Executive Summary

Background

Back and neck pain are important health problems with serious societal and economic consequences. The prevalence of back and/or neck pain in US in 2007 was estimated to be 31 percent. The costs associated with low productivity, lost-time at work, permanent disability, and healthcare are enormous. Conventional medical treatments have been shown to have limited effectiveness in the management of back and neck pain. Complementary and Alternative Medicine (CAM) therapies offer additional options for management of back and neck pain. The number of people in Western societies using CAM therapies is increasing. The most prevalent CAM therapies are spinal manipulation, acupuncture, and massage. The number of randomized controlled trials (RCTs) evaluating CAM therapies for back and neck pain has increased over the past two decades. The results of these trials are inconsistent.

The University of Ottawa Evidence-based Practice Center (UO-EPC) reviewed and synthesized evidence to better understand effectiveness and safety of the most prevalent CAM therapies in the management of back, neck, and thoracic pain in adults. The current review commissioned by the Agency for Healthcare Research and Quality (AHRQ) and National Center for Complementary and Alternative Medicine (NCCAM) aimed to address the following research Key Questions (KQ):

KQ1. What is the efficacy, effectiveness and cost-effectiveness of the most prevalent types of practitioner-based manual CAM therapies (e.g., spinal manipulation, spinal mobilization, massage; acupuncture) compared to other CAM therapies, conventional therapies, placebo, no treatment, or wait list in improving outcomes (e.g., QoL, Pain, Function, progression of acute to chronic/ or disabling BP) in patients with nonspecific and certain specific (e.g. disc herniation, spinal stenosis, facet joint syndrome, whiplash) types of back and neck pain.

  1. For any of the CAM therapies found to be effective for BP, what factors influence success of treatments?
    1. Patient-specific factors
    2. Socio-demographics (e.g., age, gender, race, education, income)
    3. Comorbidities
  2. Severity, specific causes (as identified in Q1), and duration of BP
    1. Treatment-specific factors (e.g., dose, frequency, duration)
    2. Treatment provider-specific factors (e.g., training, specialization, experience)
  3. Does the use of any of the 3 most prevalent types of CAM for BP in adults result in a decreased or increased utilization of conventional management (diagnostic tests, number of visits & dose of medications, procedures)?

KQ2. What are the contraindications and safety profile of the three most prevalent CAM therapies for BP in adults compared to that for other CAM therapies, conventional therapies, placebo or no treatment? Does the safety profile of these therapies change across subgroups of patients with comorbidities?

Methods

Data Sources and Search Strategy

A comprehensive search was conducted in the following databases: MEDLINE, the Cochrane Library, EMBASE, and Allied and Complementary Medicine (AMED) were all searched from inception until February 2010. CINAHL, Mantis, and the ACP Journal Club were also searched from inception until September 2008. Additional literature was searched through bibliographies of relevant items. The Web sites of relevant organizations/agencies, trial registries, and conference proceedings were searched for the grey literature.

Study Selection Criteria and Process

Randomized controlled trials (RCTs) reporting efficacy/effectiveness and/or economic data of the CAM therapies (acupuncture, manipulation, mobilization, massage) versus any inactive or active treatments in adults with back, neck, or thoracic pain were eligible. Nonrandomized controlled trials and observational studies (e.g., cohort, case-control, cross-sectional) reporting harms were also included.

Reports published in English, German, Dutch, Chinese, Japanese, Italian, French, Portuguese, and Spanish were eligible for inclusion. Systematic and narrative reviews, case reports, editorials, commentaries or letters to the editor were excluded.

Titles and abstracts of all identified bibliographies were screened for eligibility by two independent reviewers who later reviewed full-text reports of potentially eligible records. Discrepancies were resolved by consensus.

Data Extraction

Two independent reviewers extracted data using an a priori developed abstraction form. The abstracted data were crosschecked and conflicts were resolved by consensus.

Primary efficacy/effectiveness outcomes included pain intensity (e.g., Visual Analog Scale-VAS, McGill Pain Questionnaire-MPQ) function (Hannover Functional Ability Questionnaire-HFAQ), and disability (e.g., Roland Morris Disability Questionnaire-RMDQ, Northwick Park Neck Pain Questionnaire-NPQ, Pain Disability Index-PDI, and Oswestry Disability Index). Secondary outcomes included spinal range of motion (ROM), straight leg raise (SLR), finger-to-floor distance (FFD), and muscle strength. Harms (e.g., any adverse event, withdrawals due to adverse events, specific adverse events) were extracted as proportions of patients with an event.

For cost-effectiveness analysis, data was extracted on: a) costs in the health care sector, b) costs of production loss, c) costs in other sectors, d) patient and family costs, and e) total costs.

Assessment of Study Quality and Reporting

The risk of bias for RCTs was assessed using the criteria list recommended in the Updated Method Guidelines for Systematic Reviews in the Cochrane Collaboration Back Review Group. Depending on the number of ‘Yes’ ratings (score range: 0-4) across four domains (treatment allocation concealment, balance in baseline characteristics, blinding, and number/reasons for dropouts), the quality of individual studies was classified into three groups: good (score: 4), fair (score: 2-3), and poor (score: 0-1). The overall bias was explored using risk-of-bias graphs. The quality of observational studies was assessed using the modified Downs and Black tool. Methodological quality of economic studies was determined using the 19-item CHEC list.

Quantitative Synthesis

The results were grouped according to a type of experimental intervention (e.g., acupuncture, manipulation, mobilization, massage), pain location in spinal region (low back, neck, thorax), duration of pain (acute/sub-acute, chronic, mixed, unknown), and cause of pain (specific versus nonspecific). Study, treatment, population, and outcome characteristics were summarized in text and/or summary tables.

We pooled RCTs with similar populations (demographics, duration, and cause of pain), same types of experimental and controls treatments, which reported outcomes measured with the same instruments (and scale) at similar post-treatment followup periods. The meta-analyses of pain were based on a 1-10 visual analogue scale. The random-effects models of DerSimonian and Laird were used to generate pooled estimates of relative risks (RRs) and weighted end point mean difference (WMDs) with 95 percent confidence intervals (CIs). Statistical heterogeneity was evaluated using the Chi-square test and the I2 statistic (low: 25.0 percent; moderate: 50.0 percent; high: 75.0 percent). Subgroup (e.g., patients’ age, gender) and sensitivity (e.g., trial quality) analyses were planned to investigate the sources of unexplained heterogeneity.

If data allowed, the statistically significant pooled estimates of post-treatment pain intensity were planned to be examined in order to determine the degree of clinical importance for the observed differences between the treatment groups. The degree of clinical importance was defined as small (WMD < 10 percent of the VAS scale), medium (10 percent ≤ WMD < 20 percent of the VAS scale), and large (WMD ≥20 percent of the VAS scale).

Publication bias was examined through visual inspection of funnel plot asymmetry with respect to contours of statistical significance and the Egger’s regression-based method.

Rating the Strength of Evidence

We assessed the overall strength of evidence using the approach of grading system outlined in the Methods guide prepared for the AHRQ Evidence-based Practice Center (EPC) program. The grading method consists of four major domains: risk of bias (high, medium, low), consistency, directness, and precision. Body of evidence for a given outcome was classified into four groups: high, moderate, low, or insufficient (no evidence). The initial grade was reduced by one level (e.g., from high to moderate; from moderate to low) for each of the domains not met and by two levels in case of high risk of bias (e.g., from high to low grade; from moderate to low grade).

Results

KQ1. Efficacy/effectiveness and Cost-effectiveness of CAM Therapies Compared to no Treatment, Placebo, and Other Active CAM/nonCAM Therapies in Management of Back, Neck, and/or Thoracic Pain

In total, 265 RCTs (including 10 reporting economic data), and five controlled observational studies which provided harms data were included. Most studies included subjects with chronic nonspecific pain. Immediate and short-term post-treatment follow-ups were most frequently reported. Only the main findings for pain, disability, and mobility are reported in this summary.

Efficacy/Effectiveness

Acupuncture – Low Back Pain. In subjects with chronic nonspecific LBP, acupuncture compared to placebo led to statistically significantly lower pain intensity (Grade – moderate) but only for the immediate-post-treatment followup (10 trials; pooled VAS: -0.59, 95 percent CI: -0.93, -0.25). Acupuncture did not significantly differ from placebo in improving pain intensity scores, well-being, disability, use of medication, proportion of subjects on sick leave, and proportion of subjects with global improvement at short-, intermediate-, and long-term post-treatment. Trials using TENS-sham, laser-sham, or medication-sham compared to those using sham-acupuncture tended to produce results in favor of acupuncture in relation to pain intensity and disability.

Subjects in acupuncture group had statistically significantly better post-treatment pain intensity (three trials; pooled short-term VAS: -1.19, 95 percent CI: -2.17, -0.21; Grade - moderate), pain disability index (one trial; immediate-term PDI; Grade - moderate), function (two trials; immediate-term HFAQ; Grade - moderate), well-being (two trials; immediate-term SF-36; Grade - moderate), or ROM (one trial; immediate-/intermediate-term extension, flexion Grade - low) compared to ‘no treatment’ group.

Subjects in the acupuncture group compared with those in usual care had significantly better post-treatment pain intensity (two trials; short-/intermediate-term VAS; Grade - moderate), disability (two trials; short-/intermediate-term RMDQ; Grade - moderate), quality of life (one trial; intermediate-term SF-12; Grade - moderate), or function (one trial; intermediate-term HFAQ; Grade - moderate).

In two meta-analyses, acupuncture did not significantly differ from pain medication in reducing immediate post-treatment pain (four trials; immediate-term pooled VAS: 0.11, 95 percent CI: -1.42, 1.65; Grade – low) or disability (two trials; pooled Oswestry: -2.40, 95 percent CI: -12.20, 7.40; Grade – low). In contrast, another meta-analyses indicated significantly worse post-treatment pain (two trials; immediate VAS: 3.70, 95 percent CI: 1.50, 5.80; Grade – low) for acupuncture versus manipulation.

Subjects receiving acupuncture had significantly better immediate post-treatment pain and disability than subjects receiving physiotherapy (two trials; trial one - light, electricity and heat therapy; trial two - hot packs, ultrasound, short-wave diathermy, TENS, muscle strengthening; Grade – low). There was no difference between acupuncture and massage in healthcare utilization (one trial; intermediate-term number of provider visits, percentage of subjects using medication; Grade - low).

In subjects with acute nonspecific LBP there was no statistically significantly difference between acupuncture and usual care groups (one trial; immediate-short-intermediate-term RMDQ; Grade-low).

Acupuncture – Neck Pain. Two meta-analyses indicated no significant difference between acupuncture and sham-acupuncture in subjects with chronic specific (two trials; Grade – moderate; VAS: 0.27, 95 percent CI: -0.60, 1.13) or nonspecific pain (three trials; Grade – low; VAS: -0.24, 95 percent CI: -1.20, 0.73) for immediate post-treatment pain intensity. Trials using TENS-sham, laser-sham, or medication-sham compared to those using sham-acupuncture tended to produce results in favor of acupuncture in relation to pain intensity and disability.

There were inconsistent results for immediate- or short-term post-treatment pain intensity between acupuncture and pain medication groups of subjects. Intermediate-term followup indicated no significant difference between acupuncture and pain medication groups. Acupuncture did not differ from standard mobilization and traction techniques or laser therapy in short-term post-treatment pain intensity or disability. Immediate/short-term post-treatment pain and disability were better in manipulation than acupuncture groups (two trials, Grade - low).

Manipulation – Low Back Pain. In subjects with acute/sub-acute nonspecific LBP, manipulation was significantly more effective than placebo (five trials; Grade – moderate) or no treatment (one trial; Grade – low) in reducing pain intensity (VAS score) immediately or short-term after the end of treatment. There was no significant difference between manipulation and placebo in post-treatment pain medication intake, disability, or back flexibility (three trials; Grade -low). Manipulation did not differ from medication in reducing pain intensity (two trials; Grade - low).

In subjects with chronic nonspecific LBP, manipulation was significantly more effective than placebo in reducing pain intensity (VAS score) immediately or short-term after the end of treatment (three trials; Grade – low). Manipulation was significantly better (in immediate post-treatment pain; two trials; Grade – low) or no different (in intermediate-term post-treatment pain; one trial; Grade – low) from medication in improving pain intensity.

In older subjects with mixed duration LBP, spinal manipulation was significantly better than medical care in improving immediate and short-term post-treatment disability (Oswestry) and perception of global improvement but not pain intensity or physical function (one trial; Grade – low).

Results from studies comparing manipulation to massage or physiotherapy in improving post-treatment pain intensity (two trials; Grade – low) or mobility (three trials; grade – low) were inconsistent, either in favor of manipulation or indicating no significant difference between the two treatments.

In two large trials (UK BEAM and Childs 2004), subjects receiving combination of manipulation and exercise or best care by general practitioner improved in pain and disability compared to subjects with no spinal manipulation treatment (Grade – Moderate).

Manipulation – Neck Pain. Subjects receiving manipulation had significantly better post-treatment pain (two trials; Grade - moderate) disability (one trial; Grade - low), or mobility (extension, flexion, rotation) (one trial; Grade - low) compared to those taking placebo. Manipulation did not significantly differ from medication in reducing pain intensity (three trials; Grade - moderate) and disability (two trials - Grade – moderate).

Mobilization – Low Back Pain. There were no significant differences in pain intensity (VAS) and ROM (flexion, extension, floor to floor) between subjects who received mobilization and placebo immediately or short-term after treatment (two trials; specific acute/sub-acute and nonspecific mixed duration pain; grade – low). Subjects with specific acute/sub-acute pain receiving mobilization had significantly reduced intake of analgesic medication and duration of sick leave compared to those receiving placebo (one trial; Grade - low).

Subjects with acute/sub-acute and chronic pain (specific or nonspecific) receiving mobilization experienced significantly improved pain intensity (VAS, MPQ) and lumbar ROM (side bending) compared to subjects not receiving any treatment, immediately and short-term after treatment (two trials; Grade – low). Results regarding disability (RMDQ, Oswestry) were inconsistent, showing either a significant difference in favor of mobilization (one trial; Grade – low) or no difference between mobilization and no treatment (one trial; Grade - low). In subjects with mixed duration of pain, there were no significant differences in pain intensity (VAS) and ROM (flexion, extension) immediately or short-term after treatment (one trial; Grade – low).

In two meta-analyses, subjects with chronic nonspecific pain receiving mobilization (traditional bone setting) compared to physiotherapy (massage, stretching, trunk exercise) had significantly lower pain intensity (two trials, Grade – low; VAS score: -0.50, 95 percent CI: -0.70, -0.30) and disability degree (two trials, Grade – moderate; Oswestry score: -4.93, 95 percent CI: -5.91, -3.96) immediately after treatment. There was no significant difference between the groups in the mean finger to floor distance immediately after treatment (two trials, Grade – moderate; -0.89, 95 percent CI: -1.89, 0.12). Similarly, mobilization and physiotherapy groups did not significantly differ in ROM (Schober’s test, extension, straight leg raising) immediately or intermediate-term after treatment (one trial; Grade - low). In subjects with nonspecific pain of mixed duration, mobilization was significantly superior to physiotherapy (massage, mobilization, thermal, and electrotherapies according to the Finnish routine) in reducing disability (Oswestry), but only at intermediate term post-treatment (one trial, Grade – low). There was no significant difference between the groups in the number of sick leave days for the trial period.

The immediate- or intermediate-term post-treatment pain intensity (VAS score; one trial; Grade - low), disability (Oswestry; one trial; Grade - low), and ROM (Schober’s test, extension, straight leg raising; two trials; Grade - low) did not significantly differ between mobilization and exercise given to subjects with nonspecific chronic or mixed duration of pain.

Mobilization – Neck Pain. Mobilization was significantly better than placebo in subjects with acute/sub-acute nonspecific pain (one trial; Grade – Low), but did not differ from placebo in subjects with chronic nonspecific pain (one trial; Grade – low).

Subjects with chronic or mixed nonspecific pain receiving mobilization had significantly lower pain intensity compared to no treatment (two trials; Grade - Low). There was no significant difference between the mobilization and no treatment groups in the mean intake of analgesic medication pills and the number of sick leave days immediately or short-term after treatment (one trial; Grade - low).

Mobilization was significantly better than massage or physiotherapy (massage, stretching and exercise) in improving pain (VAS score), disability (NDI), global assessment, analgesic medication intake, and the number of sick leave days in chronic nonspecific pain at intermediate-term post-treatment followup (one trial; Grade – Low). Subjects with mixed duration nonspecific pain in the mobilization and continued general practitioner care groups had similar post-treatment pain intensity and disability (VAS and NDI; one trial; Grade – low).

Massage – Low Back Pain. Subjects with nonspecific acute/sub-acute pain receiving massage had significantly better pain intensity (VAS, MPQ) and disability (Oswestry) compared to no treatment (one trial; Grade – low) or placebo (two trials; Grade –moderate) immediately or short-term after the end of treatment. In subjects with nonspecific chronic pain, massage did not significantly differ from no treatment or placebo in improving immediate or intermediate-term post-treatment pain intensity (VAS, MPQ; two trials; Grade – low), disability (Oswestry, RMDQ; two trials; Grade – low), general health perception (one trial; Grade – low), or health status (SF-36; one trial; Grade – low).

In two meta-analyses, massage was significantly better in reducing pain compared to relaxation (two trials; Grade – low; VAS score: -1.27, 95 percent CI: -2.46, -0.08) or physical therapy (two trials; Grade – moderate; VAS score: -2.11, 95 percent CI: -3.15, -1.07) immediately after treatment in subjects with chronic nonspecific pain. In another meta-analysis, massage was not significantly better than relaxation in improving immediate post-treatment ROM (two trials; Grade – low; trunk flexion: 2.21, 95 percent CI: -1.10, 5.52). In subjects with chronic nonspecific pain, massage was significantly more effective than physical therapy in reducing pain (SF-PQ, VAS), disability (RMDQ, modified Oswestry), and the number of days off work immediately or intermediate-term after the treatment (two trials; Grade - low). In subjects with chronic nonspecific pain, there was no significant difference between receiving massage and general practitioner care in improving pain (VAS score), disability (RMDQ), or well-being (SF-36) intermediate-term after the end of treatment (one trial; Grade – low).

Massage – Neck Pain. Subjects with acute/sub-acute, chronic, or unknown duration of nonspecific pain receiving massage had significantly improved pain (≥2-point decrease on NRS-11, VAS, Pressure Pain Threshold scores) compared to placebo (three trials; Grade – Low), immediately after treatment. In subjects with chronic specific pain, massage did not significantly differ from placebo in improving range of mobility (one trial; Grade – low) or well-being (SF-36, role physical, pain index; one trial; Grade - low).

Massage, compared to no treatment, significantly improved pain intensity (NPQ, VAS scores) but not ROM (flexion, extension) in subjects with chronic or unknown duration of nonspecific pain, immediately after the end of treatment (two trials; Grade – low).

In subjects with chronic nonspecific pain (one trial; Grade – Low), massage compared to exercise significantly improved disability (NPQ) but not ROM (flexion, extension).

Cost-effectiveness

This review included results from 10 studies of full economic evaluations of acupuncture (low back pain: two studies, neck pain: one study), spinal manipulation (low back pain: four studies, neck pain: two studies), and massage (one study) for low back and neck pain.

Acupuncture - Low back pain. Two economic evaluations showed that acupuncture was cost-effective compared to usual care and compared to no treatment in patients with chronic low back pain. However, in both studies health gains were small and one study used no treatment control group and had only 3 months followup.

Acupuncture - Neck pain. One study showed that in subjects with chronic neck pain acupuncture use was associated with significantly higher total costs compared to usual care ($1,565 versus $1,496).

Manipulation - Low back pain. There were no differences in costs between manual therapy, general practitioner care and intensive therapy for acute low back pain. Costs were higher for manipulation compared with medical care without producing better clinical outcomes for patients with mixed duration of LBP (acute, subacute and chronic). This was associated with significantly more visits to chiropractic care than medical care. Spinal manipulation in addition to general practitioner care was relatively cost-effective compared to general practitioner care alone for patients with sub-acute and chronic low back pain. In chronic LBP patients, there were no differences in costs between physician consultation, spinal manipulation plus stabilizing exercises, and physician consultation alone. Results are difficult to compare due to differences in health care systems, perspectives, interventions, populations, and methods used.

Manipulation - Neck pain. One study in subjects with neck pain found that pulsed short-wave diathermy was less cost-effective compared with manual therapy or exercise/advise. In another study, manual therapy was less costly and more effective than physiotherapy (functional, active and postural or relaxation exercises, and stretching) or general practitioner care.

Massage - Low back pain. One study reported an economic evaluation of therapeutic massage, exercise, Alexander technique, and usual general practitioner care in patients with chronic low back pain showing that massage was more costly and less effective than usual care by the general practitioner.

KQ1 a-c. Patient- and Trial-specific Factors Influencing Treatment Success

The amount of evidence regarding factors potentially influencing treatment effect (e.g., age, gender race, education, income, cause of pain, type of treatment provider, dose of treatment) was relatively limited.

There was no discernable pattern indicating that the effect of acupuncture was different in subjects with specific and nonspecific pain (neck and low back pain).

In one trial (Grade – Low), the subject’s age, gender, symptom duration, or the therapist’s years of experience did not have a significant effect on the mean change for Oswestry score between spinal manipulation in addition to exercise and exercise alone. In another trial (Grade - Moderate) the beneficial effect of massage compared to physical therapy (physical modalities, exercise and traction) was similar across age (≤ 50, and > 50 years) and gender groups.

Massage was significantly better in reducing pain intensity compared to physical therapy in subjects with severe pain at baseline. The reduction in pain intensity did not differ between the massage and physical therapy groups amongst subjects with lower baseline pain scores. The baseline severity did not modify the effect of massage (versus physical therapy) measured at intermediate-term after the end of treatment (i.e., massage was significantly better than physical therapy in reducing pain across the baseline pain severity groups).

This review identified evidence on utilization of conventional healthcare (e.g., routine visits to physician, use of analgesics, hospital stay) and work absenteeism.

The use of conventional healthcare was not different in acupuncture versus self care, usual care, or massage for subject with chronic LBP (two trials). Similarly, the use of conventional care did not differ between spinal manipulation and hospital outpatient management, or physician consultation for LBP (two trials). In contrast, the use of conventional care was significantly reduced for subjects receiving spinal manipulation in addition to exercise compared to exercise alone (one trial).

The use of analgesics was significantly reduced for acupuncture versus placebo, waiting list, TENS, or usual care in LBP (four trials); and between acupuncture and placebo in subjects with neck pain (one trial). In contrast, the use of analgesics did not significantly differ between acupuncture and self care in LBP (one trail); between acupuncture and placebo, self care, or other treatments for NP (five trials). Similarly, the use of analgesics did not differ between spinal manipulation and placebo, or conventional care in subjects with low back pain(five trials, Grade – Low); between spinal manipulation or mobilization and prescription medication, no treatment neck collar, or physiotherapy for subjects with neck pain (five trials).

The extent of work absenteeism was significantly greater in exercise alone versus acupuncture in LBP (one trial); between mobilization and no treatment or physiotherapy in neck pain (two trials). Sick leaves due to pain, did not differ between acupuncture and placebo or usual care in LBP (two trails); and between mobilization and neck collar or ‘act as usual’ in neck pain (one trial).

The sensitivity analysis, performed on acupuncture trials, found no evidence that treatment effect was strongly influenced by study quality. The pooled estimates of mean difference in pain intensity (VAS score) for ‘lower risk-of-bias’ and ‘higher risk-of-bias’ trials were -0.43 (95 percent CI: -0.76, -0.09) and -0.75 (95 percent CI: -1.39, -0.11), respectively.

KQ2. Harms of CAM Therapies

The reporting of harms was poor across studies of CAM interventions. Only very few trials reported any information on adverse events. The reported information was not detailed, lacked consistency, and was not comparable. No definitions were presented. Therefore, the rates of adverse events between the different interventions could not be meaningfully compared.

Acupuncture. The reported events in RCTs were mostly of moderate transient nature amongst these most commonly reported events were soreness/pain at the site of needling, bruising light headedness, minor bleeding, dizziness, or headache. The proportion of subjects with any adverse event did not reportedly differ in acupuncture versus TENS or usual care groups. In one nonrandomized trial, discomfort or soreness in the acupuncture, chiropractic therapy, and massage groups were 5.0 percent, 8.0 percent, and 7.0 percent, respectively.

Manipulation/Mobilization. The reported events in RCTs were mostly moderate in severity and of transient nature (e.g., increased pain). In one RCT, after 2 weeks of treatment, patients with neck pain receiving manipulation were not at significantly increased risk for having an adverse event compared to patients receiving mobilization (OR = 1.44, 95 percent CI: 0.83, 2.49). In another RCT, the proportion of patients with neck pain having adverse events was similar in manipulation versus Diazepam groups (9.5 percent versus 11.1 percent). In two case control studies, subjects ≤ 45 years of age with vertebro-basilar artery (VBA) stroke were more likely to visit a chiropractic or primary care physician than subjects without VBA stroke. This association was not observed in older subject visiting a chiropractic clinic. In one case control study, the excess risk of vascular accident was observed for both, subjects undergoing chiropractic care and subjects undergoing primary care treatments. In another case-control study, subjects with cervical artery dissection were more likely to have had spinal manipulation within 30 days (OR = 6.62, 95 percent CI: 1.4, 30.0). In one cohort study, rate of complications did not differ between subjects with low back pain receiving manipulation plus mobilization versus no treatment.

Massage. In few RCTs, subjects receiving massage experienced worsening of back/neck pain or soreness of mild and transient nature. One study reported allergic reactions (rashes and pimples) in five subjects due to massage oil.

In one RCT, the proportion of patients with neck pain having adverse events in massage group was lower (7.0 percent) compared to acupuncture (33.0 percent) or placebo-laser (21.0 percent).

Conclusions and Future Research

This review identified a large amount of evidence on comparative effectiveness of single mode CAM interventions for management of back and neck pain in subjects with a wide spectrum of causes (specific and nonspecific) and duration (acute to chronic) of pain. The reviewed evidence was of low to moderate grade and inconsistent probably due to substantial methodological and/or clinical diversity, thereby rendering some between-treatment comparisons inconclusive. The differences in the therapy provider’s experience, training, and approaches (e.g., deep or superficial massage, choice of trigger points, needling techniques) may have additionally contributed to disparate results. Evidence for acute, sub-acute, and mixed specific pain was sparse compared to that for chronic nonspecific pain. Poorly reported harms data limited our ability to meaningfully compare rates of adverse events between the treatments.

Generally, CAM treatments were more effective in reducing pain and disability compared to no treatment, physical therapy, or standard care immediately or at short-term followup. Results of trials comparing CAM treatment to sham were less consistent either showing significant differences in favor of CAM or no significant differences between the treatments.

For chronic nonspecific back pain, acupuncture was better than placebo but only for improving pain intensity at immediately post-treatment. The long-term post-treatment disability and utilization of conventional healthcare did not differ between subjects with low back pain receiving acupuncture and usual care. Trials that applied sham-acupuncture tended to produce negative results (i.e., statistically nonsignificant) compared to trials that applied other types of placebo (e.g., TENS, medication, laser) between acupuncture and placebo groups. Acupuncture significantly decreased pain medication use compared with no treatment or placebo, but not so compared with self-care, Botulinum toxin, or Lidocaine injection. There was no statistically significant difference between acupuncture and control treatments for the number of visits to other healthcare providers.

Manipulation did not differ from pain medication in improving pain intensity. Manipulation was significantly more effective than acupuncture in reducing immediate post-treatment pain intensity. Results for pain intensity or disability were inconsistent regarding manipulation compared to massage or physiotherapy for treatment of LBP. Subjects receiving manipulation did not differ in healthcare utilization from subjects receiving exercise, physician consultation, medical care, or placebo.

In chronic or mixed duration of low back pain, mobilization was similar to placebo in reducing pain or to physiotherapy (which may include a combination of manual treatment and physical modality but not physical modalities alone) in improving immediately after or short term post-treatment.

For subjects with chronic neck pain, acupuncture was not different from sham-acupuncture, pain medication, mobilization/traction, or laser therapy in reducing pain or disability after the treatment. Subjects with neck pain benefited more with manipulation than placebo in terms of pain, disability, and neck flexibility.

Mobilization was more effective than placebo in improving acute/subacute neck pain but not chronic neck pain. In subjects with neck pain (chronic, mixed duration), mobilization was better than no treatment in reducing pain intensity, but not in reducing the intake of pain medication pills or the number of sick leave days immediately or short-term after the treatment. Mobilization was better than physiotherapy or massage in reducing pain intensity and disability in subjects with chronic nonspecific neck pain. Massage was not different from placebo in improving well-being or ROM in subjects with chronic neck pain.

In summary, the degree of clinical importance for the differences in pooled pain intensity observed between the treatment groups for low back pain was small (acupuncture versus placebo; mobilization versus physical therapy), medium (acupuncture versus no treatment; massage versus relaxation), or large (acupuncture versus manipulation, in favor of manipulation; massage versus physical therapy).

Due to the small number of economic evaluations, inconsistent standards of comparison, and substantial heterogeneity (diversity of healthcare payment systems across countries) it was not possible to reach clear conclusions about the cost-effectiveness of any of the CAM treatments. Spinal manipulation for back pain was not cost-effective relative to medical care. Acupuncture was cost-effective relative to usual care or no treatment in subjects with back pain. Evidence for massage was insufficient.

In several studies subjects receiving CAM therapies reported soreness or bleeding on the site of application (acupuncture groups) and worsening of pain/back pain (manipulation/massage groups).

More data from long term and large head to head trials with sufficient duration of CAM treatments and trials comparing CAM treatment to other widely used active treatments (e.g. comprehensive physiotherapy) reporting clinically relevant and validated outcomes (e.g. pain intensity, disability, direct and indirect costs, utility of conventional care, and adverse events) are needed for definitive conclusions.

Future studies should control for or examine the influence of treatment dose/duration, care provider-(e.g. certification, years of experience) and population-specific variables on treatment effect estimate.