Ben Goldacre’s March 1st Bad Science piece for the Guardian, Don’t laugh, sugar pills are the future, in which he comments on the latest research to show that SSRIs are not much more effective than placebo in treating depression is, as usual, a bit thin on the ground with the actual science itself, even if his title might turn out to be remarkably prescient. And, wonder of wonders, I even agree wholeheartedly with a substantial amount of his earlier piece on February 27th, based on the same study, A quick fix would stop drug firms bending the truth. But far more interesting is the piece by Clive Cookson in the FT, Is there an ethical way to fine-tune the placebo effect?
In commenting on the recently-published meta-analysis of both published and unpublished trials of SSRIs led by Irving Kirsch at the University of Hull, which follows his more limited analysis of published studies in 2002, Cookson writes
Anti-depressants do not work, chorused newspaper headlines this week. The truth is quite different. The research in question showed that new-generation drugs, such as Prozac and Seroxat, relieve depression very well – but so do dummy pills.
This is the crucial point. Not, as Goldacre opines
It was fun to hear universal jubilation over the new meta-analysis showing once again that some antidepressants aren’t much cop in mild or moderate depression.
Cookson goes on to say
The study, led by Irving Kirsch at the University of Hull, is the latest testament to the power of the placebo. Analysing the drug companies’ own clinical trial data, the researchers found that four widely prescribed drugs improved patients’ score on the standard clinical test for depression by 9.6 points, while placebo pills gave a remarkable 7.8-point improvement.
As the researchers said in the journal PLoS Medicine: “The response to placebo in these trials was exceptionally large.” This set the bar for demonstrating efficacy so high that – except for severely depressed patients – the difference between treated and placebo groups did not reach a “statistically significant” level.
But researchers are only just beginning serious investigation of placebo power. “We do not really know what the mechanism is,” says Dr Derbyshire. “In fact, there may be lots of different mechanisms. For example, placebo painkillers somehow activate the brain’s endorphins [natural painkillers] while placebo aspirin activates a natural anti-inflammatory effect.“
Exactly. Now we’re getting a bit closer to the mark.
I’ve talked elsewhere about the non-local aspects of homeopathic treatment and the quantum mind-like effects observable and said
I’m saying the qualitative principles of quantum mechanics have the potential to model some of the observations which have been made in respect of homeopathy and many other of the more subtle, holistic therapies. Let’s get one thing straight right away. These effects are not specific to the therapy. They just become more apparent in the context of the therapy because of its level of subtlety and its holistic nature. They’ll be occurring just as much with conventional medicine too, but will be far less obvious to observers who are looking at things in a much more focused and linear way.
Here it would seem that evidence is starting to become clearer. And we need some new terms. ‘Placebo’ can’t be used to describe the specific effect of the patient’s expectation that the pill they’re taking will help them, at the same time as being a dustbin term for all non-specific effects of treatment. Personally I think it’s time the word placebo was restricted to its original sense and use: a dummy pill administered by a physician when he wants the patient to believe he’s taking the real McCoy. Using it in respect of intangible but verifiable effects of treatment is confusing the picture and leading to a derogatory attitude to these effects when we should be studying them free from such prejudice.
I’m going to suggest that there are 3 principle components in this effect: i) the effect due to the patient’s conscious expectations, ii) the effect due to the physician’s expectations transmitted to the patient through conscious entanglement, and iii) the effect due to conscious entanglement with the nature of the substance being prescribed.
The effect of patient expectation is clear and logical enough. Physician expectation also plays its part. In a February 2000 article in the Guardian on the rise of complementary medicine, Healing in Harmony, Jerome Burne wrote
Medicine is both an art and a science, but science has been firmly in the driving seat for 40 years. The arrival of CM practitioners may allow some of the more intangible aspects of the healing profession to re-emerge into the light, such as the power of the doctor’s own belief. “When I was starting out as a doctor, my professor told me about a new migraine drug,” recalls Marshall Marinker, professor of general practice at the University of London. “I prescribed it to a number of my patients, and it worked brilliantly. Many were completely cured. But then I began thinking about its mechanisms and how to design a trial, and it somehow stopped working so well. I never again got such good results as when I totally believed in it. I don’t think you can measure that sort of thing in clinical trials, but it is absolutely vital to the way medicine should work.”
The final effect, conscious entanglement with the nature of the substance being prescribed, is possibly going to be a harder stretch for some. Yet it seems the most plausible mechanism to explain how, as Derbyshire says above, “placebo painkillers somehow activate the brain’s endorphins while placebo aspirin activates a natural anti-inflammatory effect”.
And as it happens, this hypothesis also posits a rationale for homeopathy, explaining why effects should be observed when patient expectations aren’t relevant, and also why the wrong remedy has no effect.
So when Goldacre writes “Sugar pills are the future, if only there was a way to give them with integrity, and a straight face” he may very well find that the last laugh is, resoundingly, on him. There’s plenty of integrity and straight faces among homeopaths …