Saturday, September 13, 2014

The Nutrition Debate #244: Diabetes on Rise but Complications Decline (Duh!)


“Diabetes on the Rise but Complications Decline” was the title of a video with transcript on my Medscape Alert recently. The subtitle was “Diabetes on Rise as Waistlines Expand.” (That’s an association or “correlation,” not a cause of the rise. Obesity is a symptom of Metabolic Syndrome that includes a broken fat metabolism – but I digress.)

Both conclusions headlined above are based on epidemiological analyses of “sentinel indicators of outcomes” in diabetes collected from several very large public databases. According to the narrator, Henry R. Black, MD, from the New York University School of Medicine, “They examined 5 complications of diabetes – acute myocardial infarction (MI), stroke, lower extremity amputation, end-stage renal disease (ESRD), and death from diabetic ketoacidosis (hyperglycemic crisis)…” The “they” referred to above is Gregg EW, et al., whose ABSTRACT from their New England Journal of Medicine article can be viewed here and which is also listed as Reference #5 in the Medscape piece.

“We are all aware that we have epidemics of diabetes and obesity in the United States,” Dr. Black says, so he doesn’t attempt to explain that. Indeed, the NEJM article presumes that too, then postulates that “preventive care for adults with diabetes has improved substantially in recent decades,” and concludes that “rates of diabetes related complications have declined substantially in the past two decades.” Dr. Black’s comments and speculations are about the causes of the decline in complications. In that respect, in my opinion, both Dr. Black and the NEJM authors err in several fundamental ways.

But before I tell you how, let me first relate the specific “findings” of declining complications: “The diagnosis of diabetes increased from about 6.5 million in 1990 to almost 21 million in 2010, but the rates of major complications actually fell in all 4 age groups [studied].” There was a 69% reduction in MI cases, a 53% reduction in strokes, “but the rates of MI and strokes were about the same” (“when expressed as rates for the overall population”). According to Dr. Black, there were also fewer cases of amputations (“a reduction of more than 50%”) and fewer cases of end stage kidney disease (“a reduction of about 28%”), and hypoglycemic (sic) crises declined by 65%. (Gregg, et al. reported on hyperglycemic crises.)

Pandering to get published, or worse, maintaining an unquestioned acceptance of “perceived wisdom” to get “research” funding, is commonplace in today’s medical establishment. So saying, “Preventive care for adults with diabetes has improved substantially in recent decades” is nothing to get worked up about. I can give Gregg et al. a pass on that. But Dr. Black takes it to a new level with his myopic meanderings. “It is hard to pick any particular factor,” he says, but notes that “studies have shown that risk factor control, how you organized your healthcare system, and patient support for issues such as smoking cessation are all important in improving care, but certainly new drugs (e.g. statins) and such cardiovascular procedures as revascularization might help as well.” [Did he really call statins “new drugs”? He must be living under a rock.]

Repeating himself near the end of the video, Dr. Black says, “realizing how much we are estimating using these numbers, it could be better care, and the idea of using team-based care is promoted and probably an important factor. Medications are better (especially for lowering cholesterol), as are treatment of risk factors and patient education. Competing risks for ESRD might explain why that statistic didn’t decline as much – people live longer when they are not having heart attacks.” (Wow!)

What Doctors Gregg and Black fail to mention in either the abstract or the Medscape piece is that the increase in the number of patients with diabetes is in large part due to the changes made in the diagnostic criteria for type 2 diabetes.  In 1997, in the middle of the 2-decade study period, the threshold for diagnosing type 2 diabetes mellitus changed from a fasting glucose level of 140mg/dl to 126mg/dl. Then, in 2002, the American Diabetes Association defined prediabetes for the first time. People with prediabetes are at increased risk for developing type 2 diabetes and for heart disease and stroke.

Prediabetes is described as impaired fasting glucose (IFG,) defined as a fasting blood glucose of 100-125 mg/dl. But an IFG does not recognize insulin resistance which causes impaired glucose (carbohydrate) tolerance (IGT), defined as a glucose level from 140 mg/dl – 199 mg/dl two hours after consuming a glucose-rich drink. So, the ADA then introduced in 20XX the Glycated Hemoglobin A1c test (A1c for short) which measures (as a %) the glucose on the surface of your red blood cells (which have a life of +/- 3 months), thereby accounting for the postprandial, i.e. after meal, excursions your blood sugar took. This percentage converts into an estimated Average Glucose (eAG) value which sometimes now appears on your lab test results. And then the major medical societies, starting with the endocrinologists, lowered the level for diagnosing type 2 diabetes from an A1c of 7.0% to 6.5%. Now, an A1c between 5.7%and 6.4 is defined by the ADA as prediabetes. Note, however, that an increasing number of diabetologists now define an A1c of 5.7 as full-blown T2DM.
So, if you want to define changing standards for diagnosing type 2 diabetes as “preventive care for adults with diabetes has improved substantially” or “our care of patients with diabetes is getting better,” so be it. But changing standards for screening for and treating type 2 diabetes are the reason why the ratio of complications (numerator) to number of cases diagnosed (denominator) is improving. As the denominator gets larger, the fraction gets smaller. That’s 5th grade math.

No comments:

Post a Comment