Statistics

The United States of Diabetes?

More fun can be had with epidemiology than with any other branch of investigation. It’s the only science whose practitioners regularly jump out from behind the bushes yelling Boo! And even though the strolling citizen knows the fright is coming, even though he can see the man in the white frock lurking in the foliage ahead, and even though he has walked past this very spot hundreds of times before, the finale unvarying, he never fails to shriek and cower after the trick is sprung. The citizen stands and delivers, and gratefully, too.

The epidemiologists at the United Health Group have just burst from behind the ornamental barberry with this press release, announcing a new report:

“More than 50 percent of Americans could have diabetes or prediabetes by 2020 at a cost of $3.35 trillion over the next decade if current trends continue.”

More than fifty percent! Boo!

Wait a second. Strike that. Something fishy here. By 2050, forty to fifty percent of all citizens will be under 30 years of age. This of course implies that fifty to sixty percent will be 30 or older. Now, type 2 diabetes is a disease primarily of the old, and while some under 30 may have it, most of those afflicted will be older than 50 or 60.

Thus, taking the United Health Group’s press release at its word tells us that nearly every adult by 2050 will have type 2 diabetes or prediabetes. No Halloween prank was more frightening than a statement like this.

One should never rely on press releases, whose prose sinks below even advertising copy in reliability. Let’s examine what the actual report said:

“The United States of America is on track to become the United States of Diabetes. The epidemic of type 2 diabetes and its warning sign, prediabetes, is sweeping across the country. By 2020, an estimated 52 percent of the adult population will have diabetes or prediabetes.”

Fifty-two percent! Boo! More than half the citizens of the good ol’ USA will have type 2 diabetes, a horrible disease!

Wait a second. Strike that. A closer reading reveals that everybody won’t fall prey to type 2 diabetes, but also to the strange malady called prediabetes. In technical terms, and according to our watchful government, prediabetes “means your blood glucose is higher than normal but lower than the diabetes range.”

Thus, in plain English, prediabetes translates to not-diabetes. With this definition in mind, we can see that the statisticians at United Health have grossly underestimated the number of adults living with type 2 diabetes or prediabetes. That number won’t be fifty-two percent, but will be roughly double that, a truly startling number.

Isn’t having blood glucose higher than normal but lower than a diabetes a bad thing? Well, no, not really. It is true that those who have diabetes started with blood glucose levels lower than diabetes levels yet higher than normal. But it is also true that many of those who have blood glucose levels lower than diabetes levels yet higher than normal never develop diabetes. For them, life is sweet.

The best way to state this is that those in the “normal” range are at lower risk of developing diabetes than those in the higher than normal range. And let’s not forget the crucial addendum: whoever is at risk from the disease does not have the disease. Being “at risk” is not equivalent to “having” a disease.

To risk tediousness, let’s restate the fact that prediabetes means not-diabetes. Therefore, what is the use of forecasting an increase in prediabetes rates? Except to needlessly frighten, there is no good reason at all, especially if you are simultaneously forecasting actual diabetes rates.

Just think. The population in 2050 will consist of three subpopulations: diabetes free, prediabetes, and diabetes. The first two subpopulations do not have diabetes. Stated another way, they are healthy. The prediabetes folks are not diabetics. If they were diabetics—I cringe stating something this obvious—then they would be in the diabetic group and not in the prediabetes group.

We know United Health’s purpose is to frighten—in modern parlance, “raise awareness”—because they make of point of saying, “The vast majority of people with prediabetes (more than 90 percent) and about a quarter of people with diabetes are unaware of their health condition.” The “health condition” of prediabetics is properly classed, as we have seen, as healthy. Yet they are recommending people be drug into doctors’ offices so that they can be informed of their “condition.” Baring hypochondriacs, most healthy people are aware of their health, so that this exercise would be pointless.

These kinds of reports invariably raise my blood pressure, which I just checked: 125 over 80. That puts me in the “prehypertensive” category, defined as those people who don’t have hypertension (high blood pressure) but someday might have it. I’d better get to a doctor right away!

Categories: Statistics

18 replies »

  1. Not sure what “normal” glucose levels are in statistical terms, but supposing it’s a median, wouldn’t 50% of all adults have a glucose level higher than normal? And conversly 50% lower? So basically, they’re saying that by 2050 they expect that 2% more of the total adult population will be above the current “normal”?

  2. Another way of looking at it:

    About half of the population will have higher than normal blood glucose! Also, some of those people will have diabetes!

  3. There surely must be a statistic for the percentage of people with heightened glucose levels who make it to type 2?

  4. Since I just went through this pre-diabetes fears. Here is what I have learned so far…

    The test is taken after 12 hours of fasting. ‘Normal’ blood sugar is less than 100 (I don’t know the units). Diabetic is over 125. My number came back at 104. Your glucose level bounces around throughout the day. Following a meal your blood sugar will increase. Normal glucose levels on a full stomach can be as high as 145.

    What does 104 mean? Absolutely nothing. It won’t hurt to eat less ‘white food’ and it wouldn’t hurt to exercise more. Test again in 6 months, it may return to normal range next time we test. And, it could stay above this ‘normal’ range and never move into the ‘diabetic range’.

  5. 1) A request: can you summarise the changes that have been made to the definition of Type II diabetes over, say, the last 20 years? I take it that the probability of its not having been changed is negligible.

    2) My GP said I had diabetes because my blood sugar had twice been above the threshold. I pointed out that a series of later measurements had been below. He replied (approximately) “You virtually don’t have diabetes, but by definition you do have diabetes”. I was flummoxed.

    3) We discussed the recent inspection of my heart arteries in the hospital. “They are normal” I said. “Good” said he. “But”, said I “do they mean ‘normal’ in the sense of desirable or ‘normal’ in the sense of average?” He was flummoxed.

  6. The press release seems to be an overwrought advertising pitch, since UHG uses the spectre of prediabetes to recommend

    (1) LIFESTYLE INTERVENTION: public health initiatives and the wider use of wellness programs (Oh, goody, some busybody from UHG is going to come “improve” my lifestyle);

    (2) EARLY INTERVENTION: community-based intervention programs…with lifestyle changes and modest weight reduction (because I need yet another busybody to tell me when my butt’s getting too big);

    (3) DIABETES CONTROL: more intervention is gonna get folks who are already taking meds to take their meds better, some how. This is the Gringo Theory of Communicating with Mexicans: shout louder, and end everything with an “o”; this requires a whole different class of busybodies;

    (4) LIFESTYLE INTERVENTION STRATEGIES: wider use of public-private partnerships (so folks like UHG can get on the Federal dole, probably to hire all the aforementioned busybodies.)

    So not only is the statistical argument specious, but the proposed remedies to the inflated problem seem to be serious rent-seeking. Bastards.

  7. Some may recall a previous post of mine where I said my doctor’s beliefs tended toward the fashionable. Well, this happens to be one area. I am diabetic (Type 2 so far). My father’s mother was diabetic so I was expecting it. It’s a genetic thing. My parents weren’t diabetic but myself, my brother and one of my sisters is but two are not. When I finally went for a real blood test my levels were close to 500 mg/dL. With medication, that comes down to around 100 which is within the normal range.

    The real problem comes form determining ‘normal’. Every year since my diagnosis I keep getting different normal ranges quoted to me and each quote has been narrower than the previous. The Wiki article on sugar diabetes says, “The mean normal blood glucose level in humans is about 4 mM (4 mmol/L or 72 mg/dL, i.e. milligrams/deciliter)” but interestingly the chart shown implies that 90 mg/dL is closer to ‘normal’.

    Just exactly how does one go about determining ‘normal’? if 50% of the population are above ‘normal’ doesn’t that imply they are normal?

    Cholesterol levels are another of those moving targets. Your body cholesterol level is set by your body internally and external factors, outside of medication, have little influence. Guess where those lower ‘normal’ levels are coming from.

  8. Mike Anderson,

    That brings up and interesting point. Supposedly one of the ’causes’ of type 2 diabetes is being overweight. But no one can seem to answer how that is so outside of the fact that the two are correlated. Before I started treating my diabetes i was losing weight at a prodigious rate. If my weight condition was the ’cause’, why wouldn’t it self-correct?

    When I was initially diagnosed, I was put on a restricted caloric intake. When I asked later why eating the same thing on two consecutive days had a huge difference in my sugar level he replied that was to be expected. When I asked why then the diet he replied “Because you need to lose weight.” This despite the fact that I was already losing weight!

    I have since learned that the medication I was given to control the diabetes actually causes weight gain. So go figure. My weight has been stable for two years now but I still get the “need to lose weight” lecture.

  9. I am pre-diebetic. There are certain benefits that go with this designation, one being you will not be able to buy life insurance, at least at a decent rate. Secondly, I get to go to the doctor’s office every 90 days to have my a1c checked. It is on the high end of “normal”. I also get a lecture by a nutritionist every visit to lose weight.

    I am not on any medication for diabetes, nor am I on any special diet. The nutritionist recommends a 2000 calorie per day diet, spreading those calories among five small meals.

    So, life is a bit of a risk, and I listen carefully to my medical advisors every visit. I make it a point to have a cheeseburger after every visit. too. It is either poetic or stupid, and I don’t care which.

  10. I guess, considering all the pre-something-or-others, at 61 years of age I could be considered pre-dead. I certainly show symptoms of actually being dead from time-to-time; failure to respond when spoken to (although usually only when spoken to by my wife), catatonic state in front of the TV, certain bodily functions temporarily bound in stasis, and so on.

    Since I am not actually dead, only pre-dead, I suspect I should start an awareness campaign to bring attention to this newest of pre-conditions. Perhaps I can write a peer reviewed paper and get one of those big, fat government grants that will allow me to spread the word to the ‘cohort’ most at risk of being pre-dead.

    Or, perhaps I am just being an ‘armscythe’ in my old age… 😉

  11. Sounds like a darned good way to sell the ‘pre-diabetes’ test we just invented, as well as whole shed-loads of placebos uh scratch that, ‘highly effective chronic medication.’

  12. Seriously, the “pre-diabetes” term was probably developed to enable health insurance coverage under many, if not most, health-care plans. It ‘ticks’ another box that allows your health provider to be slightly better paid for your visit. Losing weight in certain ways does lower one’s blood sugar readings, but losing in other ways doesn’t. I don’t know the difference or the why’s.

    Almost choked with laughter at the professor’s archaic word usage [emphasis mine]:

    “Yet they are recommending people be drug into doctors’ offices…”

    But then I considered it was likely a Freudian slippage type of thing and laughingly dragged myself away from the dictionary.

  13. I was reading James Taranto’s column today. He highlighted that the last Wednesday of September is National Previvor Day.

  14. It’s worse then you state it. In the last few years the government and health care professionals have embarked on an effort to identify people who have diabetes but do not know it. There is a large minority of millions of people who have diabetes but do not have obvious symptoms. This effort was a huge success and the identified about 2 million people with diabetes who had not previously known this. Then various groups whose area of special interst is diabetes used this data to “prove” that we are in a diabetes epidemic. Or to state what you referred to; that by 2050 more then half the population will have diabetes. This is phony! It is all based on data from a known source that does NOT indicate a huge increase in diabetes. The special interest groups may want to “scare” you into getting tested but make no mistake what they want more is to increase funding. They are looking at a pile of money and cannot resist a little good natured fraud to get some of it.

  15. Half of the epidemiologists at the United Health Group are sub-par, below-average epidemiologists. It’s a real problem, but they don’t know how to solve it. Solutions to alleged problems, even simple ones, escape them half the time (at least).

    The preferred solution to dietary excesses by street proles is to force us to eat what the “experts” say we should. “Ve have vays of making you eat vat ve tell you to,” etc. Experts are the new bourgeois, and we know the track record of that group.

    Of course, half the dietary recommendations will be below average, because they will be written by sub-par dietitians and nutritionists. Fortunately, the new regulations will be enforced by sub-par gummit flunkies, at least half of whom are dumber than rocks. The net effect will be a “moron circus” where well-meaning bleeding-hearts smack their own heads into brick walls. Painful to watch, but there is no cure.

  16. Do a search of the food pyramid and capture the various versions published by the government. The latest one has given in to political correctness and is heavy on all the approved food groups and light to non-existant on those things containing sugar or fat. Were any of them correct? Is the current one correct? Will the next one be correct? We are all guinea pigs for whatever the newest food fad is. Does anyone believe the new and very expensive school lunch bill will “cure” obesity? Of course it won’t so that will require even more money and a new fad diet for our kids. I have often wondered since some 30% of the kids are obese or overweight doesn’t that mean 70% are normal or skinny? So why are we punishing the normal and skinny kids? Shouldn’t the skinny kids be eating fast food so they can get healthy? Shouldn’t we watch what the normal weight kids eat so we have emperical evidence of what the diet should be? The obvious answer is the skinny kids and normal weight kids eat the same thing as everyone else does. How could that be? If fast food makes you obese and skinny teenagers are eating fast food wouldn’t that make you question your conclusions? Has anyone considered that obesity is genetic and you cannot “cure” it by eating “healthy” (whatever that is). What are the odds that a whole country’s diet is so bad that we have an “obesity epidemic” but at the same time life expectancies increase every year.

  17. GoneWithTheWind wrote: “Has anyone considered that obesity is genetic and you cannot “cure” it by eating “healthy” (whatever that is).”

    I don’t disagree with the statistical perils outlined above, and I don’t necessarily take issue with the idea that there is such a thing as a genetic predisposition, perhaps even “cause” for some individiuals. However, as someone who currently works with individuals to help them lose weight and become more fit and active, I can tell you categorically that the great majority of overweight individuals are there as a result of lifestyle (and this included me in the past), not because of an uncontrollable genetic cause.

Leave a Reply

Your email address will not be published. Required fields are marked *