Trick or Treatment (22 page)

Evidence-based tea

The core principle of the trial is simple and can be traced back as far as the thirteenth century, when the Holy Roman Emperor Frederick II conducted an experiment to find out the effects of exercise on digestion. Two knights consumed identical meals, and then one went hunting while the other rested in bed. Several hours later, both knights were killed and the contents of their alimentary canals were examined. This revealed that digestion had progressed further in the sleeping knight. It was crucial to have two knights undergoing different levels of exercise, active and at rest, as it allowed the degree of digestion in one to be compared against the other. The key point of a trial is to compare the consequences of two or more situations.

The modern clinical trial, as developed by James Lind to test cures for scurvy in the eighteenth century, is less brutal than Frederick II’s trial, but the central idea is the same. If, for example, a novel treatment is to be tested, then it needs to be compared against something else, known as the control. That is why the novel treatment is given to one group of patients and the control is given to another group. The control can be an established treatment, or a placebo or anything at all. Afterwards the patients in both groups are assessed, so that the effect of the novel treatment can be compared against the control.

Sir Ron Fisher, a British pioneer of the use of trials in the twentieth century, used to recount a story that amply demonstrated the simplicity and power of the trial. While at Cambridge, he became embroiled in an argument over how to make the ideal cup of tea. A woman insisted that it tasted worse if milk was added to the tea as compared to when tea was added to the milk, but the scientists around the table argued that it made no difference at all. Fisher immediately proposed a trial – in this case the comparison was the taste of milk added to tea versus the taste of tea added to milk.

Several cups were made with milk added to the tea, and several with tea added to the milk, and the woman was challenged to identify which was which. Although the cups of tea were prepared in secret and were identical in all other ways, the woman could indeed correctly recognize in each case whether the tea had been added to the milk or vice versa. The trial had shown that there was a difference, that the woman was right and that the scientists were wrong. In fact, there is a good scientific reason why the two forms of tea should taste different. Milk added to tea leads to a less satisfying cup, because the milk becomes superheated and this causes proteins in the milk to deteriorate – these proteins then taste slightly sour.

Fisher used this simple example as the basis for an entire book on scientific testing,
The Design of Experiments
, which went into great detail about the subtleties of trials.

Despite its sheer simplicity and powerful ability to get to the truth, some alternative therapists argue that the clinical trial is a harsh test, which is somehow biased against their treatments. But that sort of attitude betrays a skewed understanding of the clinical trial, which merely seeks to establish the truth, regardless of the type of treatment being examined. In fact, the clinical trial provides a wholly unbiased and truly fair test of any medical treatment, either conventional or alternative. The unbiased nature of the clinical trial is demonstrated by the fact that the history of mainstream medicine is littered with apparently good ideas from conventional doctors that clinical trials proved to be useless or harmful.

For example, Bill Silverman, an American paediatrician who died in 2004, was a committed advocate of the clinical trial, even though he realized that it was a double-edged sword, capable of either validating or crushing any treatment. In 1949 he began working at the newly opened premature-infant station at the Babies Hospital in New York, and within a few weeks he was dealing with a premature baby suffering from a problem known as retinopathy of prematurity (ROP), which can result in permanent blindness. The baby was the child of the hospital’s biochemistry professor, whose wife had previously had six miscarriages. As this was the first time that the professor’s wife had successfully given birth, Silverman was particularly distressed at the prospect of the child becoming blind. Grasping at straws, he decided to administer a newly discovered hormone known as ACTH (adreno-corticotropic hormone), which had not previously been used to treat newborn infants. Although it was a fairly hit-and-miss approach, with Silverman varying the dosage according to the baby’s response, the end result was that she gained weight, her eyesight recovered and eventually she went home happy and healthy.

Inspired by this recovery, Silverman continued his ACTH treatment with subsequent cases of ROP. Furthermore, he compared his results with the recovery rates of babies with ROP at Lincoln Hospital, which was not offering ACTH treatment. The comparison was striking. Silverman gave ACTH to thirty-one babies suffering with ROP – twenty-five left with normal vision, two with near-normal vision, two with vision in just one eye and only two lost their sight completely. On the other hand, Lincoln Hospital had seven babies with ROP – they all lost their sight, except one.

For many doctors, the existing data – thirty-one babies treated with ACTH with a success rate of 80 per cent versus seven untreated babies with a recovery rate of only 14 per cent – would seem convincing enough. It would have been easy for Silverman to have continued with this therapy and recommended it to colleagues as a method for preventing blindness, but instead he had the humility and courage to question his own discovery. In particular, Silverman could see that his pilot study fell short of the rigour demanded by a high-quality clinical trial. For example, the babies were not randomly assigned to the treatment or non-treatment groups, so maybe the babies at Lincoln Hospital were suffering from particularly serious problems, hence their lower recovery rate. Or maybe Lincoln Hospital’s lack of success was a result of poorly trained staff or lack of equipment. Or maybe Lincoln Hospital was just unlucky – after all, the numbers involved were relatively small. To be confident about the efficacy of ACTH, Silverman decided to conduct a properly randomized controlled clinical trial.

Premature babies with ROP were randomly assigned to an ACTH treatment group or a no-treatment control group within the same hospital. Both groups were treated identically, apart from the use or not of ACTH. Within a few months the results emerged. An impressive 70 per cent of the infants treated with ACTH completely recovered their sight. Remarkably, the results in the control group were even more impressive, with an 80 per cent recovery rate. Babies in the untreated group had fared slightly better in terms of avoiding blindness, and moreover they suffered fewer fatalities compared to babies in the group treated with ACTH. It seemed that ACTH offered no benefit to babies and also had side-effects. A follow-up study confirmed the results of Silverman’s rigorous clinical trial.

The initial results from the Lincoln Hospital were abnormally poor, which had fooled Silverman into believing that he had discovered a powerful new treatment, but he had been wise enough not to be complacent and rest on his laurels. Instead, he re-tested his own hypothesis and disproved it. Had he had not been so critical of his own work, subsequent generations of paediatricians might have followed his example and administered ACTH, a useless, expensive and potentially harmful treatment.

Silverman was a passionate believer in the randomized clinical trial as the tool for questioning and improving the care of babies, which made him an unusual figure among doctors in the 1950s. Although researchers were convinced of the importance of evidence in determining best practice, the doctors on the ground still tended to be overconfident about their gut instincts. They had faith in their own sense of what the ideal conditions should be for helping premature babies, but according to Silverman this was a primitive way of deciding serious health issues:

Doctors in the 1950s preferred to rely on what they had seen with their own eyes, and would typically respond to patients with the mantra ‘in my experience’. It did not seem to matter to doctors that their personal experience might be limited or misremembered, as opposed to the evidence from research trials, which would be extensive and meticulously documented. That is why Silverman was determined to instil a more systematic approach among his colleagues, and he was supported in his mission by his former tutor Richard Day:

Half a century later, today’s doctors are much more accustomed to the concept of evidence-based medicine, and most accept that a well-designed randomized clinical trial is crucial for deciding what works and what does not. The purpose of this book has simply been to apply these same principles to alternative medicine. So what does evidence-based medicine say about chiropractic therapy?

Manipulating patients

When patients visit a chiropractor, they are usually suffering from back or neck pain. After taking a medical history, the chiropractor will embark on a thorough examination of the back, particularly the bones of the spine, called
vertebrae
. This will include looking at the patient’s posture and overall mobility, as well as feeling along the spine to assess the symmetry and mobility of each spinal joint. Often X-ray images or MRI scans are also used to give a detailed view of the vertebrae. Any misalignment in the spine is then corrected in order to restore the patient’s health. Chiropractors see the spinal column as a complex entity, such that each vertebra affects all the others. Hence, a chiropractor might work on a patient’s upper spine or neck in order to treat pain in the lower back.

The hallmark treatment of the chiropractor is a range of techniques known as
spinal manipulation
, which is intended to realign the spine in order to restore the mobility of joints. Chiropractors also call this an
adjustment
. It can be a fairly aggressive technique, which pushes the joint slightly beyond what it is ordinarily capable of achieving. One way to think about spinal manipulation is as the third of three levels of flexibility that can be achieved by a joint. The first level of flexibility is that which is possible with only voluntary movement. A second and higher level of flexibility can be achieved by exerting an external force, which pushes the joint until there is resistance. The third level of flexibility, which corresponds to spinal manipulation, involves a thrusting force that pushes the joint even further. The chiropractor will submit the vertebrae of the spine to this third level of motion by using a technique called
high-velocity, low-amplitude thrust
. This means that the chiropractor exerts a relatively strong force in order to move the joint at speed, but the extent of the motion needs to be limited, because otherwise there would be damage to the joint and its surrounding structures. Although spinal manipulation is often associated with a cracking sound, this is not a result of the bones crunching against each other or a sign that bones are being put back in their right place. Instead, the noise is caused by the release and popping of gas bubbles, which are generated when the fluid in the joint space is put under severe stress.

If you have never visited a chiropractor, then the easiest way to imagine spinal manipulation is by analogy to an experiment you can do with your hand. Position your right forearm vertically upwards and hold your right hand flat, with the palm facing up – as if you are carrying a tray of drinks. Your wrist should be able to bend backwards so far that your flat hand begins to dip slightly below the horizontal – this is what we have called level-one flexibility. If you use your left hand to press steadily and firmly downwards on your right palm, then the wrist can be bent a little further down by a few degrees, which is level-two flexibility. Imagine –
and please do not do this
– that your left hand applied an additional short rapid thrust on your right hand, thus bending it down even further by a small amount. This would be level-three flexibility, akin to the sort of action involved in spinal manipulation via a high-velocity, low-amplitude thrust.

Because spinal manipulation is the technique that generally distinguishes chiropractors from other health professionals, it has been at the centre of efforts to establish the medical value of chiropractic therapy. Researchers have conducted dozens of clinical trials in order to evaluate spinal manipulation, but they have tended to generate conflicting results and have often been poorly designed. Fortunately, as with acupuncture and homeopathy, there have been several systematic reviews of these trials, in which experts have attempted to set aside the poor trials, focus on the best-quality trials and establish an overall conclusion that is reliable.

In fact, there have been so many systematic reviews that in 2006 Edzard Ernst and Peter Canter at Exeter University decided to take all of the current ones into account in order to arrive at the most up-to-date and accurate evaluation of chiropractic therapy. Published in the
Journal of the Royal Society of Medicine
, their paper was entitled ‘A systematic review of systematic reviews of spinal manipulation’. Ernst and Canter’s review of recent reviews covered spinal manipulation in the context of a large range of conditions, but for the time being we will concentrate on the most common problems dealt with by chiropractors, namely back and neck pain. In this context they took into account three reviews looking at back pain alone, two reviews looking at neck pain alone and one review that covered both neck and back pain.