Statistics of lives saved by heroes behind the headlines

WITH every scientific breakthrough that has a life-or-death impact there are those names and faces who receive the lion’s share of the bouquets and column inches. These people are usually very generous about the (often) multidisciplinary teamwork which actually went into the final discovery, but the collaborators are unnamed.

Many are probably happy enough to receive the acknowledgement of their peers without also craving public applause. But we should know their names. Often they work in areas the public doesn’t easily understand, but what they do can be of almost earth-shattering importance.

One scientist quietly saving lives is Professor Julietta Patnick, who will receive her honorary doctorate from Sheffield Hallam University later this year in recognition of 32 years overseeing standards for screening of breast, cervical and colorectal cancer.

As director of the NHS cancer screening programmes, she has developed and led the national approach to the early detection of cancer through screening. Another such name is Paul Meier.

Countless new drugs are developed each year, but how we know which ones are better than what went before – the knowledge that convinces governments to support the widespread use of those medicines – is down in great part to this man most of us have never heard of.

The New York-born statistician has just died at the age of 87, and for the many of us who didn’t realise his importance, it’s still not too late to doff our hats to a man hugely instrumental in dragging medicine out of the days of snake oil and into the modern world.

Anyone who’s had a successful treatment for cancer, diabetes, heart disease or HIV/AIDS or scores of other conditions during the last half-century owes their life to Meier’s work in medical statistics, including the fields of randomised drug trials and the Kaplan-Meier Estimator, which gave medical statisticians a simple way of comparing patients’ survival rates after different treatments. As a medical research tool both are now universal.

“The world is perhaps not appreciative enough of the contribution made by statisticians to medicine, but Paul Meier was a towering figure and his work was staggeringly revolutionary and radical,” says David Spiegelhalter, Winton Professor of the Public Understanding of Risk at Cambridge University. “After the Second World War the British statistician Austin Bradford Hill, who was part of the team working on the use of streptomycin (the antibiotic) in tuberculosis, introduced the idea of flipping a coin to decide who would get the new drug. This was the first ‘randomised’ trial.

“Before that the whole business was a lot more paternalistic, and doctors would choose to try new medicines on the patients they thought would benefit most. But that wasn’t a good system, and nor was simply giving a new drug to every other patient, which was too easy to fiddle.

“Before randomisation – literally flipping a coin – doctors could perhaps also look for the result they wanted to find. There was also an assumption that new drugs were better than old ones, when that wasn’t necessarily true. Randomisation was the only way of getting reliable data.”

Meier took what had been done in the UK on a smaller scale and developed the idea. Researchers randomly assigned one group of patients to receive an experimental treatment and another to receive the standard treatment. The results couldn’t be fiddled or unintentionally skewed, and if the sample of patients is large enough the two groups will be the same in every respect except the treatment they receive.

Meier’s contribution was not just in adding to the technical understanding of randomisation; even more importantly he used the sheer force of his personality to persuade the US government’s Food and Drugs Administration to demand that all new drugs be subjected to randomised trial. Where America led, the world followed.

“That strategic decision half a century ago has already saved millions of lives and those millions should be attributed to Paul,” says Sir Richard Peto, professor of medical statistics and epidemiology at Oxford University. Walter Gregory, professor of statistical methodology in medical trials at Leeds University, added: “Taking one example, randomised trials of breast cancer drugs over 30 years have dramatically improved survival rates, because randomisation has led to small incremental changes in medicines that add up to a huge difference.”

[email protected]