Chapter 111
Placebos
The study caused an outcry. How can you possibly randomize people to sham operations? Medical professional associations questioned the surgeons’ ethical stance, and also the competence of the patients who had agreed to participate in the study. But guess what happened. The operated-on patients felt better, but so did the placebo patients on whom the operation was only simulated. Knee arthroscopy really was ineffective. At the moment, surgery on the shoulder’s rotator cuff is in the same crisis of confidence.
Surgical research has long been mocked as a “comic opera,” because most surgical peer-reviewed publications contain merely a series of individual cases or reproduce only experts’ opinions. Unlike drugs, for which a certain safety and effectiveness must be demonstrated, there are no such requirements for new surgical procedures, and they can be introduced without regulatory oversight. Only about one in five surgical procedures in all of orthopedics is supported by at least one good randomized controlled trial showing that it is better than the nonsurgical alternative. In the 53 studies in which the benefit of surgeries of all kinds was compared with that of sham surgeries, most surgeries themselves were just a scam and no better than the placebo procedures.
A subsequent systematic review with meta-analysis of arthroscopies in middle-aged or older patients with knee pain, with or without osteoarthritis, concluded that there might be a small, temporary, “unimportant” benefit, but the surgery does not outweigh the downsides. Complications of arthroscopies include, in one out of 250 cases, blood clots (deep vein thromboses) that travel to the lungs, cause an infection, and in extremely rare cases lead to death. The particularly tragic irony is that the osteoarthritis patients who underwent an arthroscopy were three times more likely over the next nine years to need an artificial knee joint for the same knee.
Placebos
To be clear: Arthroscopy works. It just doesn’t work better than a sham surgery. But people feel better either way, which is why surgery is also described as the “ultimate placebo.” An arthroscopy helps just as much as a visit to the medicine man. It even has many of the same components—the journey to the place of healing, fasting, and anointing with a purifying liquid (skin disinfection) before facing the masked healer.
It is estimated that, across all different osteoarthritis therapies taken together, about 75 percent of pain reduction, 71 percent of functional improvements, and 83 percent of the reduction in stiffness are attributable to the placebo effect. Then all sorts of sham treatments sprang up like weeds, such as “low-dose” radiation therapy, in which the joint is pulsed with the radiation equivalent of 60 000 chest X-rays. This treatment does work, but no better than if the device is secretly switched off and a recording of the sound of the running machine is played.
There is an extensive literature on the placebo effect. It is hard enough to believe that sugar pills can have a clinical effect, but that’s only the beginning. Two sugar pills work more strongly than one, and green and blue sugar pills have a different effect than red and orange ones. Sugar pills stamped Bayer Aspirin help headaches better than sugar pills stamped generic aspirin, which is not surprising, because patients who were told their pills had cost full price felt a greater improvement than those who were told their pills had been obtained at a discounted price. And a needle can do more than a pill.
The placebo effect of an injection is so strong that, in osteoarthritis patients, a placebo injection relieves pain to a clinically significant degree for three months and produces clinically significant improvements in function and stiffness for six months. Here we arrive at the so-called efficacy paradox. Hyaluronic acid injections are not administered because they do not work better than placebo injections, whereas NSAID medications are administered because they do in fact have an effect beyond the placebo effect. That’s reasonable, but are you ready to hear something crazy? Because needles have a larger placebo effect than pills, injecting hyaluronic acid into the joint—the treatment that is not recommended—works better than the recommended treatment, taking NSAID pills, which would probably also be safer! So why not just stick needles into people if it works so well? We do. It’s called acupuncture.
It is actually possible to test acupuncture against a placebo. There are “sham acupuncture needles” that look and feel exactly like real acupuncture needles. In reality, however, the needle is blunt and retracts into the hollow shaft like a magic trick. If it is then secured with a bandage and sticks out, you can’t tell whether real or faked acupuncture is taking place, entirely without “needling.”
So does acupuncture work for knee osteoarthritis? Yes! But fake acupuncture works too. For this reason, the American Academy of Orthopaedic Surgeons strongly advises against it. But if acupuncture “works” and is relatively harmless, why shouldn’t one use it?
An estimated more than 100 000 Americans die each year from the side effects of prescription drugs alone, making medications one of the most common causes of death. Whatever one may think of doctors who prescribe placebos, at least they don’t kill anyone. If there are conditions like osteoarthritis for which placebos have been shown to be effective, why don’t doctors actively deceive their patients by prescribing them?
They do.
Various surveys of medical personnel found that 17 to 80 percent of physicians and 51 to 100 percent of nurses have already given their patients “pure” placebos—that is, instead of prescribing ineffective antibiotics for a cold, patients received a sham treatment such as an injection of saline solution. There are all kinds of placebos on the market that you can even buy yourself, with brand names like Obecalp (“placebo” backward), Magic Bullet, and Fukitol (no joke!).
Plato even defended the occasionally indispensable lie in medicine, writing in his Republic: “A lie is … useful to men only as a medicine …” Thomas Jefferson called it a “pious fraud.” In the medical literature it is referred to as a “modest scam.” While some doctors contemptuously call placebos quackery, others ask, especially with regard to osteoarthritis: “Why shouldn’t we use them to our advantage?” The American Journal of Medicine published a treatise on the “ethics and practice of placebo therapy,” arguing that “deception is completely moral when it serves the patient’s welfare,” and raised the question of why the “inhabitants of ivory towers denigrate the use of placebos.” Some patients might react sensitively, the treatise warns; therefore “you should never reveal to them that their precious medicine was a joke.”
Weight loss
Twin studies suggest that about half of the osteoarthritis risk is genetic. What can we do about the other half? Fortunately, there are successful methods that manage without capsules, needles, scalpels, or untruths.
Obesity may be the most important modifiable osteoarthritis risk factor, which explains why an examination of thousands of skeletal remains from prehistoric hunter-gatherers to today’s city dwellers showed a dramatic increase in incidence over the last 50 years. Compared with normal-weight individuals (BMI < 25), the incidence of knee osteoarthritis is three times as high in obese individuals (BMI ≥ 30) and as much as five times as high in those with class 2 obesity (BMI ≥ 35). Fat tissue in general, and even directly in the joints—such as in the fat pad under the kneecap—can be a source of pro-inflammatory substances that have been shown to promote cartilage breakdown.