Why don’t people make good choices around health? Why do we persist in unhealthy behaviour, choose ‘alternative’ therapies that don’t work, demand antibiotics for viral illness?
The usual answer is that the general public – lazy, feckless, a bit dim – is to blame. Or maybe the media, feeding them vaccine scares or over-blown stories about miracle cures.
A new book published this week suggests we need to look a bit closer to home for the root of the problem. How can patients make good decisions if doctors can’t inform them properly, and the research asks all the wrong questions (then presents it in the wrong way)?
Better Doctors, Better Patients, Better Decisions, edited by Gerd Gigerenzer and Muir Gray, (pub. Strungmann Forum Reports) could simply be titled Better Information. The editors, working with a range of authors, look at the barriers that must be overcome for patients to get good enough information to make a proper decision.
One barrier is a general illiteracy around statistics, and there’s no doubt that education in this area could be better. Speaking at the launch of the book in BMA House, Dr Gigerenza said this was not just a problem for the general public, but that his research showed most doctors struggled to interpret stats too.
Yet the way statistics are presented can make all the difference. Giving study results as natural frequencies – 1 in 100 people died after having this treatment – is a clear and powerful way of getting a message across. Use a percentage – the treatment had a 1% mortality rate – and you immediately introduce uncertainty. One percent of what?
It gets worse. As a medical writer, I love it when medical studies include absolute risk percentages like the example above, because I can get my head round them. They rarely do. Far more common is a relative risk, or odds ratio. (Mortality odds ratio for the treatment was 1.2, for example. It means nothing unless you know the absolute risk it’s compared to.)
Worse still is when you get a mix of absolute and relative risks. I find it quite shocking that this still happens, and Dr Gigerenza said it was the ‘moral responsibility’ of journal editors to ban this practice.
The paper will report the benefits of a treatment as a relative risk (a 50% relative risk reduction in death from breast cancer) but the risks as an absolute risk (a 1% absolute risk increase in death from heart attack). Assuming the heart attack risk rose from 1% to 2%, these risks of course cancel each other out. But the bigger figure grabs the attention, and that’s what you’ll see on the front of the Daily Mail.
This summary just scratches the surface of the book, an impressively solid tome awaiting my full attention. Reckoning With Risk, Gigerenza’s earlier work, is a more accessible introduction. Oh, and Mervyn King, governor of the Bank of England, was at the launch of the book. Should I be pleased or worried that he’s a fan?