Unintentional poisoning in preschool children and other stories . . .

BMJ 2014; 349 doi: https://doi.org/10.1136/bmj.g5720 (Published 24 September 2014) Cite this as: BMJ 2014;349:g5720

Certain expressions, like “telemedicine,” “personalisation,” and “big data” have Minerva turning to the owl on her shoulder with a grimace. They seem to be used mostly by academics and politicians to mean whatever they please. Big data can be useless data. Fruit can hang low but be largely inedible. These sour reflections are inspired by a recent survey of 32 832 patients in New South Wales (BMJ Open 2014;4:e005768, doi:10.1136/bmjopen-2014-005768), which showed a poor correlation between self reported health problems and diagnoses recorded when patients were admitted to 313 hospitals in that Australian state. There was reasonable agreement about diabetes, poor agreement about cardiac disease, and almost no agreement about obesity.

But more reliable kinds of big data derived from hospital codes can be useful and fascinating, as shown by a survey of admissions for unintentional poisoning in preschool children in England from 2000 to 2011 (Archives of Disease in Childhood 2014, doi:10.1136/archdischild-2013-305298). These have decreased by 23% in …

View Full Text

Log in

Log in through your institution


* For online subscription