Overtreatment of elderly diabetics

The last time I was directly responsible for treating diabetes was fifty years ago, when I was an intern in medicine at UCLA. In my subsequent career as a psychiatrist I was not directly responsible for diabetes care, and as an individual, I don’t have the condition. As a result, I haven’t kept up on diabetes treatment, so a June 11 article on “Diabetes Overtreatment in Elderly Individuals: Risky Business in Need of Better Management” was news to me.

The opening two sentences of the American Diabetes Association’s article on “Tight Diabetes Control” make it sound as if “tight control” should be the goal of treatment:

“Keeping your blood glucose levels as close to normal as possible can be a lifesaver. Tight control can prevent or slow the progress of many complications of diabetes, giving you extra years of healthy, active life.”

In my uninformed state, that’s how I understood how diabetes should be managed, even for over 65ers. But I was wrong.

Several paragraphs later there’s a very clear statement that elderly people with diabetes should be treated differently:

“Elderly people probably should not go on tight control. Hypoglycemia [overly low blood sugar] can cause strokes and heart attacks in older people. Also, the major goal of tight control is to prevent complications many years later. Tight control is most worthwhile for healthy people who can expect to live at least 10 more years.”

The American Geriatrics Society gives precise guidelines for the goal of diabetes treatment in over 65ers. The key measure of diabetes control is hemoglobin A1c. For healthy over 65ers with long life expectancy, the target should be 7.0 – 7.5%. For those with “moderate comorbidity” (so-so health) and a life expectancy of less than 10 years the target should be 7.5 – 8.0%. And with “multiple comorbidities” such as heart failure, cancer, and dementia the goal should be 8.0 – 9.0%.

In 1998, the United Kingdom Prospective Diabetes Study led to widespread belief that “intensive control” – typically meaning a hemoglobin A1c level below 7.0% – was the right target. But that study involved relatively healthy patients with an average age of 53, and over 65ers were excluded. But a study reported 10 years later in which the upper age limit was 79 and the average age of subjects was 63 concluded that “intensive control” produced more harm than benefit for the elderly, resulting in the new guidelines for treatment of over 65ers.

Despite the new guidelines, a recent study of more than 200,000 patients with diabetes who were over 75 and receiving insulin or oral anti-diabetic medication showed how poorly these guidelines are being followed. Fifty percent had hemoglobin A1c levels below 7.0%. In other words – half of this large group were being overtreated!

Why would this happen? The study did not probe the whys and wherefores, but I have two speculations.

First, doctors are notoriously slow to adopt new practices that contradict old ones. To some extent this reflects slow diffusion of information. I may simply not yet have learned about the new findings. But even when doctors are aware of the new perspectives – such as the guidelines promulgated by the American Geriatrics Society – we often continue to follow the familiar course of action.

This conservatism isn’t necessarily a bad thing. The stakes are high in medical treatment. Sometimes the “new certainty” turns out to be wrong. High dose chemotherapy and bone marrow transplantation in breast cancer is a prime example. Thirty thousand women received the highly toxic and sometimes fatal treatment before it was shown to be ineffective. Skepticism about pharmaceutical manufacturer-sponsored research is another reason for conservatism. In my practice, when patients asked about hot-off-the-press medications, unless there was a clear rationale for trying the new approach I typically suggested that we give the new nostrum a year to show its true colors.

My second speculation about the cause of overtreatment is that acknowledging the wisdom of the new guidelines for treatment of the elderly requires acknowledging vulnerability and mortality. We over 65ers are more sensitive to insulin and oral diabetes medications than we were earlier in life. This change puts us at greater risk for injuriously low blood sugar levels. But we, and our physicians, don’t always like to acknowledge greater fragility. Even more so with mortality. If an older patient asks “why should we control my blood sugar levels less rigorously now?” the physician may not feel comfortable saying “tighter control gives benefit in the distant future, and you are not likely to live that long.”

At 90, my late father-in-law was troubled by angina that kept him from walking as far and as briskly as he enjoyed doing. His cardiologist recommended a new surgical procedure. Early in his life my father-in-law had been in the plumbing business, and the surgery made sense to him as “clearing out the pipes.” The results were disastrous. His heart functioned much less well after the surgery. I was impressed with the thoughtfulness and generosity of his response. “My doctors meant well, but they were young men, and they were thinking of younger hearts than mine.”

Jim Sabin, M.D., 75, is an organizer of Over 65, a professor of population medicine and psychiatry at Harvard Medical School, and a Fellow of the Hastings Center. For a previous post on overtreatment, see here.

9 Responses to “Overtreatment of elderly diabetics”

  1. Bob Fenton

    This is just personal opinion, but the guidelines were changed because doctors will not prescribe insulin until they are forced into this and as a result they are afraid of hypoglycemia and being blamed for a death.

    In addition, because memory problems and various types of dementia, doctors have backed themselves into a corner and as such have raised the goals to keep patients on oral medications. This in turn will speed up the complications robbing patients of quality of life and an earlier demise.

    Some research now is saying if they had started insulin at diagnosis, the pancreas may have been able to repair itself to a degree and oral medications could possibly be used when memory problems happen. More research is still needed to confirm this, but it is becoming interesting.

    A friend who is near 77 just had an A1c of 5.3% and his endocrinologist went ballistic and asked him to leave the practice. No investigation about his blood glucose readings or anything. Mr friend has no other health conditions and is about five pounds under ideal weight. He has had diabetes for almost 10 years and tells me he was about 10 pounds overweight at diagnosis.

    • Jim Sabin

      Hi Bob

      Thank you for your comments!

      I don’t have enough understanding of the details of front-line diabetes treatment to respond to your views about insulin vs oral medications. It sounds as if your concern is more with UNDERtreatment rather than OVERtreatment. Knowing how complex science, individual biology, & the variation in MD temperaments are, I assume that both happen with some regularity.

      Re your friend with the A1c of 5.3 – it’s not at all clear to me why his physician asked him to leave the practice. Quite apart from the particulars of diabetes, “expelling” a patient from a medical practice is a very serious step from the dual perspectives of clinical care and medical ethics.



      • Bob Fenton

        Using this table will give you some idea for average blood glucose levels – http://www.diabeteschart.org/bloodsugarchart.html

        My friend had an HbA1c the last of February of 5.2% with only two excursions below 70 mg/dl and they were both in the mid 60’s. The endocrinologist had ordered him to bring it up to at least 7.0%. My friend had only said it will be what it will be.

        He is very diligent in counting his carbohydrates and injecting insulin accordingly. He obtains his medications and testing supplies from the VA. He still buys three containers of test strips per quarter above what he is supplied by the VA.

        The doctor had been the one to put him on insulin because of problems with metformin. His ability to obtain sufficient vitamin B12 from food had become impaired and he was very deficient in B12 and D. Required shots to bring him back to normal range. This has caused his neuropathy.

        In the almost three years on insulin, he has improved his HbA1c from about 7.0% to the 5.2% and has been under 5.5% three times.

        I am concerned about both over and undertreatment. Our support group sees both and we work with each other to promote the treatment that works best for the individual.

        • James Sabin

          Hi Bob

          Thank you for these further comments and the link to the glucose table, which was very helpful.

          Your description makes your friend’s relationship with his endocrinologist sound problematic. Nowadays the term “doctor’s orders” is only used tongue in cheek (at least by the physicians I know). From what I’ve read the endocrinologist had sound reasons for strongly recommending a target not lower than 7.0% for H1c. Your friend clearly had a preference for his own approach. Ideally they would discuss the pros and cons of the alternatives. If your friend understood the risks of hypoglycemia but did not change his view, I could imagine a physician talking with him about how to monitor for early warning symptoms and strategies for preventing dangerously low blood sugars.

          I recall a patient of my own who clearly had experienced an episode of mania and psychosis, and whose life history made the diagnosis of bipolar illness seem as close to certain as one can be in medicine and psychiatry. My patient (and spouse) interpreted the situation differently and were both very reluctant to use the medication (lithium) that I strongly recommended. They didn’t change their minds. I said “I’m not a worrier but I’m very worried about you. Let’s follow you closely and hope that you’re right and I’m wrong…” In fact another episode occurred, and the patient agreed to start a medication regimen.

          The difference for your friend is that the endocrinologist may have felt so uncomfortable prescribing an insulin dose that led to the low H1c levels that he concluded that in good conscience he could not continue as your friend’s doctor. In circumstances like that, medical ethics calls for transfer to another physician, ideally done in a respectful, “non-punitive” manner.



  2. Eric Reines

    Jim–The third reason for over treatment is “P4P.” Pay for performance programs of Medicare and private insurers are about ten years behind guidelines of specialty societies. A fourth reason is how some specialty societies value their own importance, which in turn may depend on the sources of income of some of the members of guideline-writing committees. Most of us physicians ignore the P4P and do what is right for the patient, and are supported in so doing by the AMA and MMS, but still P4P galls us. (p.s. request for clarification: the first link in your posting seems to be the original article upon which another article which you quote was based.) Thank you.–Eric

    • James Sabin

      Hi Eric
      Thank you for your comment. I apologize for the long delay in responding.
      If a P4P system rewards overtreatment of diabetes in the elderly that would reflect the height of stupidity. I share your distaste for P4P. I particularly loathe the common use of terms like “we should incentivize doctors to do XYZ.” When XYZ reflects good practice, the idea that physicians must be “incentivized” to do the right thing is an insult. And when the P4P target does not appear to represent good practice, physicians are faced with a choice between deviating from professional ideals versus being “dinged” for “inferior performance.”
      Thank you for pointing out my error in making a link – I’ve corrected it.

  3. Elizabeth Hamon

    I will be 87 next month and was diagnosed with Type 2 seventeen years ago. I moved a year ago and my new medical practice is very keen that I relax my goal of 7 which, combined with exercise, has served me well with no complications. However, if I allow my average reading to rise to 8.5 to 9 I am consumed by anxiety about the amount of blood sugar I have in my body so am returning to a maximum of 7.5 where I feel “safe”. So far I have never had a hypo or fallen into a coma during the night, which the nurse tried to frighten me would happen. On the contrary, if hungry during the night, I wake up suddenly. So I agree that each patient has individual needs which should be acknowledged by professionals and taken into account. I found this discussion very helpful.

  4. MO Palmer

    I just went through this. My dosage of Metformin was doubled to 2000 mg a day and I developed projectile diarrhea and fecal incontinence. My A1c was 6.2-6.7. I switched providers and am on 5 ml Lantus, My A1C is 7.2. My digestive problems – including frequent GERD – are by and large gone. I no longer soil my clothing. The new doc said an A1c of 7 or more at 73 is ok. I also have gastroparesis and have given up salads, veggies with peels, whole grain, high fiber – and have only had one episode of gastroparesis. I approve your study because treating me as a younger individual with no comorbidity about killed me, including two years of Parkinsonism from Reglan. I kept saying “one size does not fit all” but my degrees are in History, so nobody listened. Have not had one episode of incontinence while giving a speech. I’m living proof that diabetes in older people IS different. Now we need to look at BMI in older people and stop starving ourselves and working out until we drop

  5. Linda C

    My husband is 76 and has been a diabetic for 10 years, after by-pass surgery. His A1c is always under 6, usually 5.7 or lower. He has constant diarrhea and is weak. His endocrinologist is happy with his numbers. His primary doctor says he may be over medicated with 2000mg metformin per day and wants him to stop medication and monitor for 3 months. My husband is concerned with the difference of opinion.
    I have always felt that under 125 blood sugar readings after fasting were good even for someone have artery disease. Any comments?