Critical New Developments in the Safety and Efficacy of Chiropractic Treatment, Part 2

By Arthur Croft, DC, MS, MPH, FACO

In part one of this editorial (Jan. 1, 2008), I began with something of a call to arms by drawing attention to a paper out of Portugal that was critical of chiropractic, arguing that it was not well-proven in terms of effectiveness and that its safety had not been adequately demonstrated.4 The three cases they presented of serious untoward reactions to chiropractic therapy were intended to underscore their recommendation for the implementation of a "risk alert system." Although this was not further explored, one wonders what they meant by it. But if equivalent risk alerts were issued for medicine and surgery, I wonder how long it would last.

The real substance of the editorial was two papers published this year that all chiropractic practitioners must acquire, read and keep handy. I also mentioned several other papers which should also be a part of every DC's personal library. They demonstrate that, contrary to the imputations of Gouveia, et al.,4 chiropractic has been demonstrated to be both very safe and, in most studies, more effective than medicine. The paper I reviewed first was that of Gross, et al.5 They showed that after a thorough review of the published literature, including the Cochrane Register of Controlled Trials (Central), manipulation/mobilization with exercise showed "strong evidence" of effectiveness. In this highest category of scientific evidence, conspicuously absent was any medical or pharmaceutical intervention.

However, in some ways, the results might have seemed a bit ambiguous or disappointing to some. For example, for manipulation/mobilization with modalities or manipulation alone, there was "evidence of no benefit." These findings based on published research and clinical trials can sometimes fail to show an effect because of study design problems, including small sample size, lack of statistical power, bias or confounding of one kind or another. An actual treatment effect may be hard to see: When there is no control group, follow-up times are brief, or interventions are inconsistent or non-representative.

In this second part of this editorial, let's look at two researchers who have taken a different look at this literature - some might say a gentler, kinder look. This paper by Vernon and Humphreys6 is in an open-access journal, so you can download it free at:

Author Abstract

Manual therapy for neck pain enjoys a long history, with increasing popularity in recent times. The evidence base of manual therapies for neck pain consists of a reasonably large body of clinical trials, an even greater number of systematic reviews and, more recently, a number of practice guidelines. We have conducted several systematic reviews pertaining to the evidence base for both acute and chronic neck pain, as well as for the outcome of control groups of chronic neck pain subjects in clinical trials of conservative therapies.

In this review, we first provide background material on the definition and characterization of manual therapies as well as on the epidemiology of neck pain. We then review our recent systematic reviews on manual therapies for acute and chronic neck pain without whiplash. Finally, we provide brief, original reviews of the literature on the treatment of whiplash injury by manual therapies, followed by the current practice guidelines pertaining to manual therapies for neck pain. While there are several publications, especially those registered with the Cochrane Collaboration, that are currently the authoritative evaluations of the use of manual therapies for neck pain, the present review is designed to present a broad overview of the topic with a distinctive approach, emphasizing the analysis of change scores in the clinical trials. It is hoped this will benefit researchers and clinicians alike in their management of neck pain patients.


For chiropractic physicians, this is an essential document because the authors systematically reviewed a very broad literature base that has assessed the relative efficacy of manual therapy in the management of neck pain. In this context, manual therapy (MT) can include massage therapy in its many manifestations, mobilization and spinal manipulative therapy (SMT). In this context, SMT is not necessarily the equivalent of CMT. There have been papers registered with the Cochrane Collaboration that were not entirely enthusiastic, as we saw in part one of my article. It is always important to understand these studies and their methodologies and limitations. What Vernon and Humphreys did that was unique was to analyze this literature on the basis of change scores.6

The authors first discussed the epidemiology of neck pain, even though it seemed to be a bit tangential to their subject. They reported that according to the Quebec Task Force (QTF), recovery from whiplash should take only four weeks. I must admit to being a bit disappointed, since Dr. Vernon and I sat on a multidisciplinary international panel on whiplash with one of the QTF authors in the same year the QTF whiplash document was published. Dr. Vernon was the only other member of the panel who claimed to have read it and listened as I argued against the panel's blind endorsement of the dubious 1995 Spine paper.10 Specifically, I pointed out that the QTF's definition of "recovery" was simply a return to work, not the return to full health (We later critiqued the paper in Spine3 which, unfortunately, did not get the attention we'd hoped for. But I digress.) So it is odd that he and his co-author did not take some issue with it.6

For acute neck pain, there were only three trials. These were SMT or MT only.

No acute trials were found for traction, acupuncture, ultrasound or massage. The scores for these trials were low and thus did not provide convincing evidence. For chronic neck pain, however, 27 reports concerning 25 separate trials were found; again, none for manual traction. For SMT, there were 13 trials. The authors noted that in the past, other authors were looking to compare SMT or MT with other forms of therapy. The present study included several studies not reviewed by these other authors.

One group reported only percentage differences between groups and did not even report the outcome data. Another did not provide intra-group variability measures and did not analyze the degree of intra-group change at all. Their conclusion was: "The evidence did not favor manipulation and/or mobilization done alone or in combination with various other physical medicine agents; when compared to one another, neither was superior." Vernon and Humphreys6 felt it was appropriate to assess the magnitudes of change within each treatment group randomized to receive the therapy of interest.

For the SMT trials, there were no adverse effects. This was something that Gouveia, et al., failed to find in their review.4 All groups demonstrated positive change. In 12 of 13 trials, SMT demonstrated statistically significant or clinically important changes. In their interpretation, the authors noted the possibility of chance improvement, but opined that while 20 mm of improvement on a 100 mm VAS was considered by one authority to be "clinically relevant," chance improvement was very unlikely to exceed 15 mm.

All of the above studies excluded treatment for whiplash. For this condition, which was analyzed separately, there were eight trials. Most of them utilized MT as passive mobilization or massage, but two recent ones report on SMT.1,2 Vernon and Humphreys noted that, up until 2004, there were no studies demonstrating the effect of MT alone for whiplash. This would be reflected in previously developed reviews or guidelines. These two Spanish studies compared SMT to PT and found a strong treatment effect for SMT, including manipulation alone, and concluded that the addition of manipulation accelerated the recovery rate.1,2

The authors also reviewed a number of published guidelines, some of which are for general management and some of which were developed for specific types of practitioners. (My guidelines were not included because they are specifically for whiplash and the purpose of the provision of these guidelines in the paper, I assume, was to contrast those recommending MT or SMT or not recommending it.) All but one of the nine guidelines published since 2003 recommend MT and/or SMT.

I would hasten to point out that two papers published by some colleagues of mine were not reviewed, probably for methodological reasons (they were not randomized, nor were they controlled).7,9 While they may not merit inclusion in the Cochrane Reviews or this paper, both reported very favorable results in a group of chronic CAD cases treated by a chiropractor and are also worthy of owning.

One should conclude from this two-part editorial/review that the current evidence for chiropractic's efficacy for spinal conditions is quite good. In fact, a substantial amount of literature suggests it is superior to the traditional, nonsurgical, medical interventions. Safety is always a concern, and it is clear that spinal manipulation can result in catastrophic injury and even death. The profession should continue to strive to identify risk factors and take appropriate precautions. Fortunately, the risk for serious untoward consequences is much lower than other forms of medical intervention, including even simple pharmaceutical management.

Although, we did consider cost-effectiveness here, this is another area where other studies have shown chiropractic's big advantage. It should be measured not only in the doctor's fees, but in all other costs. This includes pharmaceutical and hospital costs, referrals to other providers, lab and radiographic costs, MRI and other advanced imaging costs, time loss, permanent partial or total disability, cost of vocational rehabilitation, loss of future earning capacity, and measures of quality of life such as the Disability Adjusted Life Years (DALY) and the Quality Adjusted Life Years (QALY).

While many DCs feel immured by forces looking to contain costs by eliminating non-essential complementary and alternative medicine, we should be pushing to demonstrate our real value. And some have been. In a recent study,8 DCs serving as IPA providers in an HMO have demonstrated their value as primary care providers, not as subspecialist CAM providers hoping for a referral from MD gatekeepers.


  1. Fernandez-de-las-Penas C, Fernandez-Carnero J, Fernandez AP, et al. Dorsal manipulation in whiplash injury treatment: a randomized controlled trial. JWRD, 2004;3:55-72.
  2. Fernandez-de-las-Penas C, Fernandez-Carnero J, Palomeque del Cerro L, et al. Manipulative treatment vs. conventional physiotherapy treatment in whiplash injury: a randomized controlled trial. JWRD, 2004;3:73-90.
  3. Freeman MD, Croft AC, Rossignol AM. "Whiplash associated disorders: redefining whiplash and its management." By the Quebec Task Force. A critical evaluation. Spine, 1998;23:1043-9.
  4. Gouveia LO, Castanho P, Ferreiera JJ, et al. Chiropractic manipulation: reasons for concern. Clin Neurol Neurosurg, 2007;109:922-6.
  5. Gross AR, Goldsmith C, Hoving JL, et al. Conservative management of mechanical neck disorders: a systematic review. J Rheumatol, 2007;34:1083-102.
  6. Vernon H, Humphreys BK. Manual therapy for neck pain: an overview of randomized clinical trials and systematic reviews. Europa Medicophysica, 2007;43(1):91-118.
  7. Khan S, Cook J, Gargan M, et al. A symptomatic classification of whiplash injury and the implications for treatment. J Orthoped Med, 1999:21.
  8. Sarnat RL, Winterstein J, Cambron JA. Clinical utilization and cost outcomes from an integrative medicine independent physician association: an additional 3-year update. JMPT, 2007;30:263-9.
  9. Woodward MN, Cook JC, Gargan MF, et al. Chiropractic treatment of chronic 'whiplash' injuries. Injury, 1996;27:643-5.
  10. Spitzer WO, Skovron ML, Salmi LR, et al. Scientific monograph of the Quebec Task Force on Whiplash-Associated Disorders: redefining "whiplash" and its management. Spine, 1995;20(8 Suppl):1-73.

Click here for previous articles by Arthur Croft, DC, MS, MPH, FACO.

Page printed from: