characterize those who have beneficial effects of the acupuncture treatment. 47 A systematic review found evidence supporting the cost-effectiveness of the guideline- 48 endorsed treatments of acupuncture for sub-acute or chronic LBP.29 Apart from the few trials 49 of acupuncture treatment for acute LBP, there are no studies examining the cost-effectiveness 50 51 of acupuncture for acute LBP. Therefore, we also aim at carrying out a cost-effectiveness 52 analysis. 53 54 Our hypotheses are: 55 1. Acupuncture treatment contributes to faster pain-recovery in acute LBP compared to 56 standard treatment in general practice provided in accordance with the Norwegian 57 national guidelines. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 2. Acupuncture treatment for acute LBP improves function, and reduces drug use and 4 sick leave, compared to the standard treatment in general practice provided in 5 accordance with national guidelines. 6 3. Acupuncture treatment for acute LBP is a cost-effective treatment in general practice. 7 8 9 METHODS AND ANALYSIS: 10 This is a randomised, controlled, multicenter study. The GPs administering the acupuncture 11 have received their training through the education program of the Norwegian Society of 12 Medical Acupuncture and are members of the Network Group of Medical Acupuncture, under 13 the Norwegian College of General Practice. They are specialists in general practice and many 14 of them are experienced participants in randomised clinical trials, given a parallel ongoing 15 trial examiningFor the effects peer of acupuncture review on infantile colic.30 31only These GPs are located in 16 different areas in southern Norway, both cities and rural municipalities. GPs in the same 17 practices, but without training in acupuncture, however, need to be recruited to treat the 18 19 control group. 20 A medical secretary at every doctor's office will carry out the telephonic interview in which, 21 the patients will be informed about the study, the inclusion and exclusion criteria will be 22 checked, the randomisation of the patients will be performed and the distributing and 23 collecting questionnaires for the treatment day. A two-day workshop before the start of the 24 main study will be arranged for the participating doctors and medical secretaries. The 25 workshop will contribute to the standardization of information, reporting and patient 26 management in general. A website with information about the study will be created. 27 28 29 We plan to recruit patients with acute LBP and randomly allocate them into two study-groups 30 (A and C). Patients in both groups will receive standard treatment for LBP in accordance with 2 31 the Norwegian national guidelines. One group will receive only the standard treatment, while 32 the other group also will receive additional acupuncture treatment as described below. To

33 control for potential attention bias, we will measure the time spent on each patient in the http://bmjopen.bmj.com/ 34 intervention and control groups. Outcome assessor will be blinded for group allocation of the 35 patients. 36 37 Inclusion criteria: 38 39 - Adults (20-55 years) who contact their GP’s office because of acute nonspecific LBP 40 (0-14 days). 41 on September 29, 2021 by guest. Protected copyright. 42 Exclusion criteria: 43 - Nerve root affection and/or radiating pain below the knee. 44 - LBP with suspected “red flags”, i.e. infections, tumours and metastatic disease, 45 rheumatic disease, fractures and significant deformities of the spine. 46 - LBP which starts in pregnancy. 47 - Physician-reported sick leave of 14 days or more during the month before the 48 49 commencement of the back pain, for any reason.

50 - Disability pension. 51 52 Control group C: Standard treatment in general practice is provided in accordance with the 53 Norwegian national guidelines, i.e. general advice about activity, prescription of pain 54 relievers (paracetamol), and sick leave, if needed. 55 56 Acupuncture group A: This group will receive standard treatment as the control group and 57 in addition acupuncture treatment. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 15 BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 Acupuncture: The patient sits in a chair and the doctor stimulates the acupuncture points “the 4 Lumbar Pain Points” on the right hand, with acupuncture needles of type Seirin B - 8a 0.30 x 5 30 mm. These are two points located between the second and third and the fourth and fifth 6 metacarpal bones, immediately distal to the bases of the metacarpals. Insertion depth is 10-15 7 mm. The doctor stimulates the needles in a rotating up- and down movement to impart a 8 powerful needle sensation (called “de Qi”), and this is repeated in short sequences to maintain 9 10 the needle sensation for a total of 1 min. The needles shall stay in the hand during the rest of 11 the treatment. The patient is then asked to rise and to perform slow rotating pelvic movements 12 for 2 minutes, before lying down on a bench to be treated in the local points Huatuojiaji 13 ("Jiaji") with acupuncture needles of the type SEIRIN J-8 with sleeve 0.30 x 50 mm. These 14 points are located 1.5 cm lateral to the depressions below the spinous processes, and we will 15 acupuncture Forthem bilaterally peer in the segments review of the L2-L4 (6 needles) only at a depth of 3-4 cm. 16 They are stimulated manually until the patient experiences the needle sensation. Then, the 17 patient lies quietly on the bench for another 5 minutes before all the needles are removed. 18 The whole acupuncture session lasts for a total of 8 minutes. Thus part of the advice and 19 20 prescription part is done while the patient is receiving acupuncture treatment, in order to let 21 the consultation time for the two groups be as equal as possible. The consultation time will be 22 measured and registered by the GPs. 23 24 Choice of treatment: 25 The described treatment with the specific distal and local acupuncture points has been chosen 26 after a three-step process. We started with the treatment we used ourselves in clinical practice, 27 compared it with the literature and standardized it for this trial. Then we asked several 28 experienced acupuncture doctors how they treated acute LBP and what they thought about our 29 30 suggestion of standardization. 31 Based on this feedback, we asked an expert group of physicians and physiotherapists 32 experienced in administering acupuncture about which acupuncture points they preferred in

33 the treatment of acute LBP. The end result was considered the treatment according to best http://bmjopen.bmj.com/ 34 practice for this study. 35 36 Patient flow: 37 When a patient with acute LBP contacts the GP’s office, the medical secretary provides 38 information about the trial and asks whether the patient would want to participate in the study. 39 40 If the patient consents, the medical secretary asks the patient questions regarding inclusion- 41 and exclusion criteria. The inclusion and exclusion criteria are checked both in the telephone on September 29, 2021 by guest. Protected copyright. 42 interview by the secretary and in the consultation by the GP. All contacts are counted and it is 43 recorded at what level and for what reason exclusions are done. If the patient is eligible, the 44 medical secretary randomises the patient to one of the two groups through the Web-based 45 application at the Unit for Applied Clinical Research in Trondheim 46 (http://www.ntnu.no/dmf/akf). This is used to register the patient in the appropriate doctor's 47 timetable, but the patient does not know which group or which doctor he/she is going to meet. 48 At the GP’s office, the patient is required to fill in the consent form and the first questionnaire 49 50 (T0A). This is delivered to the medical secretary prior to the consultation. 51 52 Patients who are randomised to the control group will receive one standard treatment 53 provided by a GP without acupuncture training. The consultation will include history, 54 examination, information, prescription and advices, according to national guidelines for LBP. 55 If the doctor discovers exclusion criteria, the patient will be excluded from the trial. 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 Patients who are randomised to the acupuncture group will receive one single acupuncture 4 treatment in addition to the standard treatment as described for the control group. The 5 consultation will be provided by a GP who is trained in acupuncture. If the patient’s condition 6 requires further consultations in the follow-up time, these will be recorded. These 7 consultations, however, will not include additional acupuncture treatment. 8 9 10 When the patient has finished the consultation, they will fill in a post-treatment questionnaire 11 (T0B) and submit it to the secretary in an enclosed envelope. The patient then will receive a 12 back pain diary, which he/she will be required to fill in at home at the given times. The diary 13 also contains information about how the patient can choose to give the answers in an 14 electronic questionnaire (Figure 1). 15 For peer review only 16 Measurements: 17 For evaluation of the effects of treatment, we will use standardised instruments that have been 18 validated, both internationally and nationally.32 33 The outcome measures will be filled in at 19 baseline and after 1, 2, 4 and 12 weeks, and the following outcome measurements will be 20 21 included: 22 23 The primary outcome is the median time in days to the recovery of pain, measured on the first 24 day the patient scores 0 or 1 point on the Numerical Rating Scale (NRS). Clinically relevant 25 differences between the groups are considered to be minimum 3 days. 26 27 Secondary outcome measures will be: 28 1. Pain intensity is assessed by the numerical rating scale (NRS) before and immediately 29 after treatment and at the other follow-up times. The patient indicates his perceived 30 31 pain intensity on an 11-point scale from 0 to 10 with endpoints indicating "no pain" 32 and "worst imaginable pain". Based on other studies, the clinically relevant

33 improvement is estimated as being minimum 2-3 points for individual patients and 1 http://bmjopen.bmj.com/ 34 point as the difference between the groups. 35 2. Back specific functional status by the Roland Morris Disability Questionnaire (RDQ). 36 This measures patients' perceptions of function.34 The patient answers yes or no 37 concerning 24 allegations about the activities and conditions, depending on whether 38 they feel that the statement describes them on this day. 39 3. Sick leaves, the number of days away from work due to back pain. 40 41 4. Global measure of improvement (Likert improvement assessment scale). This assesses on September 29, 2021 by guest. Protected copyright. 42 the patients’ perceptions of change, stated in whole numbers from 1 = much better to 5 43 = much worse. 44 5. Use of medication (paracetamol, eventually others), counting of daily consumption. 45 6. Number of new visits at the GP’s office. 46 7. Side effects of treatment (acupuncture and medications). 47 8. Health-related quality-of-life by the EuroQoL (EQ-5D). This enumerates 5 questions 48 that map the following areas: Walking, personal care, daily activities, pain/discomfort 49 and anxiety/depression, as well as a 100 mm VAS scale (Visual Analogue Scale).35 50 51 52 In addition the patients will be asked to fill in a Back pain diary. This refers to a small printed 53 matter where the patient completes a daily questionnaire about the condition, as an aid to 54 remember the results between each registration in the electronic questionnaire. It contains 55 question about the pain level, function, use of medications, side effects and other comments 56 every day for 2 weeks and 4 and 12 weeks after treatment. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 15 BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 The following questions will be included at the baseline for describing the baseline 4 characteristics of the included sample and for the evaluation of potential prognostic factors for 5 recovery: 6 Sociodemographic variables: This refers to age, gender, marital status, education, smoking 7 status, use of alcohol, height, weight and serious life events during the 12 last months. 8 Örebro screening form for musculoskeletal pain: These are 25 questions that map all the 9 main "yellow flags". This questionnaire about the pain and how it influences job-related 10 36 37 11 activities shows prognostic factors for acute LBP. 12 Subjective Health Complaints (SHC): Here, the patient indicates in what degree he/she has 38 13 had 29 different health complaints the last 30 days and how long the complaints have lasted. 14 15 Patients will Forfill in the questionnaire peer at baselinereview (before the first only treatment) and then 16 immediately after the first treatment at the GPs office. Patients will fill in the Back Pain Diary 17 at home every day for 2 weeks and after 4 and 12 weeks. They may choose to fill in the 18 results directly in the electronic questionnaire. If the results are not registered electronically, 19 20 the patients may have to send the hard copy version of the diary to the project leader after 12 21 weeks, which is the maximum follow-up time. 22 23 Sample size: 24 We have calculated that a total of 270 patients will be needed for the study. Based on previous 25 studies and clinical experience, we estimate that the 3 days’ difference in median time to 26 recovery between the acupuncture group and control group is clinically significant, with a 27 median time to recovery of 7 days in the intervention group.4 39 40 The probability is 80 % that 28 the study will detect a treatment difference at a two sided 5 % significance level, if the true 29 hazard ratio is 1.429. This is based on the assumption that the accrual period will be 0 days, 30 41 31 the follow-up period will be 84 days and the median survival is 7 days. In addition, we 32 calculate with up to 10% dropout in the study.

33 http://bmjopen.bmj.com/ 34 Pilot study: 35 Prior to the main study, we plan to conduct a pilot study to test the study design, the 36 assessment and reporting tools with a total of 8 patients, 4 in each group. 37 38 Randomisation 39 40 Randomisation will be performed by a web-based randomisation system developed and 41 administered by the Unit of Applied Clinical Research, Institute of Cancer Research and on September 29, 2021 by guest. Protected copyright. 42 Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway, 43 and it will use block randomisation with various sizes of the blocks. 44 Each patient is randomised after the inclusion and exclusion criteria are considered. It is the 45 GP's medical secretary who checks the randomisation online, and the patient is then given an 46 appointment with either the “control-GP” or the “acupuncture-GP”. The patient is not told 47 which group he/she is randomised to or the name of the GP before the consent form and the 48 first questionnaire are filled in. 49 50 51 Analyses 52 Data will be analysed by an outcome assessor who is blinded to group status. The groups will 53 be analysed only as group 1 and group 2, and the results presented in Tables and Figures will 54 be worked out before the codes are broken. The primary analyses will be by intention-to-treat 55 and we will restrict the number of analyses in order to reduce the possibility of Type I errors. 56 For primary outcomes, a P value of <0.05 will be considered statistically significant. For the 57 secondary outcomes a P value of <0.01 will be considered significant. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 4 Primary Outcome Analyses: 5 We will assess the difference in survival curves (days to recovery) for the 2 groups using the 6 log-rank statistic. The median days to recovery will be used to express the time to recovery 7 for the two groups. Cox regression will be used to assess the effect of treatment group on 8 hazard ratios after allowing for the days of pain duration as baseline covariate. 9 10 11 Secondary Outcome Analyses: 12 Differences between the groups will be presented as mean with 95% CI or in categories with 13 odds ratio for categorical data. A mixed model with group as a fixed factor will be used for 14 the other outcome measures. Post-hoc analyses will be conducted if there is a significant 15 difference betweenFor treatment peer groups. We review will also test for potential only confounding factors in 16 these models. Analyses of prognostic factors will be carried out by a multivariate regression 17 models. 18 19 20 ETHICS AND DISSEMINATION: 21 When the patient contacts the GPs office, he/she will be informed about the objectives of the 22 study and asked if he/she is willing to participate. The patients will be told that their 23 participation is voluntary and that they will be granted full anonymity. Written consent will be 24 required from the included patients. 25 All patients will receive the treatment they normally would receive from their GP, but half of 26 them will undergo acupuncture treatment in addition to this. Given the randomisation, some 27 patients will not meet their usual GP. Instead, another GP will consult the patient, though in 28 the same office. 29 42-44 30 The risk of side effects of the acupuncture treatment has been found to be low. We will 31 examine any side effects of the treatment during the trial. 32 The trial is registered in the ClinicalTrials.gov-register, NCT01439412.

33 We plan to apply to the Regional Ethics Committee of South-Eastern Norway when the http://bmjopen.bmj.com/ 34 protocol is completed and published. 35 36 Publication policy: 37 The results of the trial will be published in appropriate journals regardless of outcome. The 38 39 trial will be implemented and reported in accordance with the recommendations of 40 CONSORT and STRICTA. 41 on September 29, 2021 by guest. Protected copyright. 42 DISCUSSION 43 This paper presents the design and rationale for a randomised, controlled multicenter study 44 examining the effects of acupuncture on the recovery of patients with acute LBP. The project 45 will increase the knowledge about the effects of acupuncture treatment for acute LBP, a 46 47 common disorder seen by general practitioners entailing high costs for the patient and society. 48 For the individual, a faster pain relief will aid an earlier return to normal, everyday activities. 49 For the society, the effect may be that LBP patients will return earlier to their work. The 50 primary outcome is the median time in days for recovery from pain. The secondary outcomes 51 are rated global improvement, back specific functional status, sick leaves, medication, GP 52 visits and side effects. A pilot study will be conducted. In the present study, we will also 53 analyse possible prognostic factors for recovery and cost-effectiveness of acupuncture 54 treatment for LBP. 55 56 57 Collaboration: 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 15 BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 We have established an international collaboration agreement with Karolinska Institutet 4 (Sweden) and Keele University (United Kingdom). Both institutions have extensive 5 experience and knowledge within research in medical acupuncture and musculoskeletal 6 disorders. 7 8 Funding: 9 10 TS has received research grants from the Norwegian College of General Practitioners to plan 11 this trial. 12 13 Authors’ contributions: 14 TS and HS initiated the project. TS, HS, MB, MG, IS and AF all contributed to 15 conceptualisationFor and design peer of the study review and all revised this manuscript only critically for 16 important intellectual content. All authors read and approved the final manuscript. 17 18 Competing interests: 19 20 None 21 22 Acknowledgements: 23 Thanks to Magne Thoresen, Professor of Biostatistics, University of Oslo, for assistance with 24 the statistical power calculation. 25 26 27 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 on September 29, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 REFERENCES: 4 5 1. Carey TS, Garrett JM, Jackman A, et al. Recurrence and care seeking after acute back pain: results of a long- 6 term follow-up study. North Carolina Back Pain Project. Med Care 1999;37:157-64. 7 8 2. Laerum E, Brox JI, Storheim K, et al. National clinical guidelines. Low back pain with and without nerve root 9 affection. Oslo: Formi, 2007. 10 11 3. Pengel LH, Herbert RD, Maher CG, et al. Acute low back pain: systematic review of its prognosis. BMJ 12 2003;327:323. 13 14 4. Coste J, Delecoeuillerie G, Cohen de Lara A, et al. Clinical course and prognostic factors in acute low back 15 pain: an Forinception cohort peer study in primary carereview practice. BMJ 1994;308:577-80. only 16 17 5. Henschke N, Maher CG, Refshauge KM, et al. Prognosis in patients with recent onset low back pain in 18 Australian primary care: inception cohort study. BMJ 2008;337:a171. 19 20 6. Hunskår. [Textbook of general practice]. Oslo: Gyldendal Norsk Forlag AS, 2003:338-41. 21 22 7. Hopton AK, Curnoe S, Kanaan M, et al. Acupuncture in practice: mapping the providers, the patients and the 23 settings in a national cross-sectional survey. BMJ Open 2012;2:e000456. 24 25 8. Manheimer E, White A, Berman B, et al. Meta-analysis: acupuncture for low back pain. Ann Intern Med 26 2005;142:651-63. 27 28 9. Yuan J, Purepong N, Kerr DP, et al. Effectiveness of acupuncture for low back pain: a systematic review. 29 Spine (Phila Pa 1976) 2008;33:E887-900. 30 31 10. Furlan AD, van Tulder MW, Cherkin DC, et al. Acupuncture and dry-needling for low back pain. Cochrane 32 Database Syst Rev 2011:CD001351.

33 http://bmjopen.bmj.com/ 34 11. Furlan AD, Yazdi F, Tsertsvadze A, et al. A systematic review and meta-analysis of efficacy, cost- 35 effectiveness, and safety of selected complementary and alternative medicine for neck and low-back pain. 36 Evid Based Complement Alternat Med 2012;2012:953139. 37 38 12. He RY. Clinical Observation on Treatment of Lumbago due to Cold-Dampness By Warm-Acupuncture Plus 39 Chinese Medicine. Chinese Acupuncture & Moxibustion 1997;17:279-80. 40 41

13. Araki S, Kawamura O, Mataka T, et al. Randomized controlled trial comparing the effect of manual on September 29, 2021 by guest. Protected copyright. 42 acupuncture with sham acupuncture for acute low back pain Journal of the Japan Society of Acupuncture 43 and Moxibustion 2001;51:382. 44 45 14. Kittang G, Melvaer T, Baerheim A. [Acupuncture contra antiphlogistics in acute lumbago]. Tidsskr Nor 46 Laegeforen 2001;121:1207-10. 47 48 15. van Tulder MW, Scholten RJ, Koes BW, et al. Non-steroidal anti-inflammatory drugs for low back pain. 49 Cochrane Database Syst Rev 2000:CD000396. 50 51 16. Kennedy S, Baxter GD, Kerr DP, et al. Acupuncture for acute non-specific low back pain: a pilot randomised 52 non-penetrating sham controlled trial. Complement Ther Med 2008;16:139-46. 53 54 17. Lund I, Naslund J, Lundeberg T. Minimal acupuncture is not a valid placebo control in randomised controlled 55 trials of acupuncture: a physiologist's perspective. Chin Med 2009;4:1. 56 57 18. Lundeberg T, Lund I, Naslund J, et al. The Emperors sham - wrong assumption that sham needling is sham. 58 Acupunct Med 2008;26:239-42. 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 15 BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 19. Linde K, Niemann K, Schneider A, et al. How large are the nonspecific effects of acupuncture? A meta- 4 analysis of randomized controlled trials. BMC Med 2010;8:75. 5 6 20. Vas J, Perea-Milla E, Mendez C, et al. Efficacy and safety of acupuncture for the treatment of non-specific 7 acute low back pain: a randomised controlled multicentre trial protocol [ISRCTN65814467]. BMC 8 Complement Altern Med 2006;6:14. 9 10 21. Shin JS, Ha IH, Lee TG, et al. Motion style acupuncture treatment (MSAT) for acute low back pain with 11 severe disability: a multicenter, randomized, controlled trial protocol. BMC Complement Altern Med 12 2011;11:127. 13 14 22. Deadman P A-KM, Baker K. A Manual of Acupuncture. 3 ed ed: Journal of Chinese Medicine, 2008:574-80. 15 For peer review only 16 23. Heyerdahl O, Lystad N. [Textbook of Acupuncture]. Oslo: Universitetsforlaget, 2007:291-303. 17 18 24. White A, Cummings M., Filshie J. An introduction to western medical acupuncture. London: Churchill 19 Livingstone Elsevier, 2008:7-76. 20 21 25. Witt CM, Jena S, Selim D, et al. Pragmatic randomized trial evaluating the clinical and economic effectiveness 22 of acupuncture for chronic low back pain. Am J Epidemiol 2006;164:487-96. 23 24 26. Kalauokalani D, Cherkin DC, Sherman KJ, et al. Lessons from a trial of acupuncture and massage for low 25 back pain: patient expectations and treatment effects. Spine (Phila Pa 1976) 2001;26:1418-24. 26 27 27. Linde K, Witt CM, Streng A, et al. The impact of patient expectations on outcomes in four randomized 28 controlled trials of acupuncture in patients with chronic pain. Pain 2007;128:264-71. 29 30 28. Thomas KJ, MacPherson H, Thorpe L, et al. Randomised controlled trial of a short course of traditional 31 acupuncture compared with usual care for persistent non-specific low back pain. BMJ 2006;333:623. 32

33 29. Lin CW, Haas M, Maher CG, et al. Cost-effectiveness of guideline-endorsed treatments for low back pain: a http://bmjopen.bmj.com/ 34 systematic review. Eur Spine J 2011;20:1024-38. 35 36 30. Skjeie H, Skonnord T, Fetveit A, et al. A pilot study of ST36 acupuncture for infantile colic. Acupunct Med 37 2011;29:103-7. 38 39 31. ClinicalTrials. Effects of Acupuncture in the Treatment of Infant Colic (NCT00907621). 2009. 40 41 32. Mannion AF, Elfering A, Staerkle R, et al. Outcome assessment in low back pain: how low can you go? Eur on September 29, 2021 by guest. Protected copyright. 42 Spine J 2005;14:1014-26. 43 44 33. Mannion AF, Porchet F, Kleinstuck FS, et al. The quality of spine surgery from the patient's perspective. Part 45 1: the Core Outcome Measures Index in clinical practice. Eur Spine J 2009;18 Suppl 3:367-73. 46 47 34. Roland M, Morris R. A study of the natural history of back pain. Part I: development of a reliable and sensitive 48 measure of disability in low-back pain. Spine (Phila Pa 1976) 1983;8:141-4. 49 50 35. Solberg TK, Olsen JA, Ingebrigtsen T, et al. Health-related quality of life assessment by the EuroQol-5D can 51 provide cost-utility data in the field of low-back surgery. Eur Spine J 2005;14:1000-7. 52 53 36. Linton SJ, Hallden K. Can we screen for problematic back pain? A screening questionnaire for predicting 54 outcome in acute and subacute back pain. Clin J Pain 1998;14:209-15. 55 56 37. Grotle M, Vollestad NK, Brox JI. Screening for yellow flags in first-time acute low back pain: reliability and 57 validity of a Norwegian version of the Acute Low Back Pain Screening Questionnaire. Clin J Pain 58 2006;22:458-67. 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 38. Grovle L, Haugen AJ, Ihlebaek CM, et al. Comorbid subjective health complaints in patients with sciatica: a 4 prospective study including comparison with the general population. J Psychosom Res 2011;70:548-56. 5 6 39. Curtis P, Carey TS, Evans P, et al. Training primary care physicians to give limited manual therapy for low 7 back pain: patient outcomes. Spine (Phila Pa 1976) 2000;25:2954-60; discussion 60-1. 8 9 40. Hancock MJ, Maher CG, Latimer J, et al. Assessment of diclofenac or spinal manipulative therapy, or both, in 10 addition to recommended first-line treatment for acute low back pain: a randomised controlled trial. Lancet 11 2007;370:1638-43. 12 13 41. Schoenfeld D. Find statistical considerations for a study where the outcome is a time to failure 14 (http://hedwig.mgh.harvard.edu/sample_size/time_to_event/para_time.html). 2001. 15 For peer review only 16 42. Vincent C. The safety of acupuncture. BMJ 2001;323:467-8. 17 18 43. White A. A cumulative review of the range and incidence of significant adverse events associated with 19 acupuncture. Acupunct Med 2004;22:122-33. 20 21 44. Witt CM, Pach D, Brinkhaus B, et al. Safety of acupuncture: results of a prospective observational study with 22 229,230 patients and introduction of a medical information and consent form. Forsch Komplementmed 23 2009;16:91-7. 24 25 26 27 28 29 30 31 32

ACUPUNCTURE FOR ACUTE NON-SPECIFIC LOW BACK PAIN: A PROTOCOL FOR A RANDOMISED, CONTROLLED, MULTICENTER INTERVENTION STUDY IN GENERAL PRACTICE – THE ACUBACK STUDY.

For peer review only

Journal: BMJ Open

Manuscript ID: bmjopen-2012-001164.R1

Article Type: Protocol

Date Submitted by the Author: 20-May-2012

Primary Subject General practice / Family practice Heading:

Complementary medicine, Rehabilitation medicine, Pharmacology and http://bmjopen.bmj.com/ Secondary Subject Heading: therapeutics

COMPLEMENTARY MEDICINE, Back pain < ORTHOPAEDIC & TRAUMA Keywords: SURGERY, PAIN MANAGEMENT, PRIMARY CARE, Clinical trials < THERAPEUTICS, Adverse events < THERAPEUTICS

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 on September 29, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 15 BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 ABSTRACT: 4 Introduction: 5 Some general practitioners (GPs) treat acute low back pain (LBP) with acupuncture, despite 6 lacking evidence of its effectiveness for this condition. 7 8 The aim of this study is to evaluate whether a single treatment-session with acupuncture can 9 reduce time to recovery when applied in addition to standard LBP-treatment according to the 10 Norwegian national guidelines. Analyses of prognostic factors for recovery and cost- 11 effectiveness will also be carried out. 12 Methods and analysis: 13 In this randomised, controlled, multicenter study in general practice in Southern Norway, 270 14 patients will be allocated into one of two treatment-groups, using a web-based application 15 based on blockFor randomisation. peer Outcome reviewassessor will be blinded only for group allocation of the 16 patients. 17 18 The control group will receive standard treatment, while the intervention group will receive 19 standard treatment plus acupuncture treatment. There will be different GPs treating the two 20 groups, and both groups will just have one consultation. Adults who consult their GP because 21 of acute LBP will be included. Patients with nerve root affection, "red flags", pregnancy, 22 previous sick leave more than 14 days and disability pension will be excluded. The primary 23 outcome of the study is the median time to recovery (days). The secondary outcomes are rated 24 global improvement, back-specific functional status, sick leave, medication, GP visits and 25 side effects. A pilot study will be conducted. 26 Ethics and dissemination: 27 28 Participation is based on informed written consent. We will apply for an ethical approval from 29 the Regional Committee for Medical and Health Research Ethics (REC), when the study- 30 protocol is completedpublished. Results from this study, positive or negative, will be 31 disseminated in scientific medical journals. 32 Discussion:

33 This study aims to explore whether acupuncture is a useful clinical tool for treating acute LBP http://bmjopen.bmj.com/ 34 in general practice. 35 Trial Registration: 36 ClinicalTrials.gov Identifier: NCT01439412. 37 38 39 40 41 on September 29, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 4 INTRODUCTION: 5 Low back pain (LBP) is a very common disorder with consequences for the individual patient 6 as well as for the society. Up to 80% of the population experiences back pain at least once in 7 their lifetime, about 50% during the previous year. Point prevalence is 15%, and the condition 8 1 9 relapses frequently, 40% within 6 months. Back pain is the medical condition that ranks highest in terms of Norwegian socio-economic expenses.2 Most people with acute LBP 10 3 11 experience improvements in pain and disability within a month, but the median time to 4 5 12 recovery recorded in studies on back pain varies widely, from 7 days to 58 days. 13 Acute LBP is treated primarily in the primary healthcare by general practitioners (GPs), 14 physiotherapists, manual therapists and chiropractors. Recommended treatment according to 15 clinical guidelinesFor contains peer information aboutreview the condition, advice only to stay active and, if 16 possible, avoid bed rest, early and gradual mobilisation after the acute phase, pain treatment 17 with paracetamol and/or non-steroidal anti-inflammatory drugs (NSAIDs) with time- 18 2 6 19 contingent doses. 20 GPs educated and trained in acupuncture also use acupuncture for the treatment of both acute 7 21 and chronic LBP cases. There is evidence that acupuncture is effective in chronic LBP, and 2 8- 22 such treatment is therefore recommended in the "National guidelines for LBP" in Norway. 23 10 Reviews and meta-analyses conclude that the documentation of acupuncture treatment for 24 acute LBP is limited by few and poorly conducted studies, however.10 11 In a systematic 25 review conducted by Yuan et al, the authors concluded that for non-specific LBP, treatment 26 regimens of acupuncture differ by the types of reference sources, in terms of treatment 27 frequency, the points chosen, number of points needled per session, duration and sessions, 28 12 29 and co-interventions. 30 31 In 1997, He presented a Chinese study with 100 patients afflicted with LBP (5 days to 6 32 months duration) randomised to either manual acupuncture with moxibustion plus Chinese 13

33 herbal medicine or Chinese herbal medicine alone. A later Cochrane review concludes that http://bmjopen.bmj.com/ 34 this trial was of low methodological quality and showed limited evidence to back the fact that 35 the combined treatment was more effective for a global measure of pain and function in the 36 long-term follow-up.10 37 38 39 Araki et al have published an abstract of a trial where 40 patients with acute LBP (< 3 days) 40 were randomised into two groups where the intervention group got acupuncture treatment in 41 the acupuncture point SI3 bilaterally and then performed back exercises. The control group, on September 29, 2021 by guest. Protected copyright. 42 however, received sham acupuncture treatment with mimicked needle insertion, after which 43 they were asked to perform back exercises. Araki et al found no difference between the 44 effects of acupuncture and that of the sham acupuncture.14 45 46 One of the studies referred to in the Cochrane review is a Norwegian study conducted by 47 Kittang et al.15 They found that acupuncture was as effective as medical treatment with 48 49 Naproxen per os in relation to pain and stiffness, but that the “Naproxen-group” had more 50 side effects and greater recurrences in the observation period. While Kittang et al conclude 51 that acupuncture is effective, this result is based on other studies showing the efficacy of 16 52 NSAIDs for acute LBP. 53 54 Kennedy et al published a pilot study in 2008, which demonstrated the feasibility of a 55 randomised, controlled trial regarding the penetrating needle acupuncture when compared to 56 non-penetrating sham acupuncture for the treatment of acute LBP in primary care. However, 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 15 BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 the study lacked sufficient power to draw any conclusion on treatment effects.The study did 4 not reveal any conclusions of the effects of acupuncture.17 5 6 Sham acupuncture involves inserting penetrating or non-penetrating needles in points that are 7 not classical acupuncture points and/or are not located in the same segment. Several studies 8 conclude, however, that sham acupuncture is not a valid placebo control, and this is explained 9 18 19 10 by neurophysiologic effects of the sham treatment. Trials with sham acupuncture need to 11 be very comprehensive in order to demonstrate differences between the effects of real 20 12 acupuncture and the sham treatment. 13 14 In 2006, Vas et al published a study protocol of a four-branch randomised controlled trial, 15 which intendedFor to obtain peerfurther evidence review on the effectiveness ofonly acupuncture on acute LBP 16 and isolate the specific and non-specific effects of the treatment.21 The results have not yet 17 been published. 18 Shin et al are planning to perform a randomised, controlled trial with two arms comparing a 19 22 20 motion-style acupuncture with an NSAID injection. 21 22 However, we have not yet found evidence favouring acupuncture as a treatment for acute 23 LBP. Our clinical experience is that an acupuncture treatment of the distal and local 24 acupuncture points combined with small mobilising movements means faster recovery of the 25 pain. This treatment is according to some textbooks on acupuncture.23 24 None of the 26 mentioned protocols though examine this kind of acupuncture. All the points that we plan to 27 use are mentioned among the common points used for LBP in the systematic review by Yuan 28 et al, and mobilising movements are often used as a co-intervention for acute LBP.12 29 30 31 Strong stimulation of the distal acupuncture points is thought to act through activation of the 32 extra-segmental pain inhibition, "Diffuse noxious inhibitory Controls” (DNIC) and the release

33 of endorphins and possibly, other endogenous compounds in the central nervous system. http://bmjopen.bmj.com/ 34 Local acupuncture therapy immediately following the mobilisation movements may provide 35 additional benefits through segmental pain-inhibition along with the inhibition of myofascial 36 trigger points.25 37 38 Clear prognostic factors for the efficacy of acupuncture for chronic LBP have, however, not 39 40 been documented. In a large German study, the authors found that lower age, lower baseline 41 spinal function and more than 10 years of education indicated a better response to 26 on September 29, 2021 by guest. Protected copyright. 42 acupuncture. Some researchers still find that optimistic treatment expectations improve the 43 effects of acupuncture,27 28 while others find the best treatment effects in patients who are 44 neutral to acupuncture as a treatment method.29 There are no similar studies for acute LBP. 45 46 A systematic review found evidence supporting the cost-effectiveness of the guideline- 47 endorsed treatments of acupuncture for sub-acute or chronic LBP.30 Apart from the few trials 48 of acupuncture treatment for acute LBP, there are no studies examining the cost-effectiveness 49 of acupuncture for acute LBP. 50 51 52 In the present study, we aim at exploring whether a single treatment-session with acupuncture 53 can reduce the time to recovery when applied in addition to the standard LBP-treatment in 54 general practice according to Norwegian national guidelines. We will not use sham 55 acupuncture as a control treatment. Further, we will not include a pure placebo group or a 56 “waiting list group”, because this will mean not giving treatment to patients with severe pain, 57 which we consider unethical. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 Secondary aims are pain intensity, disability, sick leave and drug use, and in addition we will 4 evaluate the prognostic factors for recovery, and if possible, identify any factors that 5 characterize those who have beneficial effects of the acupuncture treatment. 6 A systematic review found evidence supporting the cost-effectiveness of the guideline- 7 endorsed treatments of acupuncture for sub-acute or chronic LBP.30 Apart from the few trials 8 of acupuncture treatment for acute LBP, there are no studies examining the cost-effectiveness 9 10 of acupuncture for acute LBP. ThereforeFinally, we also aim at carrying out a cost- 11 effectiveness analysis. 12 13 Our hypotheses are: 14 1. Acupuncture treatment contributes to faster pain-recovery in acute LBP compared to 15 standardFor treatment peer in general practice review provided in accordance only with the Norwegian 16 national guidelines. 17 2. Acupuncture treatment for acute LBP improves function, and reduces drug use and 18 sick leave, compared to the standard treatment in general practice provided in 19 20 accordance with national guidelines. 21 3. Acupuncture treatment for acute LBP is a cost-effective treatment in general practice. 22 23 METHODS AND ANALYSIS: 24 This is a randomised, controlled, multicenter study, which will involve 15 GP group practices 25 located in different areas in southern Norway, both cities and rural municipalities. The 15 GPs 26 administering the acupuncture have received their training through the education program of 27 the Norwegian Society of Medical Acupuncture and are members of the Network Group of 28 29 Medical Acupuncture, under the Norwegian College of General Practice. They are specialists 30 in general practice and many of them are experienced participants in randomised clinical 31 31 trials, given a parallel ongoing trial examining the effects of acupuncture on infantile colic. 32 32 15 These GPs are located in different areas in southern Norway, both cities and rural

33 municipalities. GPs in the same practices, but without training in acupuncture, however, need http://bmjopen.bmj.com/ 34 to bewill be recruited to treat the control group. 35 A medical secretary at every doctor's office will carry out the telephonic interview in which, 36 the patients will be informed about the study, the inclusion and exclusion criteria will be 37 checked, the randomisation of the patients will be performed and the distributing and 38 39 collecting questionnaires for the treatment day. A two-day workshop before the start of the 40 main study will be arranged for all the participating doctors and medical secretaries. The 41 workshop will contribute to the standardization of information, reporting and patient on September 29, 2021 by guest. Protected copyright. 42 management in general. A website with information about the study will be created. 43 44 We plan to recruit patients with acute LBP and randomly allocate them into two parallell 45 study-groups (A and C) of equal sizes. Patients in both groups will receive standard treatment 46 for LBP in accordance with the Norwegian national guidelines.2 One group will receive only 47 the standard treatment, while the other group also will receive additional acupuncture 48 49 treatment as described below. To control for potential attention bias, we will measure the time 50 spent on each patient in the intervention and control groups. Outcome assessor will be blinded 51 for group allocation of the patients. 52 53 Inclusion criteria: 54 - Adults (20-55 years) who contact their GP’s office because of acute nonspecific LBP 55 (0-14 days). 56 57 Exclusion criteria: 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 15 BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 - Nerve root affection and/or radiating pain below the knee. 4 - LBP with suspected “red flags”, i.e. infections, tumours and metastatic disease, 5 rheumatic disease, fractures and significant deformities of the spine. 6 - LBP which starts in pregnancy. 7 - Physician-reported sick leave of 14 days or more during the month before the 8 commencement of the back pain, for any reason. 9 10 - Disability pension. 11 12 Control group C: Standard treatment in general practice is provided in accordance with the 13 Norwegian national guidelines, i.e. general advice about activity, prescription of pain 14 relievers (paracetamol), and sick leave, if needed. 15 For peer review only 16 Acupuncture group A: This group will receive standard treatment as the control group and 17 in addition acupuncture treatment. 18 Acupuncture: The patient sits in a chair and the doctor stimulates the acupuncture points “the 19 20 Lumbar Pain Points” (Yaotongxue/Yaotongdian) on the right hand, with acupuncture needles 21 of type Seirin B - 8a 0.30 x 30 mm. These are two points located between the second and 22 third and the fourth and fifth metacarpal bones, immediately distal to the bases of the 23 metacarpals. Insertion depth is 10-15 mm. The doctor stimulates the needles in a rotating up- 24 and down movement to impart a powerful needle sensation (called “de Qi”), and this is 25 repeated in short sequences to maintain the needle sensation for a total of 1 min. The needles 26 shall stay in the hand during the rest of the treatment. The patient is then asked to rise and to 27 perform slow rotating pelvic movements for 2 minutes, before lying down on a bench to be 28 treated in the local points Huatuojiaji ("Jiaji") with acupuncture needles of the type SEIRIN J- 29 30 8 with sleeve 0.30 x 50 mm. These points are located 1.5 cm lateral to the depressions below 31 the spinous processes, and we will acupuncture them bilaterally in the segments of the L2-L4 32 (6 needles) at a depth of 3-4 cm. They are stimulated manually until the patient experiences

33 the needle sensation. Then, the patient lies quietly on the bench for another 5 minutes before http://bmjopen.bmj.com/ 34 all the needles are removed. 35 The whole acupuncture session lasts for a total of 8 minutes. Thus part of the advice and 36 prescription part is done while the patient is receiving acupuncture treatment, in order to let 37 the consultation time for the two groups be as equal as possible. The consultation time will be 38 measured and registered by the GPs. The duration of the acupuncture treatment in this trial is 39 12 40 shorter than in earlier trials, and we will only have one treatment session. This is to reduce 41 potential attention bias. on September 29, 2021 by guest. Protected copyright. 42 43 Choice of treatment: 44 The described treatment with the specific distal and local acupuncture points has been chosen 45 after a three-step process. We started with the treatment we used ourselves in clinical practice, 46 compared it with the literature and standardized it for this trial. Then we asked several 47 experienced acupuncture doctors how they treated acute LBP and what they thought about our 48 suggestion of standardization. 49 50 Based on this feedback, we asked an expert group of physicians and physiotherapists 51 experienced in administering acupuncture about which acupuncture points they preferred in 52 the treatment of acute LBP. The end result was considered the treatment according to best 53 practice for this study. 54 55 Patient flow: 56 When a patient with acute LBP contacts the GP’s office, the medical secretary provides 57 information about the trial and asks whether the patient would want to participate in the study. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 The patient will be informed that he/she will meet a GP and get the usual treatment for LBP, 4 and may or may not get acupuncture in addition to the usual treatment. During the observation 5 period of 12 months, patients are asked for not to receive any additional acupuncture 6 treatments. 7 If the patient consents, the medical secretary asks the patient questions regarding inclusion- 8 and exclusion criteria. The inclusion and exclusion criteria are checked both in the telephone 9 10 interview by the secretary and in the consultation by the GP. All contacts are counted and it is 11 recorded at what level and for what reason exclusions are done. If the patient is eligible, the 12 medical secretary randomises the patient to one of the two groups through the Web-based 13 application at the Unit for Applied Clinical Research in Trondheim 14 (http://www.ntnu.no/dmf/akf). This is used to register the patient in the appropriate doctor's 15 timetable, butFor the patient peerdoes not know whichreview group or which doctoronly he/she is going to meet. 16 At the GP’s office, the patient is required to fill in the consent form and the first questionnaire 17 (T0A). This is delivered to the medical secretary prior to the consultation. 18 19 20 Patients who are randomised to the control group will receive one standard treatment 21 provided by a GP without acupuncture training. The consultation will include history, 22 examination, information, prescription and advices, according to national guidelines for LBP. 23 If the doctor discovers exclusion criteria, the patient will be excluded from the trial. 24 25 Patients who are randomised to the acupuncture group will receive one single acupuncture 26 treatment in addition to the standard treatment as described for the control group. The 27 consultation will be provided by a GP who is trained in acupuncture. If the patient’s condition 28 requires further consultations in the follow-up time, these will be recorded. These 29 30 consultations, however, will not include additional acupuncture treatment. 31 32 When the patient has finished the consultation, they will fill in a post-treatment questionnaire

33 (T0B) and submit it to the secretary in an enclosed envelope. The patient then will receive a http://bmjopen.bmj.com/ 34 back pain diary, which he/she will be required to fill in at home at the given times. The diary 35 also contains information about how the patient can choose to give the answers in an 36 electronic questionnaire (Figure 1). 37 38 Measurements: 39 For evaluation of the effects of treatment, we will use standardised instruments that have been 40 33 34 41 validated, both internationally and nationally. The outcome measures will be filled in at on September 29, 2021 by guest. Protected copyright. 42 baseline and after 1, 2, 4, and 12 weeks and 12 months, and the following outcome 43 measurements will be included: 44 45 The primary outcome is the median time in days to the recovery of pain, measured on the first 46 day the patient scores 0 or 1 point on the Numerical Rating Scale (NRS). Clinically relevant 47 differences between the groups are considered to be minimum 3 days. 48 49 Secondary outcome measures will be: 50 51 1. Pain intensity is assessed by the numerical rating scale (NRS) before and immediately 52 after treatment and at the other follow-up times. The patient indicates his perceived 53 pain intensity on an 11-point scale from 0 to 10 with endpoints indicating "no pain" 54 and "worst imaginable pain". Based on other studies, the clinically relevant 55 improvement is estimated to as being minimum 2-31.5 points in a previous Norwegian 56 study on patients with acute LBP.for individual patients and 1 point as the difference 57 between the groups.35 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 15 BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 2. Back specific functional status by the Roland Morris Disability Questionnaire (RDQ). 4 This measures patients' perceptions of function.36 The patient answers yes or no 5 concerning 24 allegations about the activities and conditions, depending on whether 6 they feel that the statement describes them on this day. 7 3. Sick leaves, the number of days away from work due to back pain. 8 4. Global measure of improvement (Likert improvement assessment scale). This assesses 9 10 the patients’ perceptions of change, stated in whole numbers from 1 = much better to 5 11 = much worse. 12 5. Use of medication (paracetamol, eventually others), counting of daily consumption. 13 6. Number of new visits at the GP’s office. 14 7. Side effects of treatment (acupuncture and medications). 15 8. Health-relatedFor quality-of-life peer by thereview EuroQoL (EQ-5D). onlyThis enumerates 5 questions 16 that map the following areas: Walking, personal care, daily activities, pain/discomfort 17 and anxiety/depression, as well as a 100 mm VAS scale (Visual Analogue Scale).37 18 19 20 In addition the patients will be asked to fill in a Back pain diary. This refers to a small printed 21 matter where the patient completes a daily questionnaire about the condition, as an aid to 22 remember the results between each registration in the electronic questionnaire. It contains 23 question about the pain level, function, use of medications, side effects and other comments 24 every day for 2 weeks and 4, and 12 weeks and 12 months after treatment. 25 26 The following questions will be included at the baseline for describing the baseline 27 characteristics of the included sample and for the evaluation of potential prognostic factors for 28 recovery: 29 30 Sociodemographic variables: This refers to age, gender, marital status, education, smoking 31 status, use of alcohol, height, weight and serious life events during the 12 last months. 32 In addition we will ask the patients about their preferences for the two treatment options for

33 LBP and if they believe that acupuncture will contribute to faster recovery than the usual http://bmjopen.bmj.com/ 34 treatment. 35 Örebro screening form for musculoskeletal pain: These are 25 questions that map all the 36 main "yellow flags". This questionnaire about the pain and how it influences job-related 37 activities shows prognostic factors for acute LBP.38 39 38 Subjective Health Complaints (SHC): Here, the patient indicates in what degree he/she has 39 40 40 had 29 different health complaints the last 30 days and how long the complaints have lasted. 41 on September 29, 2021 by guest. Protected copyright. 42 Patients will fill in the questionnaire at baseline (before the first treatment) and then 43 immediately after the first treatment at the GPs office. Patients will fill in the Back Pain Diary 44 at home every day for 2 weeks, and after 4 and 12 weeks and after 12 months. They may 45 choose to fill in the results directly in the electronic questionnaire. If the results are not 46 registered electronically, the patients may have to send the hard copy version of the diary to 47 the project leader after 12 weeks. The last questionnaire has to be sent after 12 months,, which 48 is the maximum follow-up time. 49 50 Patients will receive a SMS reminder about the registration. If the patient does not respond to 51 the electronic questionnaire or the Back pain diary, he/she will be reminded by SMS or 52 telephone twice. Those who withdraw will be asked for the reason for doing so. 53 54 Sample size: 55 We have calculated that a total of 270 patients will be needed for the study. Based on previous 56 studies and clinical experience, we estimate that the 3 days’ difference in median time to 57 recovery between the acupuncture group and control group is clinically significant, with a 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 4 41 42 3 median time to recovery of 7 days in the intervention group. The probability is 80 % that 4 the study will detect a treatment difference at a two sided 5 % significance level, if the true 5 hazard ratio is 1.429. This is based on the assumption that the accrual period will be 0 days, 6 the follow-up period will be 84 365 days and the median survival is 7 days.43 In addition, we 7 calculate with up to 10% dropout in the study. 8 9 10 Pilot study: 11 Prior to the main study, we plan to conduct a pilot study to test the study design, the 12 assessment and reporting tools with a total of 8 patients, 4 in each group. 13 14 Randomisation 15 RandomisationFor will be performedpeer by a web-basedreview randomisation only system developed and 16 administered by the Unit of Applied Clinical Research, Institute of Cancer Research and 17 Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway, 18 and it will use block randomisation with various sizes of the blocks. 19 20 Each patient is randomised after the inclusion and exclusion criteria are considered. It is the 21 GP's medical secretary who checks the randomisation online, and the patient is then given an 22 appointment with either the “control-GP” or the “acupuncture-GP”. The patient is not told 23 which group he/she is randomised to or the name of the GP before the consent form and the 24 first questionnaire are filled in. 25 26 Analyses 27 Data will be analysed by an outcome assessor who is blinded to group status. The groups will 28 be analysed only as group 1 and group 2, and the results presented in Tables and Figures will 29 30 be worked out before the codes are broken. The primary analyses will be by intention-to-treat 31 and we will restrict the number of analyses in order to reduce the possibility of Type I errors. 32 For primary outcomes, a P value of <0.05 will be considered statistically significant. For the

33 secondary outcomes a P value of <0.01 will be considered significant. http://bmjopen.bmj.com/ 34 35 Primary Outcome Analyses: 36 We will assess the difference in survival curves (days to recovery) for the 2 groups using the 37 log-rank statistic. The median days to recovery will be used to express the time to recovery 38 for the two groups. Cox regression will be used to assess the effect of treatment group on 39 40 hazard ratios after allowing for the days of pain duration as baseline covariate. 41 on September 29, 2021 by guest. Protected copyright. 42 Secondary Outcome Analyses: 43 Differences between the groups will be presented as mean with 95% CI or in categories with 44 odds ratio for categorical data. A mixed model with group as a fixed factor will be used for 45 the other outcome measures. Post-hoc analyses will be conducted if there is a significant 46 difference between treatment groups. We will also test for potential confounding factors in 47 these models. Analyses of prognostic factors will be carried out by a multivariate regression 48 models. 49 50 51 ETHICS AND DISSEMINATION: 52 When the patient contacts the GPs office, he/she will be informed about the objectives of the 53 study and asked if he/she is willing to participate. The patients will be told that their 54 participation is voluntary and that they will be granted full anonymity. Informed wWritten 55 consent will be required from the included patients. 56 All patients will receive the treatment they normally would receive from their GP, but half of 57 them will undergo acupuncture treatment in addition to this. Given the randomisation, some 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 15 BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 patients will not meet their usual GP. Instead, another GP will consult the patient, though in 4 the same office. 5 The risk of side effects of the acupuncture treatment has been found to be low.44-46 We will 6 examine any side effects of the treatment during the trial. 7 The trial is registered in the ClinicalTrials.gov-register, NCT01439412. 8 We plan to apply to the Regional Ethics Committee of South-Eastern Norway when the 9 10 protocol is completed and published. 11 12 Publication policy: 13 The results of the trial will be published in appropriate journals regardless of outcome. The 14 trial will be implemented and reported in accordance with the recommendations of 15 CONSORT andFor STRICTA. peer review only 16 17 ISCUSSION 18 D 19 This paper presents the design and rationale for a randomised, controlled multicenter study 20 examining the effects of acupuncture on the recovery of patients with acute LBP. The project 21 will increase the knowledge about the effects of acupuncture treatment for acute LBP, a 22 common disorder seen by general practitioners entailing high costs for the patient and society. 23 For the individual, a faster pain relief will aid an earlier return to normal, everyday activities. 24 For the society, the effect may be that LBP patients will return earlier to their work. The 25 primary outcome is the median time in days for recovery from pain. The secondary outcomes 26 are rated global improvement, back specific functional status, sick leaves, medication, GP 27 28 visits and side effects. A pilot study will be conducted. In the present study, we will also 29 analyse possible prognostic factors for recovery and cost-effectiveness of acupuncture 30 treatment for LBP. 31 32 Collaboration:

33 We have established an international collaboration agreement with Karolinska Institutet http://bmjopen.bmj.com/ 34 (Sweden) and Keele University (United Kingdom). Both institutions have extensive 35 experience and knowledge within research in medical acupuncture and musculoskeletal 36 disorders. 37 38 39 Funding: 40 TS has received research grants from the Norwegian College of General Practitioners to plan 41 this trial. The study funder will not have any impact on the trial and the publication of the on September 29, 2021 by guest. Protected copyright. 42 results. 43 44 Authors’ contributions: 45 TS and HS initiated the project. TS, HS, MB, MG, IS and AF all contributed to 46 conceptualisation and design of the study and all revised this manuscript critically for 47 48 important intellectual content. All authors read and approved the final manuscript. 49 50 Competing interests: 51 None 52 53 Acknowledgements: 54 Thanks to Magne Thoresen, Professor of Biostatistics, University of Oslo, for assistance with 55 the statistical power calculation. 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 REFERENCES: 4 5 1. Carey TS, Garrett JM, Jackman A, et al. Recurrence and care seeking after acute back pain: results of a long- 6 term follow-up study. North Carolina Back Pain Project. Med Care 1999;37:157-64. 7 8 2. Laerum E, Brox JI, Storheim K, et al. National clinical guidelines. Low back pain with and without nerve root 9 affection. Oslo: Formi, 2007. 10 11 3. Pengel LH, Herbert RD, Maher CG, et al. Acute low back pain: systematic review of its prognosis. BMJ 12 2003;327:323. 13 14 4. Coste J, Delecoeuillerie G, Cohen de Lara A, et al. Clinical course and prognostic factors in acute low back 15 pain: an Forinception cohort peer study in primary carereview practice. BMJ 1994;308:577-80. only 16 17 5. Henschke N, Maher CG, Refshauge KM, et al. Prognosis in patients with recent onset low back pain in 18 Australian primary care: inception cohort study. BMJ 2008;337:a171. 19 20 6. Hunskår. [Textbook of general practice]. Oslo: Gyldendal Norsk Forlag AS, 2003:338-41. 21 22 7. Hopton AK, Curnoe S, Kanaan M, et al. Acupuncture in practice: mapping the providers, the patients and the 23 settings in a national cross-sectional survey. BMJ Open 2012;2:e000456. 24 25 8. Manheimer E, White A, Berman B, et al. Meta-analysis: acupuncture for low back pain. Ann Intern Med 26 2005;142:651-63. 27 28 9. Yuan J, Purepong N, Kerr DP, et al. Effectiveness of acupuncture for low back pain: a systematic review. 29 Spine (Phila Pa 1976) 2008;33:E887-900. 30 31 10. Furlan AD, van Tulder MW, Cherkin DC, et al. Acupuncture and dry-needling for low back pain. Cochrane 32 Database Syst Rev 2011:CD001351.

33 http://bmjopen.bmj.com/ 34 11. Furlan AD, Yazdi F, Tsertsvadze A, et al. A systematic review and meta-analysis of efficacy, cost- 35 effectiveness, and safety of selected complementary and alternative medicine for neck and low-back pain. 36 Evid Based Complement Alternat Med 2012;2012:953139. 37 38 12. Yuan J, Kerr D, Park J, et al. Treatment regimens of acupuncture for low back pain--a systematic review. 39 Complement Ther Med 2008;16:295-304. 40 41

13. He RY. Clinical Observation on Treatment of Lumbago due to Cold-Dampness By Warm-Acupuncture Plus on September 29, 2021 by guest. Protected copyright. 42 Chinese Medicine. Chinese Acupuncture & Moxibustion 1997;17:279-80. 43 44 14. Araki S, Kawamura O, Mataka T, et al. Randomized controlled trial comparing the effect of manual 45 acupuncture with sham acupuncture for acute low back pain Journal of the Japan Society of Acupuncture 46 and Moxibustion 2001;51:382. 47 48 15. Kittang G, Melvaer T, Baerheim A. [Acupuncture contra antiphlogistics in acute lumbago]. Tidsskr Nor 49 Laegeforen 2001;121:1207-10. 50 51 16. van Tulder MW, Scholten RJ, Koes BW, et al. Non-steroidal anti-inflammatory drugs for low back pain. 52 Cochrane Database Syst Rev 2000:CD000396. 53 54 17. Kennedy S, Baxter GD, Kerr DP, et al. Acupuncture for acute non-specific low back pain: a pilot randomised 55 non-penetrating sham controlled trial. Complement Ther Med 2008;16:139-46. 56 57 18. Lund I, Naslund J, Lundeberg T. Minimal acupuncture is not a valid placebo control in randomised controlled 58 trials of acupuncture: a physiologist's perspective. Chin Med 2009;4:1. 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 15 BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 19. Lundeberg T, Lund I, Naslund J, et al. The Emperors sham - wrong assumption that sham needling is sham. 4 Acupunct Med 2008;26:239-42. 5 6 20. Linde K, Niemann K, Schneider A, et al. How large are the nonspecific effects of acupuncture? A meta- 7 analysis of randomized controlled trials. BMC Med 2010;8:75. 8 9 21. Vas J, Perea-Milla E, Mendez C, et al. Efficacy and safety of acupuncture for the treatment of non-specific 10 acute low back pain: a randomised controlled multicentre trial protocol [ISRCTN65814467]. BMC 11 Complement Altern Med 2006;6:14. 12 13 22. Shin JS, Ha IH, Lee TG, et al. Motion style acupuncture treatment (MSAT) for acute low back pain with 14 severe disability: a multicenter, randomized, controlled trial protocol. BMC Complement Altern Med 15 2011;11:127.For peer review only 16 17 23. Deadman P A-KM, Baker K. A Manual of Acupuncture. 3 ed ed: Journal of Chinese Medicine, 2008:574-80. 18 19 24. Heyerdahl O, Lystad N. [Textbook of Acupuncture]. Oslo: Universitetsforlaget, 2007:291-303. 20 21 25. White A, Cummings M., Filshie J. An introduction to western medical acupuncture. London: Churchill 22 Livingstone Elsevier, 2008:7-76. 23 24 26. Witt CM, Jena S, Selim D, et al. Pragmatic randomized trial evaluating the clinical and economic effectiveness 25 of acupuncture for chronic low back pain. Am J Epidemiol 2006;164:487-96. 26 27 27. Kalauokalani D, Cherkin DC, Sherman KJ, et al. Lessons from a trial of acupuncture and massage for low 28 back pain: patient expectations and treatment effects. Spine (Phila Pa 1976) 2001;26:1418-24. 29 30 28. Linde K, Witt CM, Streng A, et al. The impact of patient expectations on outcomes in four randomized 31 controlled trials of acupuncture in patients with chronic pain. Pain 2007;128:264-71. 32

33 29. Thomas KJ, MacPherson H, Thorpe L, et al. Randomised controlled trial of a short course of traditional http://bmjopen.bmj.com/ 34 acupuncture compared with usual care for persistent non-specific low back pain. BMJ 2006;333:623. 35 36 30. Lin CW, Haas M, Maher CG, et al. Cost-effectiveness of guideline-endorsed treatments for low back pain: a 37 systematic review. Eur Spine J 2011;20:1024-38. 38 39 31. Skjeie H, Skonnord T, Fetveit A, et al. A pilot study of ST36 acupuncture for infantile colic. Acupunct Med 40 2011;29:103-7. 41 on September 29, 2021 by guest. Protected copyright. 42 32. ClinicalTrials. Effects of Acupuncture in the Treatment of Infant Colic (NCT00907621). 2009. 43 44 33. Mannion AF, Elfering A, Staerkle R, et al. Outcome assessment in low back pain: how low can you go? Eur 45 Spine J 2005;14:1014-26. 46 47 34. Mannion AF, Porchet F, Kleinstuck FS, et al. The quality of spine surgery from the patient's perspective. Part 48 1: the Core Outcome Measures Index in clinical practice. Eur Spine J 2009;18 Suppl 3:367-73. 49 50 35. Grotle M, Brox JI, Vollestad NK. Concurrent comparison of responsiveness in pain and functional status 51 measurements used for patients with low back pain. Spine (Phila Pa 1976) 2004;29:E492-501. 52 53 36. Roland M, Morris R. A study of the natural history of back pain. Part I: development of a reliable and sensitive 54 measure of disability in low-back pain. Spine (Phila Pa 1976) 1983;8:141-4. 55 56 37. Solberg TK, Olsen JA, Ingebrigtsen T, et al. Health-related quality of life assessment by the EuroQol-5D can 57 provide cost-utility data in the field of low-back surgery. Eur Spine J 2005;14:1000-7. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 15 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2012-001164 on 25 June 2012. Downloaded from 1 2 3 38. Linton SJ, Hallden K. Can we screen for problematic back pain? A screening questionnaire for predicting 4 outcome in acute and subacute back pain. Clin J Pain 1998;14:209-15. 5 6 39. Grotle M, Vollestad NK, Brox JI. Screening for yellow flags in first-time acute low back pain: reliability and 7 validity of a Norwegian version of the Acute Low Back Pain Screening Questionnaire. Clin J Pain 8 2006;22:458-67. 9 10 40. Grovle L, Haugen AJ, Ihlebaek CM, et al. Comorbid subjective health complaints in patients with sciatica: a 11 prospective study including comparison with the general population. J Psychosom Res 2011;70:548-56. 12 13 41. Curtis P, Carey TS, Evans P, et al. Training primary care physicians to give limited manual therapy for low 14 back pain: patient outcomes. Spine (Phila Pa 1976) 2000;25:2954-60; discussion 60-1. 15 For peer review only 16 42. Hancock MJ, Maher CG, Latimer J, et al. Assessment of diclofenac or spinal manipulative therapy, or both, in 17 addition to recommended first-line treatment for acute low back pain: a randomised controlled trial. Lancet 18 2007;370:1638-43. 19 20 43. Schoenfeld D. Find statistical considerations for a study where the outcome is a time to failure 21 (http://hedwig.mgh.harvard.edu/sample_size/time_to_event/para_time.html). 2001. 22 23 44. Vincent C. The safety of acupuncture. BMJ 2001;323:467-8. 24 25 45. White A. A cumulative review of the range and incidence of significant adverse events associated with 26 acupuncture. Acupunct Med 2004;22:122-33. 27 28 46. Witt CM, Pach D, Brinkhaus B, et al. Safety of acupuncture: results of a prospective observational study with 29 229,230 patients and introduction of a medical information and consent form. Forsch Komplementmed 30 2009;16:91-7. 31 32