An exclusive for the Daily Mail, courtesy of ONS

The Daily Mail yesterday splashed on a story on the number of care home residents who have died as a result of dehydration in the past five years. The story was followed up by the Daily Express and the Daily Telegraph.
There is nothing wrong with the actual figures, although they are misrepresented. The number said to have died of thirst in England and Wales in the past five years, 667, is actually those in which dehydration was mentioned as a contributory factor on their death certificates. There were many fewer cases where dehydration was the underlying cause of death – 163 over the same period.
Today’s follow-up in the Mail – "Two patients a day die of thirst on NHS wards" – is more honest. Its second sentence says that dehydration “contributes” to 800 deaths a year in NHS hospitals. The number in which it is the underlying cause (128) is more like two patients a week than two a day.
This may seem like splitting hairs. To most people the figures are shocking, whether dehydration is the underlying or merely a contributory cause. They are not, however. especially suprising to those who have followed controversies over withdrawal of nutrition and hydration to patients in their last hours or days of life, or know anything about the process of dying. Guidance from the General Medical Council indicates that artificial nutrition and hydration are not invariably appropriate treatment.
The Mail’s figures came from the Office for National Statistics, as a result of a request. So when the Mail first edition dropped, other newsdesks were not able to follow up the story by accessing the same figures themselves. They had to lift it from the Mail, or leave the story alone.
This also explains why the Mail was able to get a follow-up story to itself today from the same set of statistics. Had others been able to access this analysis, the points about underlying or contributory causes might have been made, and the public better informed. I am told that the Mail ignored cautionary advice from the ONS about the data, but that’s par for the course.
What’s the answer? It’s right that ONS should produce analyses such as these on request, and encouraging that the Mail thought of asking them to. Many stories are enhanced by the right figures. It’s possible that the data accessed by the ONS in response to the Mail’s request is already in the public domain, somewhere – I haven’t got to the bottom of that – but it’s the assembling of the data which takes time. A busy newsdesk at 11 pm at night is not the place to attempt such research, nor are those working there likely to know how to begin.
But providing figures in such a way that they cannot be challenged by others is not acceptable. The UK Statistics Authority is already on the case, and what seems likely to result is a policy under which ONS would publish such analyses and make them available to all at the same time as they are sent to the journalist who originally sought them. That would be in line with Freedom of Information procedures.
An alternative might be to publish them at the same time as the story appeared, but that’s problematical because ONS may not know when publication is taking place or may be deliberately misled.   
For anybody interested in the tables, they have been published today on the excellent Timetric website.