Total Pageviews

Sunday 3 February 2013

If you can’t score a goal, move the goal posts !


Regular readers here know we never tire of publishing the NHS annual audits regarding diabetes in the UK. Year in and year out the abysmal lack of progress is presented. Of course the situation regarding numbers of confirmed diabetics goes forever up, the epidemic of type two diabetes continues unabated. The stats on HbA1c which give an illustration of how well controlled a diabetic is, are woeful in the extreme. So, what do you do if you want to give the impression you are making some headway, what do you do if you can‘t score a goal ? Improve treatment and reduce the grim numbers, not in the UK. In the UK they move the goal posts.

Results for England. The National Diabetes Audit 2010-2011
Percentage of registered Type 1patients in England
HbA1c >= 6.5% (48 mmol/mol) = 92.6%
HbA1c >   7.5% (58 mmol/mol) = 71.3%
HbA1c > 10.0% (86 mmol/mol) = 18.1%

Percentage of registered Type 2 patients in England
HbA1c >= 6.5% (48 mmol/mol = 72.5%
HbA1c > 7.5% (58 mmol/mol) = 32.6%
HbA1c >10.0% (86 mmol/mol) = 6.8%

These results are very similar to those obtained in previous NHS audits over the past 5 - 6 years.

"An NHS prescribing advisory body has called for a change in diabetes guidance to include a minimum HbA1c target level, reflecting the weight of trial data suggesting lowering blood sugar below a certain level may harm patients.


In the light of mounting evidence that driving levels down to below the 7.5% recommended in the QOF increases the risk of cardiovascular events.
The latest MeReC Extra includes an analysis of the UK primary care trial published in The Lancet in February, and follows an editorial in the BMJ which also questioned the clinical value of an HbA1c value of less than 7.0%.
The Lancet paper, a retrospective study using data from the UK General Practice Research Database, identified that a median HbA1c level of about 7.5% was associated with the lowest risk of all-cause mortality and macrovascular disease events.
But it also showed there was a U-shaped association between increased all-cause mortality above and below an HbA1c of 7.5%. Changes above or below an HbA1c of 7.5% was associated with a greater risk, regardless of whether treatment was intensified with oral hypoglycaemic agents or insulin injections.
Compared with the reference group who had a median HbA1c of 7.5%, all-cause mortality in the 6.4% HbA1c decile was 52% higher and 79% higher in the 10.6% HbA1c decile.
The May MeRec Extra suggested GPs ‘may wish to consider the implication of the study in their discussion with patients about risks and benefits of intensifying drug treatment and setting of individual HbA1c targets.'
It concluded: ‘If this ‘U-shaped' relationship between HbA1c levels and mortality in patients with type 2 diabetes receiving combination blood glucose lowering treatment is confirmed, this would add weight to the view that diabetes guidelines might need revision to include a minimum HbA1c value.'
The NPC review said that this evidence, in addition to similar findings from the ACCORD, ADVANCE and Veterans Affairs diabetes trials, which also failed to find a consistent significant benefit of intensive glycaemic control on cardiovascular outcomes and mortality, ‘added weight' to the possibility of a minimum HbA1c target."


As you can see the deeply flawed ACCORD trial is being trotted out again.

Eddie

More here.

5 comments:

Anonymous said...

The worrying fact is that when the 2011-12 results are published and I do not think they have been yet these figures could quite well be worse.Nothing has changed in recent years and diabetes and obesity is still rising sharply.It would be most interesting to note what figures are like for 5.5% these of course never appear.

Joe

Lowcarb team member said...

Hi Joe

I read some time ago these audits may be discontinued because of the costs involved. The NHS and DUK etc. would love these audits to disappear for obvious reasons.

Eddie

Anonymous said...

Yes, they would like these awful results not to be seen, glad to see you show the truth
Jeff.

Anonymous said...

Sorry to recycle - but here is my post from ETYM on this - it is beyond belief that this can be proposed on the basis of the trials they discuss.

This makes me f**cking incandescent with anger; look at the number of diabetics who 'hit' the targets according to the audits that Eddie frequently cites; it's atrocious. The response here is to clearly say well let's lower the targets then.

It makes no sense at all that 'normal' HbA1c's should increase your risk of CVD! How can that be? Do non-diabetics suffer from diabetic complications? Our blood sugar levels have evolved to be very very precise, now because diabetic treatment is so bad that even double the non diabetic levels cannot be managed these presumptuous oafs say; well 65 million years of mammalian evolution must be wrong then and look we have the half-arsed studies to prove it.

Look at the language they are using "VADT trial finds no differences in outcomes between patients with average HBA1c of 8.4% and those intensively treated to 6.9%." INTENSIVELY TREATED is the key thing. This is exactly what the ACCORD study failed to note; if you stuff people full of powerful drugs things may go badly.

More importantly 6.9% is a very high HbA1c compared to a non diabetic. We all know the figure of 7.8 mmol/l as the level at which damage starts to be done; that equates to an HbA1c of 6.1% so to have any meaningful comparison you need one group of people below 6.1% and one group above. But you can't get the lower level group because most UK diabetics are so badly served that that is a dream level to attain.

ACCORD to me is the same as deciding to do a study of weight loss by removing people's legs with a chainsaw. Guess what; the people will lose weight they will also die, to then say 'well that just goes to show how dangerous weight loss is' misses the point by about as much as a point can be missed.

Always ask what study your HCP means when they talk about them showing a danger; if it is ACCORD ignore them. If it is the DCCT or the UKPDT then both show marked improvements in lowering HbA1c, so ignore them again.

This is Orwellian Double-Think and it will kill people.

First do no harm? Ha f**cking Ha.

Dillinger

Steve P said...

This ACCORD study has alot to answer for - and all bad 'fraid to say. It is so important to read through all notes and case studies used, unfortunately not all do.
Great comment Dillinger