Originally published in Massage Today
September, 2007, Vol. 07, Issue 09

We live in an age of evidence-based medicine, but what does this mean? What it does not mean – although this is a common misconception – is that everything you do therapeutically has to be based on scientific research that validates its effectiveness and safety.

Although in some instances, this may well be part of the foundation for clinical choices.

More reasonably, Strong, et al,1 suggest that: “Evidence-based practice should incorporate evidence of the effectiveness of interventions from research (both quantitative and qualitative – sometimes termed empirical evidence), together with information about the client’s needs and goals, as well as the therapist’s (i.e., your) clinical experience.” (Italics added)

Strong, et al, suggest there are key steps you need to follow in order to apply the principles of evidence-based practice. These steps involve you gathering and critically assessing information and evidence based on research relevant to the problems of your patient/client. This should be followed by an integration of the gathered evidence with your own clinical experience, along with consideration of the needs and expectations of the client.

Out of this mix of evidence and experience a therapeutic program or plan should emerge. This accurately could be described as an evidence-based plan. Let’s consider the hierarchy of importance that commonly is given to different levels of evidence, something that causes concern for many.2 (See Box)

It is worth reflecting that a great deal of the methodology employed in complementary and manual therapy is relatively un-researched. This means that a ranking of “2” should not be seen as a suggestion that the method should not be utilized only that further study is called for.

Tonelli3 has argued that: “Proponents of evidence-based medicine have made a conceptual error by grouping knowledge derived from clinical experience and physiologic rationale under the heading of ‘evidence,’ and then have compounded the error by developing hierarchies of ‘evidence’ that relegate these forms of medical knowledge to the lowest rungs. Empirical evidence, when it exists, is viewed as the ‘best’ evidence on which to make a clinical decision, superseding clinical experience and physiologic rationale. The relative weight given to each of these areas is not predetermined, but varies from case to case.”

 Categories of Evidence

  1. Good: Empirical evidence based on systematic
    reviews (meta-analyses).
  2. Probably useful: Based on evidence from one or more
    randomized and/or controlled trials (RCT).
  3. Possibly useful: Based on some evidence from RCTs, with inconclusive or contradictory outcomes, or with methods open to question.
  4. Opinion: Practitioner conviction, expert view, clinical experience but without reliable research evidence.
  5. Poor: Rumour, traditional use, with effectiveness doubted, or research evidence suggesting ineffective or risky.
  6. Negative: Research derived evidence suggests that this
    therapy is not useful in the condition.

Tonelli, therefore, logically maintains that these are different kinds of evidence, with one being no more important than the other in any given case. In other words, clinical experience isdifferent from research evidence, not inferior to it.

Unless evidence clearly shows that a form of treatment is potentially harmful, the weight given to these forms of evidence may be equal, or experience may be more important than evidence. It is worth reminding ourselves that lack of proof of efficacy does not represent proof of a lack of efficacy. Safety, however, is non-negotiable and if there is any evidence suggesting that a form of treatment is contraindicated, this should be the deciding piece of evidence that guides you.

Low Back Pain (LBP) Example

Taking low back pain disorders as a common and easily comprehended example. A review of the literature demonstrates there is no universally applicable technique, method, approach or modality that will always be helpful in restoring pain-free functionality, since the causes and features of the condition (LBP in this instance) are anything but uniform.

An individualized approach is demanded since two seemingly identical sets of symptoms might have completely different etiological and aggravating features, and these would benefit from quite different therapeutic and rehabilitation strategies. One possibly requiring deactivation of mysofascial trigger points, followed by postural re-education, with the other calling for joint mobilization or manipulation, supported by appropriate soft-tissue treatment, possibly involving stretching and/or core stability training.

Evidence4 shows that manipulation of an “exercise sensitive” class of back pain would offer little benefit, and likewise, specific exercises are unlikely to help “manipulation sensitive” back problems. Of course, there also are back problems that are unlikely to respond to either manipulation or exercise and some that may respond to both.

Based on your patient’s presenting symptoms and history, along with your assessment, you might therefore conclude – having considered published research evidence – that ideally this individual’s problems could be assisted by one or other of the options mentioned above: trigger-point deactivation and postural enhancement via exercise, or joint mobilization assisted by soft-tissue treatment, or by something else altogether. Whichever conclusion you reach, based on the evidence, would fulfill part 1 of your task. Marrying this with clinical experience would be the next stage.

Since by definition pain almost always is a feature of LBP, in many instances accompanied by functional restriction, methods and techniques that are likely to modulate both features (pain and restriction), also are needed. Massage therapy according to research5,6 – and possibly backed by your experience – is almost universally useful in cases of back pain and so should be able to form at least part of your evidence-based decision making as to a therapeutic plan.

Of course massage is capable of being deep or shallow, rapid and stimulating or slow and relaxing, incorporating specific attention to dysfunctional features, or of being completely non-invasive in order to offer a period of calm. These variables make it necessary to qualify what form of massage should be employed in any given case, and this too should be decision based on research evidence, if it exists. And your clinical judgement should be based on experience, as well as your understanding of the pathophysiology of the condition.

One key to successful outcomes in treatment involving back pain, or anything else, lies in identification of the processes that are operating, and as best possible, recognizing underlying causative and maintaining features.

Is there spasm? Are the tissues fibrotic? Are there active trigger points influencing the condition? Is there an underlying emotional feature? Is this back pain one that will ease on its own in a few days with or without treatment? Or is this something that needs further evaluation, with immediate input from you restricted to relaxing the individual?

Hopefully, the opinions relative to evidence, offered in this brief review, will help to re-establish clinical experience as a central part of the evidence we use when deciding on how best to assist our patients.


  1. Strong J, et al. Pain-a textbook for therapists. Edinburgh: Churchill Livingstone, 2002.
  2. Rosner A. Fables or foibles: inherent problems with RCTs, J Manipulative Physiol Ther,2003;26:460-7.
  3. Tonelli M. The limits of Evidence-Based Medicine. Respiratory Care, 2001;46(12):1435-40.
  4. Flynn T, Fritz J, Whitman J, et al. A clinical prediction rule for classifying patients with low
    back pain who demonstrate short-term improvement with spinal manipulation. Spine, 2002;27:2835-43.
  5. Ernst E. Massage therapy for low back pain. A systematic review. J Pain Symptom Management, 1999;17:65-9.
  6. Furlan A, et al. Massage for low back pain. Cochrane Database Syst Rev, CD001929, 2000.