Editor—I have for a long time thought that one of the chief obstacles to the public's understanding of medicine is the inability of the average punter to understand the concepts of probability and risk-benefit analysis that underpin most of the treatment decisions we make, and our failure as a profession to dispel that ignorance.
It was disappointing to read Smith's Editor's choice, in which he bemoans the fact that doctors seldom say to their patients: “I must warn you that the simple fact of being admitted to hospital means that you have... a one in a 100 chance of dying.”1
We don't say it because it's not true. It may well be the case that 1% of patients admitted to hospital die, but very few patients enter hospital with a one in 100 chance of dying—for most, it's much less than that. Would Smith have us tell a young, fit patient admitted for a hernia repair that there is one chance in 100 that he or she won't come out alive? If not, which patient would he choose as the recipient of this alarming message? The patient in a road crash with multiple fractures and an aortic laceration perhaps? But in that case, of course, 1:100 would be a significant underestimate of his or her chance of dying. This is not just statistical semantics; for individual patients the 1% death rate is a complete irrelevance, and suggesting that this figure is something that they need to worry about is grossly misleading.
Such a figure may make for a headline grabbing editorial (and making a splash in the tabloids seems to have overtaken the impact factor as a measure of success for the BMJ), but it is not science.
Competing interests: BB holds a conviction that the BMJ can no longer be considered a scientific journal.
References
- 1.Smith R. Editor's choice. Think harm always. BMJ 2004;329: 0-g. (3 July.) [Google Scholar]