"The NHS will last as long as there are folk left with the faith to fight for it"
Aneurin Bevan

Wednesday, 14 December 2011

Diabetes Audit

There is something a bit odd about the shock news released today that people who have a serious (but controllable) medical condition are not expected to live as long as people without it. As per usual, the BBC are reporting the news as if no one has ever been aware of it, but I came to the conclusion a year ago that the BBC have an agenda to broadcast as many articles as possible that puts the NHS in a poor light. (Not surprising really, since that is the agenda of the Department of Health and the BBC seems to be the biggest consumer of churnalism at the moment.)

Let's look at what the BBC are saying.

"Up to 24,000 deaths from diabetes could be avoided in England each year, if patients and doctors better managed the condition, a report concludes."
Let's look at the statistics. Diabetes UK say that there are 2.9 million people with diabetes (type 1 and type 2) or 4.45% of the UK population. They also say that 90% of diabetics have type 2 diabetes.

ONS says that in 2010 there were 493,000 deaths in the UK. Out of a population of 62 million that means slightly under 0.8% of the population. ONS provides detailed tables giving the cause of death.

The figures says that in 2010, 5,223 people died from diabetes. When just over five thousand people die from diabetes a year, the BBC says that almost five times more deaths can be avoided. How? Perhaps this statement from the BBC article explains why:

"Around 70-75,000 diabetic patients die every year." 
This is not the number of people who die from diabetes, this is the number of diabetics who die. Such people could be dying of cancer, stroke or heart attacks; in other words the same things that everyone else dies of. Diabetes may well exasperate these conditions, but that is not the same as dying from diabetes; ONS says so.

The BBC says:

For patients with Type 1, the risk of dying was 2.6 times higher than it was for the general population. With Type 2, the risk was 1.6 times higher.
The figures that I gave above from Diabetes UK about diabetes mortality are also in a PDF report on their website called Diabetes in the UK 2010: Key statistics on diabetes. In a section called Life expectancy and mortality it has these statistics:
More than one in ten (11.6 per cent) deaths among 20 to 79-year-olds in England can be attributed to diabetes. If current trends continue, one in eight (12.2 per cent) deaths among 20 to 79-year-olds will be attributable to the condition by 2010.
Life expectancy is reduced, on average, by:
– more than 20 years in people with Type 1 diabetes
– up to 10 years in people with Type 2 diabetes
Note the date: 2010. In fact, the figures for the reduction in life expectancy comes from a Department of Health report dated 2001. The 11.6% figure comes from a report in 2008. In other words, none of this is new.

Then the BBC report this:

The Department of Health in England said shocking variations in care and an unacceptable death toll were evident. 
Where did they get this impression from? Variation in the treatment of diabetes can be obtained from the Quality Outcomes Framework, that the Department of Health have been collecting since 2004.

Eh? They have known since 2004 that there were shocking variations and have done nothing about it? Are the Department of Health that incompetent?

The BBC report is churnalism of the press release on the NHS Information Centre. The news is not shocking, it is well known, the NHS IC merely have produced a report about something we all knew anyway. Perhaps the most interesting part of the NHS IC report is this:

"There is a strong link between deprivation and increased mortality rates. Among under-65s with diabetes; the number of deaths among people from the most deprived backgrounds is double that of those from the least deprived backgrounds."
This re-states something that was in the Diabetes Audit (pdf) from last year:
"In those aged 70 years and over, similar numbers of Q1 (12.1 per cent) and Q5 (14.8 per cent) have Type 2 diabetes, but under the age of 55 Type 2 diabetes is more than twice as common in Q5 (3.0 per cent) as Q1 (1.3 per cent). This may reflect lifestyle differences in exercise, diet and weight."
Q1 is the least deprived quintile of the population, Q5 is the most deprived. There is a clear message here: deprivation causes type 2 diabetes. The BBC didn't report that, did they?


  1. It gets worse. The press release (and BBC) quote the relative risks of dying ("Women between the ages of 15 and 34 with Type 1 diabetes were nine times more likely to die than other women of the same age. Men in the same age group were four times more likely to die if they had the condition."), without giving the absolute risks? Without knowing the absolute risks, the relative risks are, at best, meaningless. At worst they are highly misleading.

  2. I have type 2 diabetes. I have a yearly check up with blood tests, an hour with the diabetic nurse where she goes through my care bit by bit. I am on aspirin and statins as a preventative measure. I cannot fault the care given to me by my GP practice. I could not get better anywhere in the world. Stop knocking the NHS and give it the credit it is due. How many lives have been saved thanks to the NHS? Oh wait, thats not important is it.... just how much it can be belittled

  3. Interesting post, but I disagree with most points. The ONS data is incomplete as diabetes rarely recorded on death certificates. Just because previous reports concluded the same findings does not invalidate reporting of a contemporaneous study. And of course the DH has been trying to improve this since 2004 but it is a difficult situation to improve. However, the BBC cud have been clearer about distinction between dying with and from diabetes, tho.

  4. Furthermore, I think saying 'deprivation causes type 2 diabetes' is unsubstantiated by the evidence provided, and in any case over-simplifies the issue. I am a journalist, by the way, so I do consider these concerns about 'churnalism' etc carefully to avoid falling into that trap.

  5. Slightly off-topic and possible anecdotal, but I remember around the time leading up to the UK smoking ban, some misrepresentation of 'deaths from smoking-related diseases' where anyone who died of (eg) lung cancer - whether or not they ever smoked - was classed as dying from a 'smoking-related disease'.