This isn't a new topic for me, I had two posts about it on the old blog. I will summarize my 2013 posts before tackling what has changed in the last decade both in the science and in my understanding. When I wrote the original blogs, there was a social media frenzy of "Placebo Experts" arguing for clinicians to harness this "powerful effect" to improve outcomes. I didn't know then that science communication was not coping well with misinformation on social media and digital communication. That craze mostly died down, but what sprang up in its place is a broken relationship between scientific establishments and the general public. Currently, a huge backlash against institutions which previously arbitrated truth is all the rage; universities, governmental agencies, and mainstream news outlets have been supplanted by podcasts and conspiracy theorists. Another topic for another day!
Let us define placebo and The Placebo Effect, and get a little history.
Defining The Placebo Effect
The placebo effect is a phenomenon in which patients experience a perceived improvement in their condition after receiving treatments with no actual therapeutic value. This effect highlights the power of the mind and the influence of expectations on mental and physical health. This can have clinical value for conditions in which subjective symptoms, like pain, are the biggest measure of outcome. https://www.health.harvard.edu/newsletter_article/the-power-of-the-placebo-effect
Placebos work through:
Expectation: Patients can see they are receiving a treatment, such as taking a pill, getting a shot, receiving a massage; their expectations trigger changes in perception and behavior leading to real improvements in symptoms.
Conditioning: Previous experiences with effective treatments can condition patients to respond positively to placebos, like Pavlov's dogs learned to associate a bell with food.
Brain Chemistry: Placebos can stimulate the release of neurotransmitters, such as hormones, endorphins, and dopamine, which can alleviate pain and improve mood.
Patient-Provider Interaction: A supportive, positive, reassuring interaction with healthcare providers can be a treatment by itself or increase a patient’s belief that they will get better.
Reduction of Anxiety: Anxiety related to symptoms can lead to worsening symptom, getting any treatment may provide relief of anxiety.
Medicine wasn't always modern medicine
Voltaire said "The art of medicine consists in amusing the patient while nature cures the disease." In Voltaire's time (1694-1778), this was absolutely true. Before the 20th century, physician regularly killed patients with "medicines" composed of mercury, sulfur, and lead. They regularly bled patients to "balance their humors" or provided botched/unnecessary surgeries without sanitation or anesthesia. Medicine was still mired deeply in Vitalism, the Four Humors, and Folk Medicine. Because doctors were dangerous, alternative medicine (essentially harmless while keeping the patient occupied) emerged as a safer option. Many of these traditions live on today in Homeopathy, Chiropractic, and Energetic Healing, while Acupuncture and its science-sounding cousin Dry-needling are residual hangovers from Vitalism and Bloodletting. Science and modern medicine have come a long way since then; but how?
It's important to note here that not all placebos are created equally. As Dr. Harriet Hall put it, the evidence suggests that by gaming human expectation a hierarchy of placebo effectiveness emerges:
Placebo surgery works better than placebo injections
Placebo injections work better than placebo pills
Sham acupuncture treatment works better than a placebo pill
Capsules work better than tablets
Big pills work better than small
The more doses a day, the better
The more expensive, the better
The color of the pill makes a difference
Telling the patient, “This will relieve your pain” works better than saying “This might help.”
How we separate medicine from trickery?
If pretend medicine can make you feel better, we cannot rely on feelings and observation alone. Actual medicine has to outperform the sham stuff. Every expert is also a flawed observer; placebo effect, confirmation bias, regression to the mean, belief bias, attentional bias, and a whole host of logical fallacies distort reality for clinician and patient alike. Regardless of education or experience, humans will make observational errors that undermine their search for effective treatments. The more effort, time, and money one spends, the greater the incentive to ignore any conflicting information. This fact places expert opinion right at the bottom of the Evidence Pyramid:
The first step in separating placebos from medicines is to find techniques to control against the confounding factors which obscure results. Most of the story that follows was paraphrased from: Trick or Treatment: The Undeniable Facts about Alternative Medicine by Edzard Ernst and Simon Singh. A really great resource for understanding the difference between medicine and shams.
Dr. James Lind (1716-1794) first identified oranges and lemons as a treatment for scurvy, a disease that affected the outcome of many wars by sickening and killing millions of sailors. Lind designed the first recorded group study to compare six common treatments (cider, sulfuric acid, vinegar, sea water, & a paste consisting of garlic, mustard, radish root, and gun myrrh, and citrus fruit) on sailors with scurvy. It's important to note that the fruit was considered "alternative medicine" at the time and was only included by chance. The medical experts of the day would have likely prescribed bloodletting two or three times over before entertaining nutrition as a mode of treatment.
Since scurvy is a deficiency of vitamin C, those given cider got a little better and those given lemons and oranges got significantly better. A key point here is the five other treatments had previously shown promise when tested individually, but looked positively pathetic when put up against the actual cure for scurvy. The proponents of these treatments, and of bloodletting, had fallen victim to the placebo effect. Placebos will look dramatic when compared to no treatment, but once you line them up in a trial against an effective treatment, their effects seem irrelevant.
This method of taking expert opinions, applying a simple test, then weeding out bad ideas, is a key tenet of the scientific method. An interesting side-story on expert opinion, Dr. Lind didn't understand what he had done, and waited years to report his findings. No one knew about vitamin's in the 18th century. He even tried to create and sell a lemon juice concentrate (processed in a way that destroyed its vitamin C) damaging his own credibility with its ineffectiveness. The human condition must be acknowledged here; we are all fallible and being wrong grants us an opportunity to learn from mistakes. A strong reminder, also, that over-confident health gurus are probably wrong, and likely either don't know or don't care whether they are hurting people.
Massage, Manual Therapy, and Evidence
Massage is deeply rooted in nearly every cultural tradition. "Massage has been around for thousands of years" is the security blanket against criticism. The same tactic is used to defend many forms of alternative medicine. It is meant to demand respect without giving an answer to difficult "how do you know" questions. Manual therapy reasoning, while less ancient, usually starts with "Dr. So-and-so said" and ends with an anecdote or observation from said doctor. Before the inclusion of the Scientific Method, it is safe to assume that tradition was the foundation of clinical reasoning. The handing down of wisdom from master to student made sense when it was the only paradigm we had.
We tend to look back without acknowledging that ancient wisdom was built upon faulty models of disease, health, and physiology. This error in reasoning is so common as to have its own name: The Appeal to Tradition. Finding this appeal doesn't require any work when discussing massage or manual therapy. Modalities almost universally originate from one or two charismatic masters, and are then carried onward by his/her followers who use "Yes, and" to build institutions upon ancient foundation. This origin story usually has a sequel when the best students of that master become masters for their respective school, and are forced to make subtle changes to keep the method alive.
Here is the point where my position last decade needs to be updated. I argued then that because massage (and manual therapy) cannot be easily controlled in an experiment, it was very difficult to say whether the benefit was coming from the interaction (seeing a caring person who does something that feels important) or from the treatment itself. A mountain of evidence that massage is useful shouldn't be waved away so quickly. It remains true that you can find a hundred different flavors of hands-on therapy out there, and that not a single one of those modalities has reliably beat out the others for any known outcome measure. In other words, since you cannot give one group a real massage and one group a fake massage, how can anyone know the difference?
This position ignores the fact that touch is a physical interaction with known benefits by itself, even though skilled touch (defined as touch given by any practitioner with meaningful education on the topic) generally doesn't outperform unskilled touch at least in babies. Robotic touch offers similar physical benefits but fewer mood and mental health benefits. It also ignores that despite the weird rituals in which humans still engage, there is likely an underlying shared mechanism in touch beyond those describe under placebo effects. Conditions like pain, anxiety, and depression respond well to touch at least partially because they are subjective perceptions.
Massage education standards are too low in the United States. Programs are too short, standards are very lax, and education is counted in contact hours rather than years. Massage schools focus on muscle names, origins, and insertions but never touch physiology. They might even teach (though many forget) the finer details of kineseology with a smidge of pathology, but they do not teach clinical reasoning, or how to evaluate evidence. These holes leave plenty of time for a substantial dose of mythology, mysticism, and straight-up bullshit. In 2025, our local massage school proudly posts photos of students learning "energy healing" which not only wastes those students time and money, but also contradicts basic physics and reality.
How I learned to stop worrying and love the process
The feeling that massage education was hopelessly lost ultimately led me to get my doctorate in Physical Therapy. Eight years and tens of thousands of dollars later, my knowledge and scope of practice has grown but very little about my process has changed. In PT we often victim to believing that if an outcome gets better, that means the treatment worked. Hard science of what is in the body (anatomy and kinesiology) will not necessarily help to gain insights into what treatments is best. The classic Hume Is-Ought Problem comes to mind. Seeing my poorly informed colleges handing out useless supplements, chasing every health fad, doling out CBD oil, and giving poorly reasoned advice remains painful. Seeing potential patients seeking ineffective and painful work, believing they need to buy a detox kit, or paying out for expensive placebo treatments is even worse.
When I see any patient, I must address their expectations. They typically seek therapy for a specific ailment (e.g. headache, neck or back pain, anxiety, depression) and desire a specific result (to get better). How do I support the claim that I can help? I point to the scientific evidence. I know many different ways to approach that goal, but how do I know which to choose? I talk to the patient, and we agree on a method that fits their values, lifestyle, and goals. Finally, how to I know the therapy was successful? I measure something relevant when they come in, I measure it frequently during the treatment phase, and I measure it one last time when they are done. Even if some of that improvement is time or placebo effects, the patient has gotten better and I did my part to get them there. None of these things require specific methods, trademarked tools, or proprietary techniques. Good therapy relies heavily on good reasoning.
Comments