The ACCORD Trial Findings: What You Should Know

In one large, ongoing trial of people with Type 2 diabetes and a high risk of cardiovascular disease, intensive blood glucose control has been linked with a slightly higher risk of death compared with less-intensive, "standard" treatment. Participants in the trial who were receiving intensive treatment will now be switched over to standard treatment. While these findings were unexpected and have raised some concerns about diabetes treatment strategies for specific groups of people, they should not trigger a change in therapy for most people with diabetes.


The study, known as the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial, is being conducted at 77 sites in North America. It has over 10,000 participants between the ages of 40 and 82 who have been involved with the study for between two and seven years. The participants all have Type 2 diabetes and especially high risk for heart attack and stroke, measured as having at least two risk factors for cardiovascular disease in addition to diabetes or as having been diagnosed with cardiovascular disease before the start of the study. The study was designed, in part, to test whether intensive blood glucose control would reduce the risk of heart attack, stroke, death from cardiovascular disease, and other “cardiovascular events” in this high-risk group.

The participants who were randomly assigned to the intensive control group aimed to lower their HbA1c levels (a measure of blood glucose over time) to less than 6%, or close to nondiabetic levels. In reality, half the participants in this group achieved an HbA1c level of less than 6.4%. Meanwhile, the standard treatment group’s goal was to achieve an HbA1c of between 7% and 7.9%, or close to the average HbA1c of people with diabetes in the United States. In reality, half the participants in this group achieved an HbA1c of less than 7.5%. Over the course of the study so far, both groups have lowered their average blood glucose levels compared to their levels before the study began.

The intensive treatment arm of the study has been stopped 18 months early, however, because more people in the intensive treatment group died compared to the standard treatment group. The difference was small—overall, 257 people in the intensive treatment group died and 203 people in the standard treatment group died, a difference of 54 total deaths or 3 deaths per 1,000 participants per year of the study. In both groups, the death rates were lower than had been seen in other studies of people with similar risk factors.

As the ACCORD trial continues, the researchers will seek an explanation for the difference in death rates between the groups. Analyses so far have not found a specific cause and have not implicated hypoglycemia or any blood-glucose-lowering drugs or combinations of drugs in the increase in deaths.

In light of the ACCORD trial’s findings, the American Diabetes Association (ADA) has released a statement advising people with diabetes who have cardiovascular disease or multiple risk factors for it to consult with their diabetes care team about their treatment goals and to make sure that their blood pressure and cholesterol levels are being properly controlled. However, the ADA also said that it “strongly encourages people with diabetes not to alter their course of treatment without first consulting their health care team” and that it “continues to encourage good control of blood glucose for the management of diabetes and its complications.” The statement went on to say that achieving an HbA1c level of less than 7% has been shown to reduce diabetes complications and “appears to be of great benefit rather than harm.”

The ADA currently recommends a general HbA1c goal of less than 7% for people with diabetes, but notes that treatment should be tailored to the individual. For some people it may be best to work toward an HbA1c level as close to normal as possible, while others—such as children, the elderly, and those with other medical conditions—may do best with less stringent goals. Some other groups, such as the American College of Endocrinology and the American Association of Clinical Endocrinologists, recommend an HbA1c target of 6.5% for people with diabetes.

Some of the media coverage of the ACCORD trial findings has been alarmist, but it is important to note that the trial is looking at a specific, high-risk group of people and that its findings may not be generalizable to all people with diabetes. In fact, another study published this week in The New England Journal of Medicine showed a reduction in overall death and cardiovascular death associated with intensive blood glucose control in a different high-risk population. The Steno-2 study, which followed 160 middle-aged people with Type 2 diabetes and microalbuminuria (kidney damage) found that people who underwent intensive therapy—with goals of an HbA1c of 6.5% or lower, cholesterol of 175 mg/dl, triglycerides of 150 mg/dl, and blood pressure of less than 130/80 mmHg—had a 20% reduction in risk of death and a 12.5% reduction in risk of cardiovascular death compared to those who received their usual treatment.

Previous trials, including the landmark United Kingdom Prospective Diabetes Study (UKPDS) of people with Type 2 diabetes and the Diabetes Control and Complications Trial (DCCT) of people with Type 1 diabetes have shown that intensive blood glucose control can substantially lower the risk of diabetes complications such as neuropathy, nephropathy, and retinopathy. (You can read more about both of these studies in the article “Managing Hyperglycemia.”). In addition, the Epidemiology of Diabetes Interventions and Complications Study (EDIC) study—a follow-up to the DCCT—found that intensive blood glucose control in people with Type 1 diabetes substantially reduced the long-term risk of developing cardiovascular disease.

Learn more about the health and medical experts who who provide you with the cutting-edge resources, tools, news, and more on Diabetes Self-Management.
About Our Experts >>

  • Raving Rabbi

    Best guess: There has always been a correlation between higher levels of blood insulin and heart incidents (like heart attacks and strokes). Type 1 people balance their food and their insulin. Type 2 have some resistant insulin floating through their blood stream.
    My prediction is that in the future, there will be a big push to have diabetics test their blood sugar AND blood insulin levels to keep them in balance. There will also be a bigger push to get diabetics to eat mainly proteins & low-glycemic index foods that don’t shoot as much insulin into the blood stream. There will be a much bigger push to lower sugar through exercise, and to cut between-meal snacking even if it means eating larger meals. (There might even be a push to get diabetics to fast one day a week, to clear out their blood and give their pancreas a day of rest.) There will be more people on medications to destroy insulin 2-3 hours after meals.

  • Raving Rabbi

    But first, they have to establish a connection. They must ask:
    1) Were the people injecting insulin dying at a faster or slower rate than those that were on meds alone?
    2) Were those with higher cholesterol (which correlates to higher insulin levels) dying more often than those with controlled cholesterol?
    3) Was there more pancreatic damage among those that died than those who lived?
    4) Were the diets of those who died filled with fast-digesting carbs? What were the “last meals” of those that died – is there any link between what they ate in the previous 24 hours and the sudden malfunction of their heart? Did those who exercised more have lower incidence of sudden death? A higher percentage of men or women?
    5) They will have to chart those deaths, and see if there was any “bell curve” or trajectory to determine how long after these people achieved blood sugar control they died.

  • CalgaryDiabetic

    One wonders if having blood sugars in the 50 to 70 damages the heart muscle because of energy deprivation. In this range many people may not have symptoms of hypoglycemia. When the study exonerates low blood sugar that makes you wonder how could they know ? Without continuous accurate monitoring presently difficult, it is not possible.

  • Joan

    The ACCORD Report: My view after reading several reports from a variety of news sources is — — it depends on the individual’s personal health – for heavens sake!

    This is yet another “scare” topic that is familiar to most of us these days! Naturally, one must be careful. My doctor has put me on or taken me off medication due to ME and its effect on ME!

    To be aware is a necessary tool for some and not so good for others. Education is goof for everyone. It is how wisely we use the education received that counts! There are too many people with diabetes who do not rationalize at all! This in turn affects their stress levels and thus their daily control!

    Sometimes – it may pay to NOT report so much except to medical professionals! Hmmmmm????

  • Camanche Bill

    Ther is an old saying-” Statistics don’t lie, but statisticians do” There seems to me to be some of that going on here. They need to ask the right questions,as the Raving Rabbi points out.

  • CalgaryDiabetic

    I agree with Raving Rabbi that minimising insulin and replacing it with exercise whenever possible may be a good thing. I just found out that there is another study under Dr. Gerststein’s supervision funded by Sanofi that is going to look at the correlation between intensive treatment and death called ORIGIN. Has anyone out there heard of any preliminary results? Final results due in OCT 2009. Will be funny if both studies contradict each other.

  • CalgaryDiabetic

    The world study larger than the Accord study called “Advantage” reported very recently that they could NOT support the “Accord” findings. Like in politics the US seems to be unique in the world. The “Origin” study under Dr.Gerstein(a poet of insulin) may shed some light. Is insulin good or bad for the heart? The poets claim good other literature says bad. No wonder Diabetics have 3x the rate of depression of normal civilians. A review of the literature of the effects of insulin on the heart would be usefull but could be so contraditory as to be depressing.

  • Shzron1946

    I don’t know why it has taken so long for me to find this blog. I am in the ACCORD Intensive control group. It has changed my life. African American woman in her 60s, obese (300+ #)Type II diabetes, high cholesterol, high blood pressure, hypothyroid condition, a heart failure patient. This is a medical description of me. My A1C was 9.7 and ACCORD got it down to below 6. I could think better and live a better life. I had a master’s degree, but couldn’t think well enough to practice my profession, but my thinking is much clearer now.

    Here’s what I think. Something is wrong with how they are looking at the data. Maybe something is wrong with one of the meds. You know people who died may have been taking a med that was discontinued earlier in this study–I know there was at least one, a heart medication. Researchers should hunker down and be more precise about looking at the data, because one thing I know for myself that no one can change my mind about is that the lower A1C worked for me.

    I am sorry the study is ending. I know I will never have such good care. Doctors just don’t take the time and health insurance wont allow the medications. I will have to work much harder to maintain, I know.

  • MelodyL

    I am also in the ACCORD program. I go to Cornell in NYC. It was the best thing I could do for myself. My A1c was 9 point something when I enrolled. Now it’s 6.7. I have lost MOST OF MY WEIGHT.

    I went to a nutritionist. And best of all, I am now only on 18 units of Lantus once a day. No other diabetic meds. I used to be on all kinds of oral meds.

    Does anyone know when the ACCORD program ends? I will be so sorry when this happens.

  • psdaengr

    the federally sponsored trial called ACCORD was stopped in February 2008 because there were 20 percent more deaths among diabetics with heart problems who got intensive treatment compared to those who were treated more conservatively.