
Novelty in science . . . and other stories
Cancer screeningAlmost everyone believes that cancer screening is a good thing because it leads to early detection of malignancy and better outcomes for people who screen positive. An essay in the New Yorker explains why almost everyone is wrong ( It’s worth reading even by doctors who think they understand lead time bias, predictive values and Bayesian probability. Muir Gray and colleagues (BMJ 2008 doi:10.1136/bmj.39470.643218.94) were right nearly 20 years ago when they wrote: All screening programmes do harm; some do good as well, and, of these, some do more good than harm at reasonable cost.Novelty in scienceScientific papers often boast about the novelty of their findings. But novelty doesn’t necessarily imply usefulness or importance. By contrast, investigations that replicate previous work can be valuable without being novel. In an attempt to measure novelty and find out whether it matters, an open competition to devise and validate indicators of scientific…
Source link