What’s in a Number? Part 2
A few days ago, I wrote an entry about the fiasco surrounding the new cholesterol guidelines recommended by the American Heart Association (AHA) and the American College of Cardiology (ACC). What I didn’t discuss in that entry is the benefit of having predetermined goals as an aid to actually achieving the goal, whatever it might be. This is behavioral psychology 101 and the principles apply to a wide range of situations, including health care; having specific goals can be very important, whether it is a goal of achieving a particular cholesterol number, body weight, or diabetes glycemic control goal using HbA1c numbers. With respect to the cardiologists, it is clear that their goal was an overarching one- to decrease patient risks for the development and progression of cardiovascular diseases, clearly a worthy goal. But, people need more concrete goals if they are to work year after year at decreasing their cardiovascular disease risks; it is not enough just to say, “trust me, take this pill for the rest of your life, exercise, don’t get fat, etc., and don’t worry about those silly blood tests for monitoring your cholesterol level.”
HbA1c Numbers
I will assume that all of you know quite a bit about HbA1c, the blood test that quantifies the mean plasma glucose level over the previous few months and has been shown to be a reliable measure of a diabetic patient’s risk for the development and progression of diabetes chronic complications (if you never heard of HbA1c, check out my entry archives). The HbA1c test is an excellent example of how having specific number goals can improve patient care when patients and their health care providers set specific HbA1c goals and monitor the patient’s success in achieving the goals. You may not appreciate just how powerful this approach to care can be. These days, HbA1c testing is more or less standardized around the world with the test numbers providing well-validated information regarding patients risks for the development and progression of diabetes complications based on the landmark Diabetes Control and Complications Trial results. Patients and their health care givers have been bombarded with HbA1c numbers nonstop since 1993 ( like cholesterol numbers) and, at least in the U.S., few people with diabetes do not know what HbA1c measures and what the numbers mean and what is a “good” HbA1c level and a not so good level.
A few weeks ago, I met an 88 year old woman at my car repair shop. We were both waiting on our vehicles. We struck up a conversation and it wasn’t long before I knew everything about her life, including the fact that she had diabetes and that her “A1 thing” was 5.2 and both she and her doctor were pleased. She asked if I knew anything about that “A1 thing,” and I told her I had heard about it. We didn’t discuss her cholesterol level, but I would be willing to bet she knew her number and if it was where it was supposed to be. Anyway, number goals really do matter and patient’s “get it.”
Sweden’s screw-up
If you still don’t believe me about the importance of specific number goals in health care, let me tell you about what happened in Sweden a few years ago. The Swedes have been at the cutting edge of diabetes care for a long time. Well before HbA1c standardization to DCCT numbers was accomplished, but after the DCCT numbers had been published, the Swedes made changes to their HbA1c assay such that the test numbers reported country-wide were about 0.5 units lower than they had been for a long time. All physicians were made aware of this change. Yet, follow-up studies showed that after the assay was changed, patients’ HbA1c numbers nation-wide rose by about 0.5 units in relation to DCCT numbers, a clinically meaningful change. Of course, what happened was that the old HbA1c numbers were so well imprinted on patient and doctor brains, that when they saw a HbA1c test result they interpreted the meaning of the value based on their prior understanding of what an acceptable value was- they didn’t appreciate that they needed to add 0.5 to the reported test value to make sense of what the number really meant with respect to risks for diabetes complications. Horror-struck, the Swedes, readjusted their reference lab values to report the old numbers, and within a year patients’ HbA1cs were back down 0.5 units on average.
Goals do matter
So, if I go to see my primary care doc next week and she tells me I need to lose some weight, but doesn’t tell me roughly how much, how likely am I to hop right on her orders? And if she suggests I take one of those nasty medicines to help me lose weight and that I shouldn’t bother monitoring my weight, but just stay on the medication forever, does that make any sense? The cardiologists should have thought about this stuff before they caused chaos last week.