London

A scare over the safety of a widely prescribed medicine is gripping the nation. As a prominent scientist, you're being interviewed live on television. Then comes the dreaded question: “Is it safe?”. How do you respond? Insist that it is and risk sounding dismissive? Reach for a reassuring statistic and be accused of hiding behind numbers? Or admit that “we just don't know”, and satisfy nobody?

Help may soon be at hand. Late last month, a group of scientists, journalists, politicians and risk-management experts met at London's Science Media Centre to discuss how to handle this question. The centre, launched in April by the Royal Institution to improve representation of science in the media, intends to use the results in a leaflet on how best to communicate risk. “When scientists are asked the question 'is it safe' and realistically it is, then we want to give them some options to draw on,” says Fiona Fox, the centre's head.

Three volunteers agreed to undergo mock interviews at the hands of BBC science correspondent Pallab Ghosh, who confronted them with a barrage of questions on issues that have captured the media's attention in the past six months: the risk associated with the combined measles, mumps and rubella (MMR) vaccine, hormone-replacement therapy (HRT) and railway safety.

Opinions varied as to how the risks involved should be communicated. Some said that a humble “in my opinion” or an “I am confident” help to suggest relative safety. Others recommended relating the risk to other activities — “I've already done five riskier things today,” said Colin Berry, a pathologist at the Royal London Hospital. And Evan Harris, the Liberal Democrat science spokesman and Member of Parliament for Oxford West, said phrases such as “if you think sex is risky you should try lifelong celibacy” can help to make clear the benefits of taking a particular risk.

Other phrases that went down well with the audience included “breast cancer is a risk of being alive”, and “I can understand that people are worried but...”. For MMR, the participants liked the idea of stressing the fact that several studies had looked for a link with autism, but none had found one. But scientific jargon, such as “uncertain in a compound way”, is best avoided, the audience said. One participant was advised against describing the risk associated with rail travel as “unquantifiable” — even if the statement is technically correct.

As well as preparing suitable answers, Ghosh said that scientists should bear in mind that they are talking to the audience, not the interviewer. But Harris added that addressing all of the audience is impossible. “There are some people that you can never and will never persuade, whatever the evidence, so I try and give answers to the large number of people who are swayable,” he said.

Statistics provoked the most heated debate. Some argued that figures bring much-needed perspective; others said that they cloud the issue. For example, media coverage of the suspension of a recent US HRT trial focused on the apparently alarming 26% increase in the incidence of breast cancer, although this is equivalent to just 8 extra cases per 10,000 women.

Some scientists have already proved to be capable of putting such figures into context. In a 10 July interview on BBC Radio's Today programme, David Purdie, an obstetrician and gynaecologist at the Hull Royal Infirmary, discussed with interviewer John Humphrys the apparently huge increases in risk posed by HRT trials. The figures, he said, are an example of “the Judas factor” — a small number of cases can cause an alarming-sounding percentage if the total number of samples is small. “Christ was betrayed by 8.5% of his disciples,” Purdie told Humphrys. “But when you consider how many actually did the job, it was just one out of 12.”

→ http://www.sciencemediacentre.org