Evidence-based medicine: finding out what really helps

First the gums started to swell and bleed. Then the teeth fell out, one after the other. The muscles lost their power, wounds did not heal anymore. The sailors affected began to hallucinate, lost their eyesight – and finally died, from a disease called scurvy today. Some ships lost more than half of their crew on their long journeys across the oceans. Scurvy was the most horrific spectre for seamen from the late middle ages to the early modern era.

As late as 1747, the Scottish doctor James Lind laid the foundations to end this terrible nightmare. He assumed that the disease once also called ‘mouth rot’ could be cured with acids. But which acids were the right ones? Lind subdivided twelve scurvy-ridden seamen into six groups of two. One group was given vinegar three times a day, another group 25 drops of sulfuric acid. Other men were given similar delicacies. However, only those two sailors who had been given lemons and oranges were recovering from the effects of scurvy.

First study heals sailors from scurvy

It took further decades before these insights found a way into the regular diet on the ships, and even longer to recognise that not the fruit acids healed and prevented scurvy, but the Vitamin C contained in the fruit. But Lind had discovered the cure for scurvy in a manner that is -- to this day -- seen as the almost optimal manner to gain new medical insights: Lind had conducted the very first known comparative study.

Even 270 years after Lind’s discovery, medical researchers still ask themselves the same questions: Is a new medicine against a certain type of cancer superior to the ones previous prescribed? How does one’s diet impact upon rheumatism? Does a catheter insertion into the human heart outweigh the risks of such an operation? What are the advantages and disadvantages of live-long medication to limit the effects of a chronic illness?

The more complex the results of medical research, and the more possibilities derived from it, the more often doctors reach their limits in terms of application and experience. And all the more if the treatment or medication does not result in success as clearly and obviously as it did with Lind’s research. That’s when researchers come into play: they determine the actual effect in scientific studies, ideally with thousands of participants.

The most reliable results often come from a more refined version of the kind of studies James Lind conducted: for example, when the impact and the risks of a new medication are to be assessed. Patients are randomly assigned to two groups. One group takes the new medication, the other group either previous standard medication or one without any active substance, so-called placebos. Neither the patients nor the researchers know who is assigned to which of the two groups. That’s the only way for a comparative study to yield objective results. Because the participants are randomly assigned to one of the groups, the experts call this a randomised study.

Studies like that are one of the most efficient tools in medical treatment based on scientific research. Even better if there are a number of these studies that can be compared, contrasted and evaluated in order to achieve a meaningful understanding of a new method of treatment.

One impassioned advocate of this so-called evidence-based medicine is Professor Gerd Antes. He has countless anecdotes on how this research method saw the light of day and eventually reached Germany.

For example that evidence-based medicine originated in Canada. And that an important milestone was the publication of a book by Archie Cochrane in 1972: the publication laid down the principles of that new way of thinking medical research. The International Cochrane Collaboration is named after him. So is its German representation, the Cochrane Institute in Freiburg. Gerd Antes was in charge of the Institute from its foundation in 1998 to 2018.

In Germany, the new method arrived as late as the mid-1990s. In 1995, the expression Evidenzbasierte Medizin first appeared in German, although its English equivalent had been in use for a long time by then. This was partially a language problem, as evident means ‘obvious’ or ‘crystal clear’ in German. But since the term had been in use internationally for such a long time, it was finally accepted in German medical circles (and abbreviated to EbM).

‘It’s about the best knowledge available’, Antes explains. Randomised comparative studies can ideally provide that. But EbM is so much more: for example, follow-up studies after the licencing of new medication can uncover rare side-effects. Data from the registers can provide information on the risks of certain artificial replacement joints. Or indicate how well patients with a certain ailment are looked after by the medical profession. Antes would accept the results of inferior methods when there’s no randomised studies available – but only with a pinch of salt.

Personal impression is often wrong

But is it not a bit one-sided to rely on medical research only? For most patients, something else is far more important: they want to judge for themselves if a certain treatment is advantageous for them. But personal impression often gets it wrong. An allegedly healing impact is often based on the so-called placebo effect. For example, if randomised studies compare real medication with active-ingredient-free pills, the latter ones often also show a certain effect. It’s the power of belief. Or simply the feeling that someone is looking after the patient. The real medication, however, would have much stronger healing powers than the placebo the patient took.

Some fallacies in medical history demonstrate how important and proper it is to stick to thorough scientific research. These fallacies were the result of poorly conducted studies, or no studies at all – until good studies improved our knowledge.

One rest, now move your body

Today, women in the menopause are advised to take hormones only sparingly, for a short time and in low dosage. Good research has identified the risks of such medication. For a long time, people with a bad back were advised to rest or even stay in bed. Nowadays, it’s recommended to stay active and to exercise. The same goes for heart disease: if done in accordance with medical advice, exercise is actually quite good for the patients. People with arthritis of the knee used to undergo an endoscopy to rinse out the knee. Studies have since shown that a pretended operation had the same effect as the endoscopy.

Many alternative therapies only work through the placebo effect. Homeopathy might be the best example. Random raw materials – from the neurotoxic arsenic to herbal extracts to dog poo – are diluted, often to an extent that the original substance is not traceable anymore. The claim that they have healing powers nevertheless contradicts everything we know about the human body. And still, homeopathy has been given ‘a second chance’, says Dr Christian Lübbers, an otorhinolaryngologist. Studies had shown a certified effectiveness, he adds. However, all eleven survey studies so far – four of them conducted by homeopathists at a British institute – have failed to prove its effectiveness.

Lübbers has been treating patients who weren’t even benefitting from the placebo effect of homeopathica in his surgery for a long time. He once removed globuli from the auditory canal of a four-year-old suffering from an inflammation of the middle ear. For Lübbers, that was the final straw: he started to publicly advocate an honest information policy on homeopathy.

Double standards for medication

Homeopathica – like anthroposophical and herbal drugs – benefit from a special clause in the German drug approval law: a mere registration with the appropriate board is often enough. For drugs that need a formal approval, the threshold is rather low. That means that a strict proof of effectiveness is not required. ‘That’s double standards’, argues Lübbers. ‘It should not be tolerated any longer for reasons of patient safety and consumer protection.’

Apart from diluting mistakes with poisonous raw materials, there is no direct danger, he adds. However, the indirect danger is that necessary and effective treatments are delayed or even avoided. ‘The provision should be based on scientifically proven effectiveness’, says Dr Dagmar Lühmann. ‘That is true for therapies, diagnostic tools, risk factors and disease progression.’ Lühmann is Deputy Chairwoman of the German Network Evidence-Based Medicine, a group campaigning for science-based concepts since 1998.

Since then, a lot has changed in terms of guidelines for doctors, course contents for medical degrees, or the question of what health insurers have to pay for. However, says Lühmann, there is a lot to be desired when it comes to the practical implementation in surgeries or hospitals.

Evidence-based medicine is no degradation of medical practitioners, as it’s based on three pillars (see graphic above): the highest possible level of knowledge, the experience of the medical practitioner, and the requirements of the patient. Doctors can deviate from the guidelines in individual cases. They can prioritise in case a patient suffers from a multitude of diseases.

Ultimately, the patient takes the decision on the appropriate therapy, and not the scientists. For example, if arthritis of the knee is treated with pain killers or if a joint replacement is required instead. If one of the pillars crumpled, the whole EbM building would collapse.

Science is always in flux. Any day, new findings could overthrow old dogmas. Therefore, nothing protects patients more efficiently from ineffective or even harmful treatment than up-to-date knowledge. Nothing guarantees the best possible treatment more than research. James Lind would certainly agree. Thanks to his comparative study, scurvy was prevented and cured, and the spectre for medieval sailors driven away.