Published: 08 July 2010
by LEWIS WOLPERT

THE distinguished biologist Peter Medawar claimed: “If a person is poorly, receives treatment to make him better, and then gets better, then no power of reasoning known to medical science can convince him that it may not have been the treatment that restored his health.” 

As Francis Bacon pointed out several hundred years earlier, we have the habit of seeing order and making causal connections because they appeal to us. This has been true for most doc­tors until quite recently. 

It is hard to believe, but true, that it took so long for medicine and its treatments to achieve a reliable scientific basis. 

The doctors from ancient times till quite recently were convinced their medicines actually worked and made their patients better, even when they did not. Many patients were killed by the medicines given to them. How eventually doctors began to take testing of drugs seriously is the core subject of Druin Birch’s important book, Taking the Medicine.

There was no view in ancient times that a drug should be tested to see if it cured the patients. One that seemed to need no testing as its effects were so obvious came from the poppy, which we now know contains opium. Second-century Rome called it the strongest of drugs but very few ancient drugs – the physician Galen referred to hundreds – did any good, and he was also very keen on bleeding as a treatment. From the poppy came morphine and a chemical modification by the firm Bayer resulted in heroin, which they promoted as a treatment for res­piratory illness. There were no studies which could have shown that it was addictive. 

There were many new discoveries, like the finding that the bark of a tree from Peru could reduce fever and so could help with malaria as it contained quinine. Oliver Cromwell refused to take it as he regarded it as tainted with the papacy, and he died. But it saved, probably, King Charles II.

In 1662 John Baptista Van Helmont published a book in which he described what he considered ought to be done to test whether a drug really worked – it was the first description of a randomised clinical trial. But he never did one. If others had, they might have discovered what harm doctors were doing with their medicines. 

The first clinical trial was carried out at sea in 1747 by James Lind who divided sailors suffering from scurvy into pairs and gave them different treatments; those who received fruit recovered best. But, claims Burch, himself a doctor, he did not appreciate what he had done and it was only nearly 200 years later that Archibald Cochrane started proper trials, and helped to change the way doctors viewed the nature of evidence. 

In the early 19th century, Alexandre Louis questioned the value of bleeding and he proposed a clinical trial, but he never did a good one. 

In 1860 Oliver Wendell Holmes said that the all medicines could be thrown into the sea and it would be better for mankind, but worse for fishes. He also attacked homeopathy, which was invented by Samuel Hahnemann who proposed that a drug which produced the same symptoms as the disease can be used to cure it. There are still many today who believe in homeop­athy although all the trials show it does not work. 

The book is full of fascinating accounts of the researchers and how drugs were discovered. Millions of soldiers in the First World War lost their lives due to infections and there was a search for compounds to kill bacteria. The ability of the mould penicillium to kill bacteria had been observed some 50 years before Alexander Fleming started his studies, and he failed to isolate the active compound. It was a group in Oxford during the Second World War who did this.

Statistics took a long time to establish itself for examining evidence in medicine as an editorial in the Lancet in 1921 makes clear. There were also problems about the ethical issues in assigning patients randomly to different treatments. Doctors were even quite recently still using drugs without understanding what they did, and Thalidomide is an example.

Burch argues that the worldwide system for drug regulation is poor and that doctors can give an untested drug to their patients. 

He does ack­nowledge how difficult proper testing is. In the 1960s fewer that 10 per cent of the drugs doctors prescribed were suppor­ted by reliable evidence. It is thus a relief that things have much improved and the figure is now round 80 per cent.

Yet there are studies suggesting that the millions of antidepressant doses taken actually are no better than a placebo, except for severe depression. 

Randomised clinical trials are essential to deal with both medicines and the claims from alternative med­icine. An important and very interesting book.

• Taking the Medicine – A short history of medicine’s beautiful idea, and our difficulty swallowing it. By ­Druin Burch. Chatto and Windus paperback £9.99

• Lewis Wolpert is Professor of Biology as Applied to Medicine in the Department of Anatomy and Develop­mental Biology of University College, London