Whether you think you can, or whether you think you can’t, you’re right.
—Henry Ford
Professor George Lewith is a family doctor who shares my interest in placebo research. He is also one of the nicest and most helpful people I know. Soon after we first met at a conference, I decided to ask him for some advice about a back problem that started when I was a rower. It was not an emergency, so there was no need for me to go to the hospital, but it was Saturday afternoon and I didn’t want to disturb George on the weekend. I didn’t think he would take the call, so I called his mobile number with the intention of leaving him a message. I was surprised when he answered.
“Great to hear from you, Jeremy,” he said. “How are you?”
“Very well, thanks, how are you?”
“I’m really happy, because I’m at the hospital where my grandson has just been born,” he answered.
Shit, I thought, not only did I call him on the weekend, but I interrupted his special family time. I took a deep breath and said, “Let me call you back, because my question is not very important.” But he insisted that he had time to talk, listened to my question, and gave me a helpful answer. I thanked him, thinking what a nice guy he was.
“It’s my pleasure,” he said. Then he added, “Feel free to call me anytime you have a question.” And he meant it, because I have since called him whenever I have a medical question and he always takes the time to answer.
George’s colleagues know that he is very nice to his patients, too. But some of them think that spending so much time and energy being pleasant and positive is a waste, and that he should focus on prescribing the right drugs.
One of his physician colleagues, Bruce Thomas, told George that his kindness might give patients a warm, fuzzy feeling, but it didn’t make them get better. Dr. Thomas knew all about pathology and physiology, but he clearly didn’t believe that the George’s bedside manner was all that important. In 1987, George challenged Bruce to try being positive with his patients to see what happened. Dr. Thomas accepted the challenge, presumably thinking he would prove George was wrong.
For his trial, Dr. Thomas prepared a series of cards and wrote positive on half of them and negative on the other half. Then he shuffled the cards and put them in a drawer in his desk. As patients came to visit him, he first diagnosed them to see whether he believed they had a life-threatening ailment that needed referral to a specialist or an ambulance. If not, he reached into his drawer to draw a card.
Dr. Thomas did this until he had treated one hundred patients with negative consultations and one hundred patients with positive consultations. He found that 64 percent of the patients who received positive consultations got better within two weeks, whereas only 39 percent of the patients with negative consultations got better. That means patients who left their consultation with firm, confident diagnoses were about twice as likely to get better than those whom Dr. Thomas had left with less confidence about their prospects. Dr. Thomas called the resulting research paper “Is There a Point in Being Positive?”
One problem with Dr. Thomas’s study was that the outcomes were subjective, because it was up to the patients to report whether they got better or not. It is possible that some of the patients might have said they got better, when in fact they didn’t feel any better. After Dr. Thomas told them with so much certainty that they would feel better, they might have felt scared to contradict him. Yet dozens of studies conducted since then with more robust and objective outcomes have confirmed that there is a point in being positive.
Fabrizio Benedetti, a famous placebo researcher based in Italy, compared what he called open and hidden treatments. Patients treated openly knew they were getting a treatment and so expected to get better. Patients in the hidden group didn’t know they were being treated and so didn’t expect to get better. Normally you cannot give a treatment to a patient without their knowledge, but Benedetti thought of a clever way to get around that problem.
In the hospital where he works, a lot of thoracotomy operations are performed. A thoracotomy involves cutting through the chest wall to access the heart and lungs. Thoracotomies are usually performed to remove tumors, to confirm diagnosis of a lung or chest disease, to reinflate collapsed tissue, or to remove fluid from the chest. Forty-two such patients were already connected intravenously to fluid bags at the hospital.
All the patients in Benedetti’s study received morphine through the preconnected intravenous line, but only half of them were told that they were being given morphine. Doctors informed these patients that they were receiving a powerful painkiller, before telling them their pain would go away within a few minutes. The remaining twenty-one patients were not told they were receiving the same morphine dose, and the use of programmable intravenous devices meant no doctor or nurses entered their room.
Patients in both groups were asked to rate their pain in a diary thirty minutes and sixty minutes after receiving the morphine, using a ten-point scale on which zero meant no pain and ten meant unbearable pain. The morphine delivered by a doctor who told patients they were receiving a powerful painkiller was 20 percent more effective at reducing pain than the hidden administration of morphine.
Benedetti’s clever study shows that positive messages from doctors do more than make patients feel better in a subjective, fluffy way. They work to make the patient actually get better. Using the same open-versus-hidden treatment design, Benedetti and his colleagues hooked ten patients with Parkinson’s disease up to electrodes. They then underwent a procedure called deep brain stimulation (DBS), which is used to help counter the neurological symptoms of the condition. The procedure involves electrical impulses being sent to parts of the brain that control movement to block abnormal nerve signals.
Half the study participants were told when the electrodes were turned on and half were not. For some reason, researchers often like to use complicated words for simple things. The slowing down of movements in people with Parkinson’s is called bradykinesia. Before and after their treatment, those in the study underwent a test commonly used to assess bradykinesia in Parkinson’s patients. They were asked to move their index fingers to a target in response to a light being switched on. The patients with the open stimulation—those told they were receiving DBS—improved their reaction times by twice as much as the patients who received hidden treatment.
In 2017, I conducted a systematic review of sixteen randomized trials on the effects of positive expectations on treatment for pain. These involved almost 4,000 patients, all of whom went through tests similar to the one Dr. Thomas did to challenge George’s belief.
I concluded that positive messages reduce pain by as much as or more than most common drugs for treating pain, but without the side effects. The doctors’ positive messages also reduced nausea, asthma symptoms, anxiety, and depression. The other thing I found is that positive messages didn’t just affect “soft,” “psychological” outcomes but could affect physical outcomes, too. Other trials have shown that positive thinking can improve lung activity in asthma patients and how much blood flows through the arteries of patients with chest pain. The positive messages also reduced the amount of medication patients used and reduced their length of stay in the hospital. But how do positive expectations work?
Positive messages can lead to health benefits in many ways. The first is through the reduction of stress and anxiety. If you think something really bad is going to happen, like your cold might be serious pneumonia, your headache might be a tumor, or your back pain will paralyze you, you are likely to become anxious. As we saw earlier, anxiety and stress increase the risk of suffering heart disease and many other conditions. Several trials also show that depression is more likely and pain levels higher in those suffering from anxiety. This seems to be because anxiety causes the release of a hormone called cholecytoskinin, which increases the intensity of pain messages from the body to the brain. If a doctor reassures you that everything is going to be okay (which it usually is) or that they have a plan (if it is more serious), your anxiety is likely to be reduced, and this can reduce your pain.
Imagine you are a caveman just back from your very first buffalo hunt. You noticed that the caveman whose spear hit the buffalo got a “high five” (or whatever the caveman equivalent was). You like the idea of getting a high five, but the next hunt is more difficult. You get close to your target, but the herd hears you and stampedes away. You have to circle a long way around to approach them from downwind. After a day of walking with barely any food or water, you are tired. Your head sags, your feet are bleeding, and your body is telling you to lie down and sleep. The only thing that keeps you going is the group of older, more experienced hunters, who keep pushing you forward. After what seems like many hours of walking, you look up to see the buffalo within striking distance. Remembering your desire for the high five, you get a surge of energy. You forget you are tired, sneak quietly up to the fearsome animals, and launch your spear. Hit in the neck, a buffalo tries to run but stumbles, falls, and dies. Triumphant, you accept your hard-won high five and experience another energy boost.
The energy boosts you just received came from the brain’s reward system, which basically makes us feel good when we do things that help us survive. Things like getting food, earning money, or engaging in activities that help you produce healthy offspring are good for your survival. One way the brain rewards you is by increasing the amount of dopamine in your body. Dopamine feels good and makes you want to do more stuff that increases your survival prospects. You can also get that good feeling by thinking about something good happening. That is one reason why the caveman got a rush when he got near the buffalo and thought about actually killing it.
If dopamine sounds like the kind of thing you can get addicted to, that’s because it is. It is the reason we can become addicted to things like food, sex, and money. While these things are good for our survival up to a point, they are detrimental beyond that point. And if we are honest, most of us indulge in at least one of these a bit too much. We often eat past the point we are full, work past the point where we have earned enough money, and keep lusting beyond the point where we have had enough. Many illegal recreational drugs target the reward system. Cocaine, for example, boosts dopamine levels in your body by preventing the brain from reabsorbing it.
Dopamine also explains why a positive expectation reduced Parkinson’s symptoms in Benedetti’s study. One cause of Parkinson’s disease is that the part of the brain that produces dopamine stops working properly. In one study, patients with the condition were given either apomorphine, which increases dopamine levels, or a placebo, but were not told which group they were in. The results were striking: dopamine activity in both groups increased 200 percent. The patients given the placebo thought they were getting the real drug, which caused them to expect a positive reward and increased the amount of dopamine in their bodies. Drugs for Parkinson’s often work in precisely the same way.
Pharmaceutical therapies for some other conditions also work by increasing dopamine levels. For example, the psychostimulants used to treat ADHD increase dopamine and norepinephrine levels in the brain. (Norepinephrine is a substance that causes your body to “wake up” by raising blood flow to muscles, increasing blood pressure, and releasing glucose for energy.)
Dopamine isn’t the only drug your body produces when you think positive thoughts or when doctors tell you to expect success.
The expectation of a positive reward also produces endorphins, which are a natural form of the painkilling drug morphine. In another one of his cool studies, Fabrizio Benedetti gave people placebo treatments that he said were powerful, giving them a positive expectation. Along with the placebo treatments, he gave some of them a drug called naloxone. Naloxone stops morphine and endorphins from working. He then used a tourniquet to cause all the study participants some pain. The people to whom he gave placebos experienced less pain than those who got the placebos plus naloxone. Benedetti’s study showed that the naloxone had blocked the effects of positive expectations. This proved that positive expectations cause the body to produce endorphins.
In high school, I was one of those kids who got good grades but got picked last when we were playing team sports. When I went on to study engineering at Dartmouth College in New Hampshire, I admired my popular athletic peers but saw myself as a geek. Thankfully, one day I bumped into the university’s new rowing coach (I literally bumped into him), who was out on campus approaching any students at least six feet tall to ask if they would try out for the team. I was just over six feet tall so he told me and his other potential new recruits that we were going to be really fast rowers. As fast as we wanted, in fact.
Being tall is a big advantage in rowing and, at just over six feet tall, I am on the short side for the sport. So I was not actually that great a candidate to be very good at it. But I didn’t know any better, and I also admired and trusted the coach, so I trained hard—very hard. I often did two training sessions per day, each one lasting between one and two hours. Had I not believed I was going to be good, I would not have endured the gut-wrenching agony of those sessions. Thanks to the hard training, I became a pretty good rower. Academics, who love impressive-sounding jargon, call believing you can achieve your goals self-efficacy. I prefer the term self-belief. Having a positive self-belief motivates you to do things (in my case, train hard).
Belief in our ability to achieve our goals can have other effects, beyond improving rowing ability. As I said earlier, research shows that back pain can be lessened through exercise. Sometimes this can feel like tough love: painful, but good for you. And motivating ourselves to do exercise even if we do not have back pain is challenging. If you have negative expectations and say to yourself, “It is never going to get better,” you might not bother doing any exercise at all, which usually makes the pain worse. The same goes for quitting smoking: If you think it is impossible, why try to begin with? You need to believe you can succeed to even try!
Since diet and exercise can reduce the risk of diseases like diabetes, heart disease, cancer, and obesity, the belief that you can improve what you eat and do more exercise can actually lower the risk of developing these conditions. Take diabetes: Research has shown that it can be prevented and its symptoms can be lessened through exercise. Of course, it is a lot easier to plan to go jogging more often, eat more lettuce, and cut out chocolate bars than it is to do these beneficial things. But if you expect to get better as a result, you are more likely to be able to motivate yourself to achieve your goals. And when you stick to the healthy program, your diabetic symptoms will be reduced.
Self-belief is important, because it is not always easy to change our behavior, even when we know it is good for us. I find it difficult to motivate myself to eat fewer sweet things. Ever since I was a small child, I’ve liked sugary snacks and even now my coworkers know they should keep the cookies out of my sight or they disappear. Once I almost convinced myself that I had a gene that made me have a special desire to eat chocolate cake and other sweet things. However, genes don’t determine those things—more about that in Chapter 13. The fact is that my belief that I could stop eating sugar actually helped me stop. Besides positive beliefs, it often helps to have friends who tell you that you can achieve things you find difficult.
The Pygmalion experiment discussed in Chapter 2 (see here) demonstrated how teachers’ beliefs about students can affect these students’ performance. At least seventeen trials have replicated the original Pygmalion experiment in teaching, leadership, and management.
Pygmalion effects also occur in medicine. In one of a number of trials that have shown this, sixty-three elderly residents at six nursing homes were given a comprehensive check of their mental health and their physical abilities. They were then randomly assigned to a “high-expectancy” or “average-expectancy” condition. Nurses were told that the elderly people in the high-expectancy group would recover more quickly and respond better to treatment.
After three months, a researcher, who didn’t know to which group the participants had been assigned, gave them another checkup. Those in the high-expectancy group did better in three out of the four outcomes measured: they were less depressed, were less likely to be admitted to the hospital, and had higher mental status than the average-expectancy group. They only did worse in one of the outcomes: Patients treated by the high-expectancy group needed more assistance from nurses than the other patients.
The way Pygmalion effects work in medicine is probably similar to how they work in a school. If a doctor believes they are prescribing a treatment that is really going to help a patient, they might treat that patient differently from someone else being given a placebo. Because of their belief that the patient is going to get better, they might be more encouraging, which could give the patient positive expectations. On the other hand, if the doctor believes that a patient is being given a “mere” placebo, they might communicate less confidence about the outcome and generate negative expectations.
In the same way that positive expectations can lead to positive outcomes, negative expectations can lead to negative outcomes. This has been demonstrated in research on the pain children experience when doctors give them injections. In one such study, doctors were asked to tell their young patients either “This is going to hurt” or “This might hurt a bit.” A third group of doctors gave children their injections while their mothers distracted them.
The children were much more likely to cry when the doctor said that the needle would hurt because they expected the pain. When the mother was distracting the child, the child had no expectations and didn’t feel as much pain. This shows how doctors who use negative words can exacerbate symptoms.
A group of researchers in China compared the effect of no words with negative words in more than five hundred patients who had hysterectomies (a surgical procedure to remove the uterus or part of it). The patients all had a syringe containing pain-killing drugs attached to their intravenous lines. They had access to a patient-controlled analgesia (PCA) pump that allowed them to increase the amount of painkiller they received when they felt more pain. Half the patients received negative messages from the doctors such as, “Please, it was useless, do not trust the PCA pump.” The other half were not subjected to negative messages.
The researchers then measured how much pain participants experienced, as well as how much morphine they asked for. The group who received the negative messages—and hence had negative expectations—reported feeling more pain and required more morphine than the control patients. Those patients who were not told that using the PCA pump was bad took an average of 45 milligrams of morphine over two days, whereas the negative-messages group consumed an average of 72 milligrams—an increase of more than 60 percent. An average difference of 27 milligrams of morphine is a lot, given that the suggested dose of morphine for adults is 5 to 20 milligrams.
Several systematic reviews have confirmed that negative expectations can harm. Colleagues of mine at Oxford University have started to investigate in more detail how negative expectations work. It seems that the expectation of a negative outcome activates brain regions associated with feeling pain.
Negative words also increase patients’ anxiety and stress levels, which can have negative health impacts. While placebos have beneficial effects because of positive beliefs, the nocebo effect—the scientific term for the harms caused by negative words—is the result of negative expectations and beliefs.
A problem is that most of us have too many negative thoughts. Even people who think of themselves as optimistic are more pessimistic than they think. One interesting study asked people the following questions:
Most people said their colleagues would have a heart attack before them, that they would retire with more money than their peers, and that their neighbors’ kids would be more likely to take drugs than their own. But this can’t be right, because on average half of us will end up better and half worse than our peers. Although we appear to be more positive than we have the right to be, we also have a lot of negative thoughts. It is hard to capture thoughts in a way that can be measured, but when scientists try to do this, they find many to be negative.
Dr. Raj Raghunathan reports studies showing that two out of three of our spontaneous thoughts are negative. Dr. Raghunathan did a bit more digging into the kinds of negative thoughts people had and found that they fall into three main categories:
I am sure we all catch ourselves having these types of negative thoughts from time to time. Some negative people say they are only being realistic and honest about life, and to them life is bad, so they’re just telling the truth. It is true that the world can be terrible sometimes. However, just like the teachers’ expectations in the Pygmalion experiment made students perform better, having negative thoughts and expectations can make things worse than they would otherwise have been.
The good thing about these negative thoughts is that they can be changed. Thoughts of inferiority come from society telling us it is important to excel (which literally means to rise above others). This leads to constant comparison and inferiority feelings, because there is always someone bigger and better (and also worse!). Negative thoughts about love arise from the fact we are told that life without a “soul mate” is not worthwhile, which is not so. And thoughts about control come from the false idea that you can control the world. Since many of these negative thoughts come from learned beliefs, it means that we can, with some effort, unlearn them. Before suggesting how, I am going to show you how medical bureaucrats who see themselves as guardians of patient safety force negative beliefs down patients’ throats.
Have you ever read about the possible side effects of aspirin? According to Britain’s National Health Service, common ones include:
The rare side effects include:
Scary, right? That is why we don’t read the small print. While we probably take too much aspirin, millions of people take aspirin without their brains bleeding. Reading about the multiple and varied side effects can induce stress. While it is important for some people to know about these, most of us can take aspirin safely without worrying.
People who enroll in clinical trials are forced to read the small print. Doctors and researchers involved in trials must inform patients about all the potential benefits and side effects of a new treatment, no matter how small the risks involved, in order for participants to be able to give their full, informed consent. A review of trials of Parkinson’s disease and depression treatments showed that between 5 and 10 percent of the patients who took the placebo treatment dropped out of the trials, usually due to side effects. But the placebo treatment could not possibly have caused the side effects. These effects can only have been caused by negative beliefs.
Unfortunately, current ethical regulations require patients to be fully informed, even if the patients don’t want to know all the gory details. This is both unnecessary and harmful and needs to change. My colleague Sir Iain Chalmers proposed a model for informed consent that is interactive and responds to the needs and wishes of patients. It would go something like this:
Good morning, Mrs. Jones, my name is Dr. Smith. Please sit down and make yourself comfortable. Your general practitioner has probably explained to you that he has asked me to see you because your breathlessness doesn’t seem to be getting any better, and he wondered whether I might be able to suggest ways of helping. I hope I will be able to do so, but this may well mean seeing you on several occasions over the next few months and working together to find the best treatment for your condition.
I’m more likely to be able to help if I can get to know more about you and your priorities and preferences. As this is the first time we’ve met, I thought it might be helpful to mention briefly how I will try to do this. Patients vary in the amount of information that they want to give and receive from their doctors. Most patients seem to get less information from their doctors than they want, but others would rather not be told some of the things that some doctors assume they must want to know. Because you and I don’t know each other yet, I’m going to need your help in learning how much information you want about your problem, and about the possible treatment options.
I’m going to depend on you to prompt me to give you more information if you think I’m not being sufficiently forthcoming, or to tell me that you’ve heard enough if you think I’m overdoing it. You also need to know that I will never lie in response to a direct question from you, and if I don’t know the answer I will do my best to find it for you. Does that seem to you to be an acceptable way of proceeding?
One randomized trial has found that this model of informed consent reduces the side effects patients suffer compared with when they are forced to worry about unusual and rare side effects. These studies show that negative words have negative health effects.
Here is a sample of words that were used by clinicians in randomized trials that were shown to have a positive effect on patients, especially for pain.
You can be your own doctor and give yourself the same messages as above—some studies show that reading positive messages (even if a doctor does not give them to you) can reduce pain. Modify the words for doctors listed above so that you are talking about yourself. For example:
Dealing with negative thoughts (such as the opposite of the statements above) is also important . . . but challenging. If I knew an easy way to banish all negative thoughts from myself—or you—I would tell you. So I will just share three tools that have helped me. The second one is based on positive psychology exercises, which, as I mentioned earlier, are supported by systematic review evidence.
After you complete the letter, put it aside for a few minutes. Then read it. Let the words sink in. Feel the encouragement, support, compassion, and acceptance. Read the letter whenever you feel down about that aspect of your life. Remember: accepting yourself is the first step toward making a positive change.