Twist of faith
When the University of Hong Kong released the results of a study into the effectiveness of acupuncture in treating depression recently, it wasn't the fact that 19.4 per cent of patients reported an improvement in their symptoms that was surprising.
The surprise was that 8.8 per cent of the patients in the study showed an improvement from so-called placebo acupuncture, which used needles that look like the real thing but don't do what real ones do.
Patients' health, it turns out, can improve simply because of the expectation of improvement from receiving medical attention. In other words: the placebo effect is about a lot more than sugar pills.
'The placebo effect is complicated; we don't know how much it's related to expectation and how much to biochemical effects,' says one of the men behind the HKU study, Dr Roger Ng Man-kin, chief of service in the department of psychiatry at Kowloon Hospital.
What's more, Ng says, expectation and medicine aren't mutually exclusive: 'If you have hope about something and then expose the brain to a scan, you'll see changes. Adjusting your mind activity can result in changes to the brain. It's hard to separate physical and mental effects, because they both cause activity in the brain.
'The placebo effect is even more prominent in providing psychotherapy. There's quite a bit of expectation and quite a bit of support that are part and parcel of the treatment.'
Medical wisdom acknowledges that the placebo effect works. The view came to widespread attention in 1955, when doctor and medical ethics expert Henry Beecher published his paper The Powerful Placebo. Various studies popped up over the years that appeared to support his view. But it wasn't until 2001 that a systematic review of all the available material was conducted - and it contradicted medical orthodoxy by finding that there was no robust evidence for the effect, except perhaps for treating pain.
The 2001 review has been controversial ever since, mostly because it aggregated results from many different diseases. The effect of placebos varies wildly depending on which type of condition you try to treat with them - from zero for blood poisoning, to 80 per cent with duodenal wounds.
There are a few things known about placebos. They seem to work better on children than adults, for example, and they don't work at all on people with Alzheimer's, because their brains have lost the ability to form expectations.
Neuroimaging shows placebos seem to affect the dorsolateral prefrontal cortex, the part of the brain responsible for high-level functions such as working memory and attention. Naloxone, a drug that is usually given to counter overdoses of opiates such as heroin, was found in 1978 to block the effectiveness of placebos, suggesting that endogenous opioids - the pleasure chemicals, such as endorphins, that the body generates to counteract pain - are involved in the effect.
The placebo effect also has a converse: the nocebo effect (in Latin, placebo means 'I shall please', and nocebo, 'I shall harm'), a term coined in 1961 to describe the harm an inert substance can do if it's perceived by the patient to be dangerous. A study published in the journal Pain in 2009 found that in every placebo-controlled trial ever done on a migraine drug, people in the control group who received a dummy sugar pill felt similar side effects to whatever drug they thought they were getting. Those who got the placebo instead of the painkiller, for example, reported digestive problems.
Here's another twist: in a study published in PLos One in 2010, researchers found that placebo pills benefited patients even when doctors explained what they were taking. Researchers divided 80 study participants who suffered from irritable bowel syndrome (IBS) into two groups. One group received no treatment for their condition while the other received sugar pills that they took twice a day. Three weeks later, 59 per cent of study participants who knowingly took the placebos reported reduced symptoms and adequate relief from irritable bowel syndrome, whereas only 35 per cent of the control group reported similar results.
Placebos seem to have more of an effect on conditions that involve subjective patient reporting of symptoms, and less on those where symptoms can be externally and objectively verified. Pain is particularly amenable to placebo treatment, along with chronic fatigue syndrome and depression.
'In psychiatric medicine, what you notice is that it takes about two weeks for the body to absorb medication,' says Dr Justin Grayer, a member of the Hong Kong Psychological Society's division of clinical psychology. 'Some people say they feel better a lot quicker than that; it could be their bodies, but it could also be the placebo effect. I think that if people expect to feel better, there's a chance they will feel better.'
A study from Harvard Medical School's Programme in Placebo Studies and the Therapeutic Encounter, established last year, found that the physical symptoms of asthma patients receiving placebo treatment didn't diminish, but they described themselves as feeling better.
'Thoughts and feelings influence the body,' says Dr Edward Shen, clinical psychologist at the Matilda Medical Centre, who specialises in traditional Chinese medicine and energy medicine, the form of Western alternative medicine that emphasises a putative qi-like life force.
'This is where placebos fit - into the workings of the mind. If you don't believe that the mind can affect physical matter, then you don't believe in placebos. If you believe in energy affecting matter, then the placebo effect is central.'
The main use of placebos, of course, has nothing to do with deceiving people. As in the HKU study, they're used as a control in medical trials, a point of comparison so that any effects of the treatment being tested can be isolated. Patients not knowing what they're getting is a necessary evil; giving people placebo drugs for actual medical treatment is always ethically dubious.
'It's part of professional ethics that people need to provide informed consent,' says Grayer. 'You need to know what something is in order to properly consent to it, so that rules out placebo drugs completely.'
It also raises something called the placebo paradox: it may be unethical to prescribe a placebo, but it may also be unethical to choose not to prescribe something that has a chance of being effective.
Fortunately, physical treatments, including pills, and also creams, inhalants, injections and acupuncture, aren't the only types of placebo. Shen argues the whole edifice of clinical practice is a kind of placebo: if part of the effectiveness of all treatments is rooted in patient belief, then clinical practice has moulded itself to reinforce positive associations that help people get better. People are likelier to get better if their doctor expresses confidence that they will - so in a way, the doctor can be a placebo.
'The only way out is to package it in non-physical terms,' says Shen. 'So, it becomes about the credibility of the physician. If you walk into a doctor's surgery in Hong Kong, they have all their certificates and diplomas displayed on the wall. The way they dress, the spectacles they wear ... It's all about creating an impression that will make you believe in the doctor's ability. That's the placebo effect.
'You can even say hope is a kind of placebo.'
Could the wealth of information available on the internet erode people's impression of medical omniscience, and cause the placebo effect to diminish?
'I would say yes, because now there's so much medical knowledge among patients,' says Shen. 'Sometimes a patient knows more about their condition than their doctor.'
At present, however, the opposite seems to be happening, with the effect becoming more pronounced over time - possibly a result of information overload, or of reporting bias. 'People who are trialling a treatment are much more likely to announce positive results,' says Ng.
'Then it goes onto the internet, where it's picked up by people who may or may not be experts. There's a material from uncontrolled studies that might not be scientifically sound. When you type a word in, you might think that the treatment is going to be effective - and that's why the placebo effect is increasing. If people are looking for a novel treatment, they're probably quite desperate about their condition and haven't responded to treatment. Their hope increases with a new treatment, and so the placebo effect is going to be increased.'
So the placebo effect is helping to shape the future of medicine. With so many variables, its workings are still a mystery, but it has all sorts of potential implications for the role of the mind in physical well-being.
'The mind and the body are so close together,' says Ng. 'One affects the other. It's why mental health care and physical health care are getting closer together. People who have a psychological illness are likelier to be physically ill, and people with a physical illness are likelier to be psychologically ill.'
- The percentage of patients who, in a 2010 study, knowingly took placebos but reported reduced symptoms and adequate relief from irritable bowel syndrome