Thinking about placebos, and healing. Thank you, Dad
Patients and doctors want to know which treatments “really work”. That is not too difficult when dealing with something mechanical, like fixing a broken bone, stopping bleeding or removing a stone. The problem comes when trying to relieve pain when it has no identifiable and fixable cause. Many patients with headaches or back pains – or stomach pains – are faced with a bewildering variety of treatments promoted by traditional and alternative therapists.
The gold standard scientific test of efficacy is the randomized controlled trial. Patients are assigned, with their consent, to receive the active treatment or a placebo (such as a “sugar pill”). They are “blinded” to the allocation (they don’t know whether they got the active or placebo pill), and followed to see if the clinical problem resolves more often in the patients receiving the active treatment.
Some may be surprised to know that this process has also been used to evaluate surgical treatments when they are controversial. Patients receive the active surgery or a “sham” procedure, which to them looks and feels identical. They get the anesthesia and skin incisions, but nothing more. The ethics and feasibility of such “sham surgery” studies are debated, but they do produce interesting results. A recent big study in patients with painful knees showed that merely looking in (with arthroscopy) was as effective as also mending the ligaments.
I have learned a lot about all this in recent years. If I have any readers (thank you), some will know that my medical career focused on a procedure called ERCP. We pass a flexible endoscope through the mouth down past the stomach to the “ampulla of Vater”, where the bile and pancreatic ducts drain their digestive juices. We can then pass instruments through the ampulla to detect diseases and to treat some of them. A good example is removing stones from the bile duct, a technique first described in 1974 from Germany and Japan. The stones are grasped and removed with a little basket after opening up the ampulla by cutting the surrounding sphincter muscle.
This cutting procedure, called sphincterotomy, is used very widely, and for many reasons. One is “sphincter of Oddi dysfunction” (SOD), a condition often suspected in patients with pain after having the gallbladder removed (cholecystectomy). The concept is that the ampullary sphincter muscle that controls the flow of bile does not relax, or goes into spasm. This results in back pressure in the duct, and attacks of pain. It is therefore logical to believe that the problem could be relieved by performing a sphincterotomy. We did a lot of those procedures. Many patients were happy, but not all, and some suffered complications, especially post-procedure pancreatitis.
More than 10 years ago we started a research project designed to find out if we could better predict which of those patients would respond to sphincterotomy, or not, in order to inform future practice. We thought it might be the precise nature of the pain, the patient’s psychosocial status or the pressure in the sphincter (we can measure that during ERCP).
For many reasons, simply looking into these and other factors in existing patients does not give clear answers. There are many biases, which is why we chose to introduce the placebo/sham element in a randomized trial.
With lots of care, patient education, institutional review board approval, and funding from the National Institutes of Health, we launched the EPISOD study (Evaluating Predictors and Interventions in SOD). This involved patients with “SOD type III”. They have pain post-cholecystectomy, but no objective evidence of biliary obstruction, such as elevated liver tests. All of the patents underwent the ERCP procedure, with measurement of the sphincter pressure. Then, 2/3 of the patients got the sphincterotomy, and the rest had no active treatment (sham). The patients and those caring for them afterwards, and the researchers, were all blinded to the intervention. No one knew which patients had had the active treatment. After one year we were very surprised to discover that the sham-treated patients did as well as those who had undergone sphincterotomy. Indeed, at 5 years, they were actually doing better!
Publication of these important results has hopefully changed practice (1). It is surely no longer appropriate to offer ERCP/sphincterotomy to those patients. We also wonder about the value of sphincterotomy in some other contexts, such as SOD type II, and idiopathic recurrent pancreatitis. Research is ongoing.
Having said all that, I come to my reason for venting today. Why on earth did the sham-treated patients do so well? Half of them had little or no pain 5 years after having “no treatment”.
I had always thought of placebos as inactive, with maybe only some temporary effect. Delving into the placebo literature revealed the extent of my ignorance, and made me think much more about the essence of “healing”. The literature is huge. I recommend reviewing papers by WB Jonas (2). It is abundantly clear that the results of any treatment, active or sham, is greatly influenced by the environment in which is given, by whom and how, and the patient’s expectations.
In his marvelous book “How healing works” (3), Jonas describes patients recovering from various severe disabilities after treatments which had no discoverable effectiveness, and records the similarities in these phenomena across cultures and religions. He quotes an orthopedic surgeon called Green, who might have been thinking about referral to an ERCP expert like me when comparing surgeons with shamans, when stating: “Shamanistic healing measures include: journeying to a healing place, fasting, wearing ritual garb, ingesting psychotropic substances, anointment with purifying liquid, an encounter with a masked healer, and inhaling stupefactive vapors. These steps are followed by a central ritual activity that may include extracorporeal, surface, and penetrative components. Postoperative ritual activities reinforce the suggestive value of the healing. These experiences increase a patient’s suggestibility, thereby enhancing the likelihood of a favorable outcome”.
So, it is clear that the performance is all-important, how treatments are “sold”. This is a slippery slope that charlatans travel with great profit. But it is also the essence of “doctoring”, how we interact with our patients, whatever the context. I learned much of that from my Dad, a country doctor, going with him on house calls a long time ago. I watched him sit with patients and their families, hold their hands, listen quietly, and offer comfort and reassurance. He had very little else to offer in the 1940s. That was before antibiotics.
All of this makes me wonder about my own practice. How much of the (apparent) success of my thousands of ERCP procedures was due to the clever procedures, and how much to my Dad’s teaching?
I conclude with a favorite quote from another mentor, Solly Marks, the grandfather of gastroenterology in Cape town. After hearing me pontificate about my procedures, he said “Peter, remember to deal with the whole patient, not just the hole in the patient”. How wise.
Thanks for sticking with me, if you have. Comments are welcome.
- Cotton PB, Durkalski V, Romagnuolo J, et al. A multicenter, randomized trial of endoscopic sphincterotomy for suspected sphincter of Oddi dysfunction in patients with pain after cholecystectomy – the EPISOD trial. JAMA 2014;311:2101-09.
- Jonas WB, Crawford C, Colloca L, et al. To what extent are surgery and invasive procedures effective beyond a placebo response? A systematic review with meta-analysis of randomized, sham controlled trials BMJ Open 2015;5:e009655.
- How Healing Works: Get Well and Stay Well Using Your Hidden Power to Heal. Jonas WB 2018. Available at Amazon.com