ACCORD Study: Wrong and Wronger

The media loves a good scare. So they’ve jumped all over the failed ACCORD study (Action to Control Cardiovascular Risk in Diabetes) with headlines making it sound like tight blood glucose control is now proven to be “bad for you.” Ugh.

As if it weren’t difficult enough educating the population about diabetes — and educating so many people with diabetes about how to treat it — now we’re supposed to question whether we should forgo blood sugar targets altogether? Geez.

I’m certainly not the first to point out why and how WRONG these sensational headlines are. But I felt compelled to compile some of the best points being made:

* The media is for the most part NOT MAKING IT CLEAR that the ACCORD study results do not extrapolate to the lion’s share of the diabetic population, and that younger people not at risk of heart disease have nothing to fear

* As Close Concerns notes, the study organizers are NOT DISCLOSING DETAILS on drug combinations and doses. “Usually announcements like this come with publishing of data, but not in this case” — which makes it even harder to draw conclusions

* The intensive treatment group in this study had a target blood sugar goal of LESS THAN 6 PERCENT. As anyone with diabetes knows, striving to go that low means frequent hypoglycemia (or risk thereof) which to my mind can’t be good for people with cardiovascular issues — even though the study organizers claim it wasn’t insulin shock that killed all those patients

* Expert endo Dr. Irl Hirsch believes it may have been the speed of the A1c drop that resulted in trauma — he references a similar study about retinopathy that showed something along these lines

* From the latest NY Times story, another theory offered up by Dr. John Buse, vice chairman of the study’s steering committee and the president of medicine and science at the ADA: the sheer stress of the program may have been a factor. “The intensity of what we did is done virtually nowhere on the planet… It’s far beyond what’s common in clinical practice,” he says. More proof that most of us have nothing to worry about striving for “normal” tight glucose control

[According to that article, "Many patients with diabetes feel stressed when they fail to meet blood sugar goals set by their doctors." I have no words...]

* Those not effected by this study: Type 1s, newly diagnosed Type 2s, and any and all younger patients who do not have cardiovascular disease

* If you DO have cardiovascular disease or high risk of it, “The most important thing is get your blood pressure controlled, cholesterol controlled, and do a reasonable job on your diabetes, but don’t go wild,” Buse says. “We are backing away from notion that we always have to push, push, push to get blood sugar lower.”

[Kind of the message we were hoping to bring across in our Know Your Numbers book]

So I hope a few things were learned here from this sweeping study. It’s just a damn shame that 257 people had to die in the process.


16 Responses

  1. Neal
    Neal February 8, 2008 at 12:37 am | | Reply

    Glad to see your comments on the ACCORD Study. I read this news story form a variety of sources trying to convince myself that something was up. How could reaching near normal hypoglycemic targets be advised against? This is one thing that internet junkies and particularly the diabetes community needs to be aware of, as we are constantly combing the web for news/etc. on how to improve our control. Stories like this wherein the media jumps in before all the details/outcomes have been elucidated serve no purpose but to confuse/hype. It would be nice to have news stories peer-reviewed, by objective & informed sources, before they’re published. But deadlines and revenues do not allow for this. This is yet another example of “soft journalism” being passed off as “hard science.”

  2. Not Funny
    Not Funny February 8, 2008 at 5:26 am | | Reply

    I feel so much better now after having read what you wrote today considering that I have Type 1.

  3. Brian Peterson
    Brian Peterson February 8, 2008 at 5:48 am | | Reply

    You forgot to mention how the media is blaming Avandia. I would be surprised if the endos’ phones aren’t lighting up right now with patients asking to be switched to another medication.

  4. Carey
    Carey February 8, 2008 at 6:01 am | | Reply

    Hi Amy,

    Maybe we need infiltrate all media outlets with a representative from the D community. Just to keep an eye on them.

    I work for a major news agency. I was pretty shocked when our story hit the wire with this headline:

    ‘Diabetics Who Tightly Control Sugar Have Higher Death Rate’

    We also didn’t specify that this was a study involving high-risk type 2 patients until very deep into the story.

    I immediately spoke to the editor of the story and told him that we can’t have this. I told him the headline alone was not only misleading, but jarring enough to give healthy type 1s a heart attack. 20 minutes later, the story had a different complexion.
    Headline was:

    ‘Intense Therapy for Type 2 Diabetics Tied to Higher Death Rates’

    And in the story, we now noted that the results of the study did not apply to type 1 diabetics.

    At least one media story eventually got it right.

  5. AmyT
    AmyT February 8, 2008 at 7:15 am | | Reply

    Wow, good for you, Carey!

  6. gretchen
    gretchen February 8, 2008 at 7:19 am | | Reply

    Type 2s can often get their A1cs below 6 without hypos, especially if they’re not on insulin or sulfs.

    It’s more difficult for type 1s, and it’s especially difficult for type 1 children on high-carb diets.

  7. Kathy
    Kathy February 8, 2008 at 9:01 am | | Reply

    I agree that the reporting of those results has been poor at best. However, I wonder about the relationship of high blood sugar on cholesterol levels. When my A1Cs were high years ago, I also had high triglycerides. My doc at the time told me better control would bring those down, and voila, that’s what happened.

    As a T1 the results don’t ring any alarm bells for me. It just tells me that T2 really is a different disease.

  8. Anne
    Anne February 8, 2008 at 9:04 am | | Reply

    I think about 200 people died in the less intensive group, so the difference between super intense and moderately intense control groups was on the order of 50. Still, it is 50 too many.

    Although my A1c has been a little higher than possible over the years, I have exercised 5-6 days per week for about 19 years (since I was diagnosed). I don’t have any complications and wonder how much the exercise plays a role in that. I wish there was a study that looked at the role of exercise in preventing complications. Maybe there has been?

  9. Anne
    Anne February 8, 2008 at 9:05 am | | Reply

    also, if I remember correctly, both treatment groups fared better than those left to their own devices? maybe I’m wrong on this.

  10. Hans Tilsner
    Hans Tilsner February 8, 2008 at 11:24 am | | Reply

    It is no secret that after years with A1cs of 9 and 10 and higher, as was the entry for ACCORD participants, any Type2 will go hypo when getting below 150. And after a few weeks of systematic BG lowering he will still show clear hypo-symptoms when getting below 100!

    So with their diabetes histories there is no need for ACCORD participants to get below 50 to experience a real good low. And that’s what they must have done and almost all day long to achieve what great medic brains had set as their healthy goal. – Sorry, but that was therapy by crowbar :-(

    Hans, type2 since 1990, Germany
    A1c 5, Apidra & Lantus

  11. Big_Dave_T
    Big_Dave_T February 8, 2008 at 11:46 am | | Reply

    Thanks for your perspective, Amy. I was getting worried that you were getting too commercial and not going to comment on this type of stuff anymore.

    I do appreciate hearing finally that there COULD BE a downside to strict control, which I always believed anyway. But big pharma has been pushing, pushing, pushing for more pills, gadgets and the like so that T2s like myself can reach a number that seems unrealistic and, now as we find out, probably unhealthy too.

  12. Jules
    Jules February 9, 2008 at 9:39 am | | Reply

    I still kind of wonder if it isn’t the insulin that does most of the harm. Type 2′s usually still produce insulin. Since they are resistant, they need more insulin. By taking additional insulin, they continue to add more and more. I try to keep my total daily insulin below 30 units. Maybe a study should be done to actually measure circulating insulin and the risk of complications from higher levels of insulin rather than glucose levels.

  13. wschaf
    wschaf February 10, 2008 at 12:44 am | | Reply

    Several things about the study need to be stressed in order to understand what the study was apparently trying to accomplish. The study was an attempt to see which is more important for prolonging life in high cardiac risk patients: tight control of blood glucose, tight control of lipids, or tight control of blood pressure. It certainly was successful in finding one of the end points, that overly tight BG control can be detrimental. One thing that should be emphasized is the degree of control attempted – the NHLBI site shows that half of the participants maintained A1c’s at or below 6.4 [compared to 7.5 for standard control]. This may indicate that the goal of less than 6.5, advocated by many endocrinologists may be too stringent. The tight BG control group with the 20% excess deaths – the PROPER way to express the results – may, indeed, have been overstressed, as seen in excess deaths for both cardiac and non-cardiac [mostly cancer] causes. Stress can kill.
    It will be interesting to see what the results of the blood pressure and lipid trials will be. I would imageine that the tight control for lipids is LDL less than 70 and systolic BP less than 120 versus 100 and 139 respectively. [And personally so because it is easier fop me to control BP and lipids than blood sugars - SBP's 118-128, total chol. at 100, LDL at 55 and A1c's in the 7-7.8]. I am a Type1-LADA for 32 years.

  14. HealthBlawg
    HealthBlawg February 11, 2008 at 9:07 pm | | Reply

    Grand Rounds Vol. 4 No. 21: The Valentine’s Day Edition

    Happy Valentine’s Day (almost), and welcome to Grand Rounds, the weekly rotating carnival of the best of the medical blogosphere, hosted improbably this week by — horrors! — a health care lawyer and consultant. I’ve organized this edition around themes

  15. Migraineur
    Migraineur February 13, 2008 at 9:45 am | | Reply

    The other thing about this study is that they only used one method of tight control – drugs, drugs, and more drugs.

    I wonder what they’d see if they got patients’ A1c down by drastically reducing sugar and starch consumption.

  16. Dan Fahey
    Dan Fahey February 15, 2008 at 6:53 am | | Reply

    I’m a Type 1 of 40+ years, whose a1cs are generally at or slightly below 6.5.
    For the first time my a1c dropped below 6 [5.8] and I WASN’T pleased!
    To me that means I’ve been doing too good a job: I’m mean I struggle with lows when I’m in the mid to low 6′s.
    Getting a 5.8 was accidental I can assure you, and I feel this way not because of the recent article about deaths but rather becuase I know the balance between good control and the effects of hypoglycemia is so important.

Leave a Reply