The placebo effect - how does it work?

Back in the 50s, it was discovered that someone believing they were getting treatment would report that they felt better even if in fact they weren't getting any treatment at all. It was as if telling the patient they were getting well was somehow helping them to do just that. All sorts of reports were published confirming the power of positive thinking, and opinion divided fairly sharply. To some, it was confirmation that the majority of people claiming to be ill were simply malingerers, whereas others thought that this represented an important medical discovery. If it was possible to improve people's health outcomes simply by telling them they were getting better, this was both a cheap and effective way of getting people well again.

It didn't work out quite like that because in 2001, there were systematic clinical trials which investigated this effect. Disappointingly, it showed that although the effect was real enough, patients reported they were getting better even when they weren't. That meant it wasn't a real treatment – it wasn't improving patient outcomes, even though those patients reported they were feeling better. And that's the problem.

The Cochrane Collaboration, an organisation which performs meta-analyses on published clinical research confirmed that despite patients reporting they felt better, their clinical status remained unchanged. It wasn't a treatment, it is a subjective effect.

Expectation and anticipation

In understanding the placebo effect, it is useful to know a little about how the brain works. Our brains maintain a model of the physical world which allows us to predict how things will behave. We expect the world to behave in a certain predictable way and having a correct model is important to us, not least for our safety. When the world turns out not to behave that way, we correct our mental model and behave differently. For example, if we anticipated that snakes were wet, our model would be corrected when we subsequently discovered they were dry. Next time, we'll expect something a little different. That expectation actually affects our brain chemistry.

The interesting thing is that our brain anticipates by changing its chemical state. Before we touch the snake, the brain has already responded as if the snake would be wet, anticipating the sensation of wetness. That enables it to test the accuracy against the real sensations when we discover the snake is dry. The same applies when we catch a ball. Before we actually catch it, the brain areas associated with the action are activated in anticipation.

We can detect these changes in brain scans. When an individual empathises with another, they actually feel what they think the other person is feeling. If someone in front of us is feeling sad, we make tiny adjustments to our own facial expression, tiny changes in muscle tension, which affect our brains in such a way as to create a similar feeling of sadness. Our feeling of empathy is exactly the mechanism of anticipation and prediction. The brain affects the state of our body, and the state of our body affects our brain as well.

And that is exactly what the placebo effect is. It is the change in the brain brought about by the expectation and anticipation of treatment.

A patient going to the doctor or therapist is primed to expect and anticipate some form of treatment that will make them feel better. Their brains respond by anticipating just that feeling of improvement, and it takes place whether or not the patient receives any treatment. Every GP knows this very well and uses it in the traditional “bedside manner”.

This effect is not just theoretical. It can be observed using functional magnetic resonance imaging (fMRI scans) which can show increases in brain activity of different regions in real time. The anticipation of pain reduction is associated with the release of endorphins and other opioids in just those regions of the brain associated with pain reduction. Although the effect is similar to actual pain reduction, it is a much lower response. But anticipating pain relief, or some pill that will provide treatment, produces a brain reaction which will make the patient feel better even if no treatment is being provided.

Of course, such a placebo effect is not the same as real treatment of the source of the pain. Nevertheless, the patient will report that they feel better. The underlying condition will not have changed. When we go to a doctor or therapist, we have been conditioned to expect a positive outcome, to anticipate pain relief and real treatment. The conditioning reinforces the placebo effect. And so long as the anticipation and expectation lasts, there will be some placebo effect.

So can we use placebo as a treatment?
The question arises as to whether or not we can employ the placebo effect as a real treatment. If sympathy and a good manner can generate positive results, surely we should accept it as treatment? Unfortunately, the placebo effect is remarkably fickle. Even if it works on one occasion, it may not work a second time, or in different circumstances. Even the same patient may experience a placebo effect in one treatment and not in another. If the anticipation or expectation changes for whatever reason, the placebo effect may disappear. Nevertheless, if it is continually reinforced, the placebo effect can last a long time, even years. Devoutly religious people experience the placebo effect associated with their prayers which they can maintain over very long periods of time.

In the case of alternative medicine therapies, these placebo effects will persist as long as the customers are convinced that the therapy is doing them some good. For this reason, many of the rituals associated with alternative therapies are concerned to convince the customer that something is being done to actually treat them whether or not anything real is happening. But the, if they really are feeling better, that's a result isn't it?

Unfortunately, the answer has to be an emphatic no. If it was possible to treat people using the placebo effect, the whole medical establishment would welcome it with open arms: it's cheap, non-invasive, easy, requires barely any training. The problem though is that it only affects the patient's mental model, their subjective state, and not their underlying physical state. If patients go away thinking they are better when they're not, that's seriously misleading. If they suffered from cancer, they may actually be led to avoid effective treatment altogether and die.

In 2001 and 2004, Hróbjartsson and Peter Gøtzsche studied 156 clinical trials to compare the outcomes to see if placebo was better than no treatment. That's different from comparing placebo to real treatment – those trials are already routine and extensively published. These were to compare the effectiveness of placebo itself, against non-treatment. If placebo could be shown to have a better effect, then it could be considered a treatment. But the placebo results were no better than no treatment – placebo is not better than no treatment at all.

So when alternative medicine practitioners are relying on placebo as evidence of their successful treatment, they are not offering any proof at all. It is no better than no treatment. To justify their therapies, they need to show clinically controlled, double-blind trials which demonstrate efficacy. These are notoriously absent even from the journals of alternative medicine practitioners.

Interestingly, the brain reacts in the same way when it expects a worsening of the condition as a result of treatment. In this case, the nocebo effect, the brain anticipates a worse condition and the patient reports that they are getting worse, even when they are getting better.

The bottom line is that the placebo effect is not a treatment and that alternative therapies based on it are sham treatments, no better than nothing at all.

Leaving the Land of Woo