Picture, if you will, a decrepit wooden sailing ship not much different from those that carried the early pioneers. On board the HMS Salisbury, the ship’s doctor is desperate. Twelve sailors are suffering from advanced gum rot, weakness, and malaise—a well-known, frequently fatal condition expected to strike those at sea for more than a few weeks. Since the doctor has few choices, and knowing some of the men are doomed anyway, he splits them into six groups and asks each pair to take a different substance. The pair given oranges and lemons recovered and were able to care for the rest until they returned to dry land.
Later hailed as the father of the systematic clinical trial, that naval surgeon—Scotsman James Lind (1716–1794)—documented his findings in his Treatise of the Scurvy (1753), which eventually contributed both to the British navy’s decision to make lemon juice a mandatory addition to sailors’ diets from 1795 onward and to the evolution of the modern randomized controlled trial (RCT). However, it took the British admiralty 42 years to reach this decision (a year after Lind’s death).

Modern Scientific Experimentation: Its Accidental Birth

Lind’s scurvy experiment is often held up as a significant turning point in the history of medicine, heralding the birth of modern scientific experimentation (and an easier life for sailors). Yet, in fact, he was neither the first to introduce the idea of systematically testing different variables, nor did he perfect the method (See “Evolution of RCTs” on page 50 for more on this). Lind did not even understand the importance of what he discovered when he wrote his treatise.
Even though Lind had devised a valid scientific experiment, he missed the significance of his results and instead interpreted them as an indication of a digestive issue. Lind lacked confidence in his findings,1 which, in part, was responsible for the long delay in authorities realizing the value of his experiments.
Lind’s story is valuable not only because it lit the spark that eventually led to more rigorous scientific experimentation, but also because it is a cautionary tale, since we now know:
• Rigorous scientific evidence must underpin all medical and therapeutic practice.
• We should ensure that all therapeutic measures we apply are based on research.
• The RCT is the gold standard of evidence,2 and published findings should not be questioned.
Is that not the case? Or did I just make a few rather glaring and inaccurate assumptions in these three bullet points? I certainly did, and deliberately so.

Making Assumptions: Naïve to Borderline Dangerous

We make assumptions on a daily basis about a range of issues, and neither the practitioner nor the researcher is immune to them. Assumptions might be relatively harmless (such as my earlier assumption that readers are familiar with scurvy, a potentially fatal disease caused by a severe lack of vitamin C), yet they may also be entirely misguided. Take, for example, the aforementioned assumptions, where I either stated or implied that:
• Readers know and agree that all therapies must be evidence-based.
• RCTs provide the highest grade of evidence.
• The classic RCT performed in a lab is appropriate for the investigation of all types of therapy.
• We must trust researchers to ask the right questions, design studies appropriately, and provide us with the evidence for or against therapies currently in use.
• All published RCTs can be trusted.
• If there is no evidence that something works, then it is the same as saying it doesn’t work.
If these points truly applied, then we would probably have a very different health-care landscape worldwide. As it stands, some of these assumptions are naïve, and some are borderline dangerous.

Proceed with Caution

Overreliance on the RCT has come in for criticism from a variety of quarters across medical disciplines and health-care stakeholders, and its limitations are being more openly debated. Despite this, public health policy, research funding, and public opinion continue to be more strongly influenced by what are perceived as the results of “hard science.” This does not invalidate the RCT as an important tool in research, but it does mean we need to proceed with caution, as it has been noted repeatedly that its purpose is to assist and support clinical decision making—not replace it.
The general consensus within evidence-based decision making suggests a model whereby clinical reasoning, patient preferences, and available research evidence suitably appraised should all be included in responsible practice, with critical appraisal and responsible judgment being applied to weigh them on a case-by-case basis.3 Secondly, it has been established that when exploring the manual therapies, the application of methodology designed for biomedical investigations is problematic in itself and needs to be considered with care.4
It is crucial to avoid extremes. The preceding bullet-point statements are not entirely true, but neither are their opposites, and blind insistence on either side will certainly lead to error. It has been argued that “Randomized trials have developed such high scientific stature and acceptance that they are accorded an almost religious sanctification . . . If relied on exclusively they may be dangerous.”5 This should not be interpreted to mean they should be ignored: the word exclusively is key here. It should go without saying that “rigorous research is key to the professionalization of any health field;”6 but it should be further understood as emphasizing the need to understand and judge the quality of research (a skill known as research literacy), so that this can be responsibly combined with informed clinical judgment.
Before challenging the assumptions of others, we must begin with our own, and it is all too easy to fall into a binary logical fallacy driven more by ideology than rational thought. A more subtle, pragmatic approach is vital. Research trials are both useful and necessary, and this discussion should in no way be interpreted as lessening their value. However, they incorporate a lot of moving parts, and the potential for errors of judgment or interpretation is always present. When focusing specifically on the research within the field of massage and bodywork therapies, there are additional specialized issues to consider.

Four Key Domains of Research for MTs

In light of this, we need to ask ourselves the following: How do we recognize and appraise the difference between a robust study that should lead us to reconsider how we practice and one that should be taken with great caution? How do we balance clinical experience and training with the evidence? Which thought leaders should we trust? How do we appraise the evidence available to us? How do we convincingly participate in a professional discussion of controversial topics where evidence is inconclusive or lacking? How do we educate clients and handle their queries or concerns? How do we establish credibility? And how can we actually contribute to improving the situation?
In the past 20 years, much has been written and discussed about the importance of research literacy and clinical research development in the context of massage therapy. Agendas have been carefully developed, issues clearly identified and articulated, and material produced and disseminated through numerous channels. The topic continues to be frequently raised at professional conventions large and small, and these questions continue to concern not only the massage therapy community but related professions across the spectrum of bodywork and holistic medicine.7
Some useful perspectives are embedded in the Massage Therapy Research Agenda,8 as a result of the specially convened working group (the Massage Research Agenda Workgroup or MRAW) set up in 1999. In the introductory section, the authors outline four key domains of research, each of which includes specific methods and questions, such as:
• Domain 1: Perception of the Field. The first domain covers the perception of the field within itself and by wider society. It includes questions regarding education, epidemiology, and related issues. Domain 1 questions might be “What do patients think of massage therapy?” or “What conditions do people normally seek help for from massage therapists?”
• Domain 2: Practitioner-Patient Rapport and Relationship. The second domain deals with MT practice and technique, dosage, and the therapeutic alliance—the practitioner-patient rapport and relationship. Domain 2 questions might include “What techniques seem to work better when dealing with low-back pain?” or “What frequency and duration of sessions seem to have best results?”
• Domain 3: Comparing MTs to Other Practices. The third domain compares massage therapy to other practices. Domain 3 explores questions such as “How effective is the addition of stretching to Swedish massage techniques?”
• Domain 4: Physiological Mechanisms.The fourth domain looks at the physiological mechanisms. A Domain 4 question might seek to answer the question: “How does it work?”
The first two domains require mostly observational, qualitative retrospective research (looking at existing records or conducting surveys), whereas the last two require a more formal laboratory trial setup.9
In establishing these domains, the authors point out that “Because of the prominence of the biomedical and reductionist models of medical research in our society, there is a strong tendency for people to honor research in Domains 3 and 4 over research in Domains 1 and 2.”
Quantitative laboratory research tends to be considered more important and is usually better funded, and few realize that one needs a deeper understanding of several areas in Domains 1 and 2 before being able to successfully ask questions in the other domains. The reason for this is that the first two domains essentially define and identify the character of the field under investigation, establishing the key players and areas where inquiry is needed. It’s like looking at a full map of a city before zooming in to see the individual neighborhood and street you are trying to reach. You need to orient yourself before becoming too specific, or the danger is that you will miss the mark.

Beware of Confirmation Bias

There’s definite irony in the following statement by Lind: “The mischief done by an attachment to delusive theories and false hypothesis is an affecting truth . . . It is indeed not probable, that a remedy for the scurvy will ever be discovered, from a preconceived hypothesis; or by speculative men in the closet, who never saw the disease, or who have seen, at most, only a few cases of it.”10 Even though he stumbled on the cure for scurvy, Lind was so focused on his own false hypothesis of its cause, thinking scurvy was due to digestive and excretory dysfunction rather than nutrition, that he missed his findings altogether. In modern terms, this is known as confirmation bias. Lind was right, however, in noting that those who speculate with no, or little, clinical experience are unlikely to ask the right questions.
Confirmation bias is as important now as it was in Lind’s day. If we are not asking the right questions, and indeed, if we do not know precisely why we are asking them, then it doesn’t matter how much research we have with the words massage therapy in the title, because it ultimately offers little to the profession itself. That is why it is essential for practitioners to participate actively in the building and appraisal of the research base—in all professions. And, although many inroads have been made since this agenda was published two decades ago (thanks to the dedicated work of leaders in this field), a quick survey of the available research literature confirms that the overwhelming majority of studies involving “massage” all fall within Domains 3 and 4, which are mostly supportive of the effects of massage as a complement to the management of various conditions. Apart from some notable exceptions in specialized journals, though, very few provide substantial support for issues highlighted in the agenda.

No Need to Reinvent the Wheel

There is no need to reinvent the wheel in attempting to address the issues already outlined in the work that has gone before. However, we urgently need to revisit the issues, consider what progress has been made, and—much like those early investigators who realized the value of Lind’s discovery 40 years after the fact—take a fresh look at what we already have, and see what we may be missing.
An astute point made in the MRAW agenda reads: The first research task for MBT should work to uncover the features of MBT practice that make it distinct (Domain 2) and that motivate people to either study it or seek it for health care (Domain 1) . . . MBT needs to know more about itself scientifically before it can accurately design research to compare itself to others or even to fully understand its own mechanism of action.11
The steps toward acquiring research literacy, and from there to research capacity (the ability to conduct research), are many.12 The first step on this path is considering our own assumptions about what we do, why we do it, and considering the questions that arise.

In this series, future columns will provide and examine more detailed explorations of each question raised in this column. The purpose of this article was to provide food for thought—and to note key historical examples that may benefit current inquiry. We will move to the more specific explorations in coming columns.

Homework

I encourage you to explore the endnotes and seek out the original sources—especially the four articles laying out the methodological issues in investigations of massage/bodywork therapy. Then, consider the following questions:
• What assumptions do I hold that are challenged by what I read here?
• How comfortable am I with reconsidering new ideas?
• Where do I feel resistance in considering new ideas?
• Where does my resistance come from?
• What questions remain unanswered, in terms of the way I understand my profession?
Seasoned practitioners may already be familiar with much of this material, in which case it might be valuable to consider how their professional experience has changed in the past couple of decades. Yet, whether a veteran or a more recent professional, all readers may wish to ponder the additional questions:
• What does massage/bodywork research mean for my practice, and how do I feel about it?
• Do I want to contribute to the objectives of the MRAW agenda? Why? Why not?

Evolution of RCTs

Basic forms of randomized controlled testing (of different variables) have been reported from the time of Hippocrates and the ancient Persians. The RCT as we know it today developed through an impressive instance of cross-border transfer of knowledge via psychology investigations evolving in the mid- to late 19th century—when psychology was still in its infancy—across three different countries.
Additional statistical testing was developed in the 1930s, and the notion of placebo in the 1950s. The standardization of the RCT and its use in modern medical experimentation only became a requirement for drug approval by the FDA in 1980.13
At its simplest, the RCT model breaks down into the following:
• Researchers divide the participants into a minimum of two groups. One will receive the treatment being tested, and one will not. The one that receives no treatment is known as the control group.
• The participants do not know which group they are in (single-blinding), nor do the researchers in direct contact with them (double-blinding).
• To avoid the participants in the control group from suspecting anything, they are often given a harmless “false” or sham treatment, known as a placebo. This is to avoid psychological factors affecting the results.
Properly conducted, the RCT is considered the gold standard for scientific research, and must be conducted before a drug, for example, can be approved by the FDA. The higher the quality of an RCT (that provides supporting evidence for a given therapy), the more likely it is to be met with approval by the medical establishment.

Notes

1. Michael Bartholomew, “James Lind’s Treatise of the Scurvy (1753),” Postgraduate Medical Journal 78, no. 925 (December 2002): 695–96, https://doi.org/10.1136/pmj.78.925.695; Iain Milne, “Who Was James Lind, and What Exactly Did He Achieve?,” Journal of the Royal Society of Medicine 105, no. 12 (December 2012): 503–08.
2. Eduardo Hariton and Joseph J. Locascio, “Randomised Controlled Trials—The Gold Standard for Effectiveness Research,” BJOG 125, no. 13 (December 2018): 1716, https://doi.org/10.1111/1471-0528.15199; David S. Jones and Scott H. Podolsky, “The Art of Medicine: The History and Fate of the Gold Standard,” The Lancet 385, no. 9977 (April 2015): 1502–03, https://doi.org/10.1016/S0140-6736(15)60742-5; Mira G. P. Zuidgeest et al., “Series: Pragmatic Trials and Real World Evidence: Paper 1. Introduction,” Journal of Clinical Epidemiology 88 (August 2017): 7–13, https://doi.org/10.1016/j.jclinepi.2016.12.023.
3. P. J. Devereaux and S. Yusuf, “The Evolution of the Randomized Controlled Trial and Its Role in Evidence-Based Decision Making,” Journal of Internal Medicine 254, no. 2 (August 2003): 105–13, https://doi.org/10.1046/j.1365-2796.2003.01201.x; Christopher M. Booth et al., “Evolution of the Randomized Controlled Trial in Oncology Over Three Decades,” Journal of Clinical Oncology 26, no. 33 (November 2008): 5458–64, https://doi.org/10.1200/JCO.2008.16.5456.
4. Claire M. Cassidy and Jacqueline A. Hart, “Methodological Issues in the Scientific Investigation of Massage and Bodywork Therapy: Part I,” Journal of Bodywork and Movement Therapies 7, no. 1 (January 2003): 2–10, https://doi.org/10.1016/S1360-8592(02)00102-X.
5. René G. Favaloro, “Critical Analysis of Coronary Artery Bypass Graft Surgery: A 30-Year Journey,” Journal of the American College of Cardiology 31, no. 4 (1998): 1B–63B, https://doi.org/10.1016/s0735-1097(97)00559-7.
6. Cassidy and Hart, “Methodological Issues in the Scientific Investigation of Massage and Bodywork Therapy: Part I,” page 2.
7. Janet Kahn, Massage Therapy Research Agenda (Evanston: AMTA Foundation, 2002); Glenn M. Hymel, “Advancing Massage Therapy Research Competencies: Dimensions for Thought and Action,” Journal of Bodywork & Movement Therapies 7, no. 3 (July 2003): 194–99, https://doi.org/10.1016/S1360-8592(03)00021-4; Trish Dryden and Rona G. Achilles, Massage Therapy Research Curriculum Kit (Evanston: AMTA Foundation, 2003); Cassidy and Hart, “Methodological Issues in the Scientific Investigation of Massage and Bodywork Therapy: Part I”; Claire M. Cassidy and Jacqueline A. Hart, “Methodological Issues in Investigations of Massage/Bodywork Therapy: Part II,” Journal of Bodywork and Movement Therapies 7, no. 2 (April 2003): 71–9, https://doi.org/10.1016/S1360-8592(02)00125-0; Claire M. Cassidy and Jacqueline A. Hart, “Methodological Issues in Investigations of Massage/Bodywork Therapy: Part III,” Journal of Bodywork and Movement Therapies 7, no. 3 (July 2003): 136–41, https://doi.org/10.1016/S1360-8592(03)00035-4; Claire M. Cassidy and Jacqueline A. Hart, “Methodological Issues in Investigations of Massage/Bodywork Therapy: Part IV,” Journal of Bodywork and Movement Therapies 7, no. 4 (October 2003): 240–50, https://doi.org/10.1016/S1360-8592(03)00067-6.
8. Kahn, Massage Therapy Research Agenda.
9. Cassidy and Hart, “Methodological Issues in Investigations of Massage/Bodywork Therapy: Part I.”
10. Milne, “Who Was James Lind, and What Exactly Did He Achieve?”
11. Cassidy and Hart, “Methodological Issues in Investigations of Massage/Bodywork Therapy: Part I”; Cassidy and Hart, “Methodological Issues in Investigations of Massage/Bodywork Therapy: Part II.”
12. Hymel, Research Methods, 5.
13. Marcia L. Meldrum, “A Brief History of the Randomized Controlled Trial: From Oranges and Lemons to the Gold Standard,” Hematology/Oncology Clinics of North America 14, no. 4 (August 2000): 745–60, https://doi.org/10.1016/S0889-8588(05)70309-9; Devereaux and Yusuf, “The Evolution of the Randomized Controlled Trial and Its Role in Evidence-Based Decision Making”; Bruce E. Wampold and Kuldhir S. Bhati, “Attending to the Omissions: A Historical Examination of Evidence-Based Practice Movements,” Professional Psychology: Research and Practice 35, no. 6 (December 2004): 563–70, https://doi.org/10.1037/0735-7028.35.6.563.