(originally published in Fall 2016 of Massage & Fitness Magazine)
The worldview of massage therapy as a profession seems to be shifting. There is growing emphasis in our community for “evidence-based practice” and a growing desire by members of the profession to seek out evidence for their treatment protocols of choice. We see the term “evidence-based practice” — and sometimes its cousin “science-based practice”— pops up often in massage therapy forums. It’s not uncommon to see massage therapists sharing research from various sources in their discussions. This shift is a decidedly good thing; as healthcare providers, we should strive to critically evaluate our own practices to ensure the best quality of care possible for our patients.
Outcome studies serve to measure the outcomes of a particular treatment protocol or comparing several treatment protocols, ideally against control groups and shams treatment groups. In other words, they are focusing exclusively on the observable results of the treatment(s) for patients experiencing a particular impairment or condition.
Mechanism of action studies focus more on whether a particular physiological response, which may plausibly cause or strongly influence the outcome, is observed during or after a treatment. Examples include the increased or decreased presence of a particular hormone in the bloodstream, or greater or lesser local blood volume in the anatomical area being treated. The idea is to see if a particular response exists, is reliably and measurably initiated or affected by the treatment, and could plausibly be responsible for the outcome.
The outcome of a treatment, and the proposed mechanism of action behind the outcome, are two completely distinct phenomena.
It’s entirely possible to be wrong about the proposed mechanism of action behind a treatment and still have positive patient outcomes. For example, a therapist might believe a treatment “removed toxins”, which are causing the patient’s symptoms. Many of their patients may report feeling better after a treatment, even if “toxins” weren’t plausibly involved at all. The positive outcome isn’t reliant on that specific mechanism of action being accurate – it just means the actual mechanism (or combination of mechanisms) is very likely something else. The outcomes of the treatment don’t validate the proposed mechanism of action for that treatment.
In a manual therapy setting, we can only observe outcomes. We can collect information on how our patients report feeling and moving after treatment. We don’t have the tools or the skills to measure mechanisms of action behind the treatment, since the vast majority aren’t directly observable on a macroscopic level without specialized equipment. We really only get half the picture.
Dr. Harriett Hall, a retired M.D. and well-known skepticism and critical thinking advocate, coined the expression “Tooth Fairy Science” to describe doing research on a phenomenon before that phenomenon is shown to exist.
The argument goes: “You could measure how much money the Tooth Fairy leaves under the pillow, whether she leaves more cash for the first or last tooth, whether the payoff is greater if you leave the tooth in a plastic baggie versus wrapped in Kleenex. You can get all kinds of good data that is reproducible and statistically significant. Yes, you have learned something. But you haven’t learned what you think you’ve learned because you haven’t bothered to establish whether the Tooth Fairy really exists.”
This argument is a good metaphor of conflating outcomes with mechanisms of action. The money received is the observable outcome, and all the data being collected relates to that outcome. None of the data shows that the proposed mechanism of action, the tooth fairy, actually exists and is responsible for the outcome. We can only reliably gather data on the outcome, there’s nothing in the outcome alone that justifies a particular mechanism.
Many people (massage therapists included) tend to point to outcomes studies as evidence of that a particular mechanism of action is accurate. For instance, pointing to a study on Reiki outcomes as evidence that the “universal life energy” Reiki proposes must exist in some format. This is akin to pointing the money under the pillow and saying “so the tooth fairy must exist!”
This is especially prevalent when the actual mechanism of action is unknown, which is often the case in health and medicine. Sometimes we don’t know why a particular therapy had (or has) an effect, or even whether the therapy itself is responsible for the outcome. However, this unknown doesn’t make any implausible or unproven mechanism any more likely to be true, and so we shouldn’t assume them to be true. If we don’t know where the money comes from, that can’t be used as evidence the Tooth Fairy really exists insofar as science goes.
Hall’s “Tooth Fairy” metaphor is great at demonstrating the need for distinction between outcomes and mechanisms of action, insofar as research goes. However, in practice, the patient’s goal is arguably always the outcome, not the mechanism. They care about the money earned, not whether the Tooth Fairy exists, or however else the money got there.
This is why outcomes studies and measurements are still very important in manual therapy (and other forms of healthcare), even in the absence of a definite mechanism of action. If a particular therapy demonstrates positive outcomes (superior to sham, placebo and no-treatment) in research and with patients in a real-world setting, there is no reason it should be discontinued on that limitation alone.
A lack of evidence for a particular mechanism does change the way we should communicate with our patients about the therapy, though. We need to be honest with our patients. Not only is it the ethical thing to do, but informed consent requires that we provide accurate information about the therapy insofar as is possible. Sometimes, that means telling a patient that while we have had good outcomes with a particular treatment protocol (which is ideally backed up by research), that we aren’t sure why the outcomes occurred.
It’s OKAY to tell a patient that the mechanism of action behind a treatment is unknown or hasn’t been proven empirically. What’s NOT okay is telling a patient a particular mechanism of action is responsible for the outcome when that mechanism is unknown or not reliably proven to be empirically measurable.
We also have to factor in the lack of empirical evidence for a mechanism against the risks inherent to the treatment. For instance, if a particular therapy has a risk of injuring a patient or being a significant burden to them, financially or otherwise, the need for justification for applying that therapy is greater. If there are comparable outcomes using treatments which have a lesser risk of harm, and with better understood mechanisms of action, it is often more appropriate to apply those treatments instead. Sometimes that means recommending no treatment, or a free home care activity, or otherwise.
Why Studying Mechanism of Action Matters
It’s true that our patients care about outcomes more than mechanisms of action (in most cases). However, researching and understanding the most plausible mechanism of action (or combinations of mechanisms, there isn’t always just one) whenever possible is important for several reasons. The narrative we share with our patients should be as accurate as possible. If a mechanism of action is known and well- researched, it may be important to share the mechanism with the patient, especially if there are any risks associated with that mechanism.
Complicated health histories and current medications may play a role in the outcomes of a treatment. Understanding how the treatment works can help us predict how that treatment may interact with other forms of therapy the patient is undergoing, and whether cautions or contraindications are in effect. If we are basing our understanding on mechanisms which are implausible, we may alter treatments unnecessarily, or miss something that puts the client at risk.
If we understand the underlying mechanism(s) of a treatment, we can potentially apply those principles in other forms of therapy and home care to maximize the outcomes of those activities. Having a greater understanding of how our therapies work also helps us in dealing with other health care providers, insurance companies, and other stakeholders in massage therapy. It also helps us to create a larger body of valid research for our own practice, as well as lobbying efforts.
Don’t Conflate Outcomes and Mechanisms
Before sharing that next piece of research or making a claim about how a treatment works, think about whether the author (or you) are conflating the outcome of a treatment with the mechanism of action. If a piece of research does make a claim about how a therapy works, read through the paper in detail and see whether they can support that claim reliably, or if it is based on assumption.
And remember, it’s OKAY if we don’t know why a particular outcome happens!