Wednesday, February 03, 2010

Good Targets, Bad Methods

There has been a lot of discussion in the media and the blogosphere about an Early Online Publication in the Lancet on 27 January 2010.

Survival as a function of HbA1c in people with type 2 diabetes: a retrospective cohort study

"Methods
Two cohorts of patients aged 50 years and older with type 2 diabetes were generated from the UK General Practice Research Database from November 1986 to November 2008. We identified 27 965 patients whose treatment had been intensified from oral monotherapy to combination therapy with oral blood-glucose lowering agents, and 20 005 who had changed to regimens that included insulin. Those with diabetes secondary to other causes were excluded. All-cause mortality was the primary outcome. Age, sex, smoking status, cholesterol, cardiovascular risk, and general morbidity were identified as important confounding factors, and Cox survival models were adjusted for these factors accordingly


[snip]

Interpretation

Low and high mean HbA1c values were associated with increased all-cause mortality and cardiac events. If confirmed, diabetes guidelines might need revision to include a minimum HbA1c value."

What has saddened but not surprised me is that the reaction in on-line medical discussions has been much the same as that when the ACCORD and ADVANCE studies appeared. No-one has questioned the methods used to attain targets, all appear to accept that there may be a problem with attempting to aim for tight targets for type 2 diabetics. Thus, the targets are bad and should be eased.

I was not going to write about it because so many others have, but eventually I responded in Present where a doctor posed the following question based on this study: "Are we treating our patients to death?"

This was my response to the doctors.

The answer depends, of course, on the method of treatment and the individual nature of your patient's diabetes.

All of these studies, including ACCORD, ADVANCE and these recent ones have a common thread. They presume that the only way the physician can reach lower A1c and blood glucose goals in a patient is by medication. None of them consider using lifestyle – diet and exercise - changes to complement minimal medication or insulin to achieve those goals.

You all presume that a patient would not use diet and exercise. And, of course, to some degree you are correct, especially when the most generally prescribed diet is directly counter-productive for achieving better A1cs and blood glucose levels.

So you prescribe diet and exercise, the numbers go up, you presume non-compliance and prescribe metformin or a sulf or add insulin, they keep going up so you increase those and add more meds. But you also stress that the patient should eat more carbohydrates – and less fat - to be absolutely sure there are no hypoglycemic episodes as a result of the sulfs or the insulin and so the cycle continues, chasing it's tail.

Lower blood glucose and A1c targets do not cause higher mortality and morbidity. I have read success stories from many thousands of pro-active diabetics of all types on many forums since I was diagnosed eight years ago who have clearly demonstrated the opposite is true.

On the other hand, over-medication to attempt to counter poor dietary advice DOES cause higher mortality and morbidity; that is the consequence these studies are showing.

The solution? First, stop promoting the terrible “heart-healthy” high-carbohydrate AHA diet to your patients, or allowing the dieticians you send your patients to to do so. Instead, suggest that the patient use their meter at their post-prandial peak blood glucose timing to find out what foods are killing them and they will quickly reduce those foods, and their levels, and substitute others. The technique is described here:
http://loraldiabetes.blogspot.com/2009/04/test-test-test.html

Some patients will do that and succeed; some will need further medication, but much less than you usually prescribe. And others will be non-compliant whatever you prescribe. They are the ones to prescribe higher medication to – but they are also the ones you should set easier targets for. Because they are the ones dying from over-medication.


Cheers, Alan.

Everything in Moderation - Except Laughter.

4 comments:

Anonymous said...

I truly do not understand these people's reasoning, If they think a low A1C is so dangerous, then what about non-diabetics? Are they dropping like flies because they don't have a 6-7 A1C? (Other than the high carb heart attacks.) So it would seem to me that closer to "normal" should obviously be better, trending the health risks down toward the non-diabetic's risks.

By the way, I've run into your name on several diabetic forums and find your posts are always helpful and gentlemanly. Unfortunately, some of your "opponents" can be, let's say less than polite, which is why I won't be joining any forums. Keep up the good work. The knowledge you've gained has certainly been of great help to me and who knows how many others.

Bob Fenton said...

Thank you Alan! This is why I enjoy reading your blog. You have a way of making a point that others only hint about and leave before making a solid point.

Or if they do make the point, leave you wondering why they even bothered - often missing a key component or element of the study.

Bob

Michael Barker said...

Most excellent. These studies are never viewed properly. They should be seen as an indictment of diabetes education and the tools which we are provided to accomplish tight control. The goal should not be questioned. What should be questioned is the poor methods that are put forth to us in order for us to meet this goal.

Rekarb

trinkwasser said...

Wow. you've been busy while I was looking the other way!

"I truly do not understand these people's reasoning, If they think a low A1C is so dangerous, then what about non-diabetics? Are they dropping like flies because they don't have a 6-7 A1C?"

I posed just that question on a UK doctors' forum, I suggested if they thought it was so advantageous THEY should put their A1c up first.

No reply, obviously