SMT/Adjustment-centric practice: Uniting around the evidence, part 4

In the previous parts of this article series, [Part 1] [Part 2] [Part 3] I have argued that, based on a fair interpretation of the peer-reviewed clinical evidence, and of all the possible interventions for non-cancerous, non-infectious, non-traumatic instability, neuro-musculospinal health issues (uncomplicated spinal health issues), the most evidence-based with respect to effectiveness, cost-effectiveness, and safety, is chiropractic thrust SMT/adjustment (and general spinal and overall fitness exercises and healthy lifestyle advice). This is true for interventions within and outsie chiropractic education and scope of practice.

I realize that this is not always the conclusion of systematic reviews and/or clinical guidelines, which often rate the evidence for SMT as similar, or inferior to, other interventions. I contend that this is due to a bias against SMT in clinical studies, in systematic reviews, and in the evidence rating scales such as the JADAD scale, due to a failure to take into account variables which can significantly affect outcomes such as dose of care, timing of outcome measures in relation to end of active care, and differentiation between thrust SMT and non-thrust mobilizations.

Most clinicians simply rely on the conclusions of systematic reviews as they lack the familiarity with the literature, scientific methodological expertise, or time to properly critique such reviews by reading each individual study reviewed, assess its methodological validity, and/or recognize which studies have been excluded based on biased selection criteria. Most clinicians are also unaware that the JADAD and other study quality rating scales do not differentiate between studies that include two or three SMT sessions vs studies that include 8-12 SMT sessions, studies that measure outcomes weeks or months after the last SMT session vs immediately after the last SMT session, or studies that include only non-thrust mobilizations vs thrust SMT.

As Bronfort et al. pointed out in their 2008 review in The Spine Journal regarding the lack of standardization of dose of care and timing of outcome measures in the published studies, “The number of SMT treatments varied from 1 to 24 and follow-up from immediate posttreatment to three years.”1 Inexplicably, and pathetically, little has changed since 2008; the lack of standardization in SMT study methodology makes lumping all these studies together into a single systematic review invalid and using the JADAD scale or any other study quality rating scale cannot correct for this issue.

This lack of standardization of dose of care and follow-up time also creates large heterogeneity of results, which is used to justify downgrading the level of evidence for SMT in systematic reviews. As an example from the recent 2017 Paige et al. systematic review in JAMA, “The quality of evidence was judged as moderate that treatment with SMT was associated with improved pain and function in patients with acute low back pain, which was downgraded from high due to inconsistency of results.”2

In other words, the actual data analysis showed high quality evidence that SMT was associated with improved pain and function, which by the way would make SMT the clear leader in quality of evidence, but the authors chose to downgrade the level of evidence to moderate because of heterogeneity (inconsistency) of results. The most frustrating part is that the authors make no effort to explain this heterogeneity of results based on dose of care, timing of outcome measurement, or thrust vs non-thrust SMT interventions.

As a point of example, the highest rated study in terms of quality in the Paige et al. review is the Hancock et al. study published in the Lancet that concluded that SMT failed to provide any clinically relevant benefit in terms of time to recovery, pain, or disability. This Hancock et al. study is consistently rated as the highest quality study of SMT in systematic reviews. There is no better example of the shortcomings of the JADAD study quality rating scale. The Hancock study receives a score of 8 out of 10 (very high) NOT because it is well designed to answer the questions it poses or that its methodological design or conclusions lack evidence of bias, but because it was well-blinded and thus lacks bias in terms of data collection and analysis, which, is the most highly weighted aspect of the JADAD rating system.

The methodological problems of this study, and the bias of the authors, is clear to see for anyone who understands research methodology and is willing to read the full study rather than just the abstract. First, every patient in the study receives paracetamol 4x/day, including the so-called non-treatment group as well as the NSAID and SMT groups. This is like assessing the effectiveness of paracetamol in a study where every subject receives SMT; it is methodologically absurd. If you want to compare SMT to paracetamol, or SMT to the NSAID diclofenac, or if you want to compare diclofenac to paracetamol, then you do so by having groups that receive one treatment or the other as well as groups that receive placebo, and thencompare outcomes.

Lack of standardization of dose of care and follow-up time creates a large heterogeneity of results….
© joyfotoliakid / Adobe Stock

Second, the SMT group did not receive SMT, they received mobilizations which were labelled as SMT. “Participants allocated to spinal manipulative therapy had treatment two or three times per week [at the physiotherapist’s discretion] to a maximum of 12 treatments over 4 weeks [the average was 2.3 treatments per week].” “Most participants had several low-velocity mobilization techniques [232/239, 97%] with a small proportion also having high-velocity thrust techniques [12/239, 5%].”3

Only 5% of the SMT group received thrust SMT, the other 95% received mobilizations. Most disturbing is that the authors falsely claim that thrust SMT and mobilizations are clinically synonymous and use a bogus citation to support their unfounded claim. “A systematic review of spinal manipulation concluded that there is no evidence that high-velocity spinal manipulation is more effective than low-velocity spinal mobilization, or that the profession of the manipulator affects the effectiveness of treatment.”3

The reference they provide is not a systematic review of spinal manipulation at all and certainly not a systematic review of a comparison of thrust spinal manipulation vs mobilizations or the effects of the profession of the manipulator on outcomes. The study they cite is a review of NSAIDs: van Tulder MW, Scholten RJ, Koes BW, Deyo RA. Nonsteroidal anti-inflammatory drugs for low back pain: a systematic review within the framework of the Cochrane Collaboration Back Review Group. Spine 2000; 25: 2501–13.

Here is what the authors conclude. “If patients have high rates of recovery with baseline care [paracetamol] and no clinically worthwhile benefit from the addition of diclofenac or spinal manipulative therapy, then GPs can manage patients confidently without exposing them to increased risks and costs associated with NSAIDs or spinal manipulative therapy.”3

The authors cite no evidence of increased risk or costs associated with SMT compared to paracetamol or NSAIDs, and they also fail to provide a shred of evidence of “high rates of recovery with baseline [paracetamol] care.” This is understandable: there has never been a shred of evidence for the effectiveness of paracetamol. One wonders how GPs “can manage patients confidently” with an intervention that has zero evidence of effectiveness. I guess they can at least feel confident knowing they are not exposing patients to the increased risk and costs of dangerous and ineffective SMT! Give me a break. The bias against SMT and for an intervention with not a shred of evidence is pathetically obvious, intellectually dishonest, and scientifically unfounded. Keep in mind, this study not only passed peer-review, it is consistently rated as one of the highest quality studies of SMT in systematic reviews.

Contrast this with the systematic review by Hidalgo et al. published in the Journal of Manual and Manipulative Therapy. This is the only systematic review that I am aware of that distinguishes between thrust vs non-thrust SMT. “Two stages of LBP were categorized; combined acute-subacute and chronic. Further sub-classification was made according to MT intervention: MT1 (manipulation); MT2 (mobilization and soft-tissue-techniques); and MT3 (MT1 combined with MT2).4

The authors conclude that thrust manipulation is superior to non-thrust manipulation or mobilizations and to placebo for both acute-subacute and chronic low back pain. “Firstly, in comparison to previous reports of limited evidence showing no difference between true and sham manipulation, the results of this systematic review show moderate to strong evidence for the beneficial effects of [thrust] manipulation in comparison to sham manipulation. “These differences are demonstrated in terms of pain relief, functional improvement, and overall-health and quality of life improvements in the short-term for all stages of LBP.”4 This rating makes thrust SMT the most highly rated intervention for both acute-subacute and chronic low back pain. Imagine if they also controlled for valid doses of care.

It’s not just that many “SMT” studies do not include manipulation, many also include only 1-3 SMT treatments. The results of these studies then get pooled in systematic reviews with the very few studies that include valid doses of thrust SMT care. What do you think this does to the rating of the level of evidence of effectiveness for SMT in systematic reviews? I remind you again, neither dose of care nor timing of outcome measures, nor thrust vs non-thrust SMT is accounted for by the JADAD quality rating scale.

For just a few of many examples: A study published in the Spine Journal in 2016 that concluded that SMT elicited no clinically meaningful benefit for patients with chronic low back pain only included three once-weekly sessions of SMT over a month, and the outcomes were assessed a full month after cessation of care.5 Another study in 2015 published in JAMA included 3-4 sessions of SMT over four weeks, measured outcomes at three months and one year, and concluded SMT elicited no clinically meaningful benefit.6

Not only are two to three SMT treatments not representative of clinical practice, this dose of care is not representative of the peer-reviewed literature. As Haas et al. point out in one of the few valid studies conducted on dose of care, “Therefore 12 sessions of SMT is the current best estimate for use in comparative effectiveness trials.”7

If we want valid answers about the effectiveness of SMT we need to conduct valid studies and write valid systematic reviews with valid selection criteria and quality rating scales. We have failed on all accounts. You will be hard pressed to find more than a very few SMT studies that include a valid dose of thrust SMT care or systematic reviews that take these very significant variables into consideration.

We need to standardize how we study chiropractic SMT. We need to operationally define SMT as high-velocity low amplitude thrust SMT and not allow mobilizations to be defined as SMT. Mobilizations need to be defined and studied as a separate intervention. We also need a standardized frequency of SMT treatments or interventions for a standardized duration. My suggestion would be to make research mimic clinical practice and the conclusions of the Haas et al. study. Why not set a standardized frequency of 3x/week for two weeks, measure outcomes, then, if improvement is shown but not complete, do another period of 3x/wk for two weeks and then remeasure the outcomes.

This would provide a lot of data regarding whether or not, and how much, patients improve in terms of pain and function at two weeks, and, if they are not completely resolved, if a further two weeks of care elicits further improvement. If more improvement has been seen and yet the patients are not fully resolved, then it would be easy to add another two weeks of care on an ongoing basis until symptoms are resolved and function is restored.

With properly controlled studies with a placebo or sham manipulation group this would provide very important data. Any comparators should be for the same duration. If we are going to compare chiropractic SMT to drugs or any other comparator, then the dose of care and timing of data collection must be standardized and made equal between the groups.

It would not be difficult to standardize frequency and duration of care for studies, and to make mandatory inclusion of a control or sham SMT group and standardized duration of any comparators. This is research methodology 101; I cannot understand why we have not done this. Every systematic review on SMT laments that few published studies meet selection criteria with respect to validity, yet we keep funding and conducting low quality studies. This is a waste of valuable limited resources and leads to more heterogeneity of results, which leads to invalid downgrading of the level of evidence for SMT.

There is so much more to discuss but space will not allow. Instead I will conclude with a plea for the standardization of SMT studies with respect to dose of care, timing of data collection, and thrust SMT vs non-thrust mobilizations and, in the meantime, for systematic reviews to address these issues of validity in their selection criteria and data analysis.

It is clinically and scientifically absurd to continue to conduct and/or pool together in systematic reviews, data from studies that have, “The number of SMT treatments varied from 1 to 24 and follow-up from immediate posttreatment to three years,” or that include mobilizations instead of thrust manipulations but are classified as SMT studies.

REFRENCES

1. Bronfort et al. Evidence-informed management of chronic low back pain with spinal manipulation and mobilization. The Spine Journal 2008 (8): 213-225.
2. Paige et al. (2017) Association of Spinal Manipulative Therapy with Clinical Benefit and Harm for Acute Low Back Pain. Systematic Review and Meta-analysis. JAMA;317(14):1451-1460
3. Hancock, M.J. et al. (2007) Assessment of diclofenac [Voltaren] or spinal manipulative therapy, or both, in addition to recommended first-line treatment for acute low back pain: a randomized controlled trial. Lancet. 2007;370:1638-1643
4. Hidalgo, et al. (2014). The efficacy of manual therapy and exercise for different stages of non-specific low back pain: an update of systematic reviews. Journal of Manual and Manipulative Therapy 22(2): 59-74
5. Castro-Sanchez, A.M. et al. (2016) Short-term effectiveness of spinal manipulation therapy versus functional technique in patients with chronic nonspecific low back pain: a pragmatic randomized controlled trial. The Spine Journal 16: 302-312.
6. Fritz, J.M. et al. (2015) Early physical therapy vs usual care in patients with recent-onset low back pain: A randomized clinical trial. JAMA 314 (14): 1459-1467.
7. Haas, M. et al. (2014) Dose-Response and Efficacy of Spinal Manipulation for Care of Chronic Low Back Pain: A Randomized Controlled Trial Spine J. 2014 July 1; 14(7): 1106–1116.

Dr. James Chesnut B.Ed, M.Sc, DC, C.C.W.P., recently developed the Evidence-Based Chiropractic and Lifestyle Clinical Protocols which include evidence-based spinal health exams, reports, and patient education. He also developed, wrote the texts, and still teaches the Evidence-Based Chiropractic and Lifestyle post-graduate certification program accredited through the International Chiropractors Association (ICA).