Friday, December 30, 2011


A tiny amount of rain fell today, but that wasn't the reason I didn't exercise. I hadn't had a rest day since last Friday, and I felt I was due. And the weather is forecast to be good tomorrow, so I'm planning to have either a long run tomorrow or a long bike-ride with a friend, if I can get the schedule worked out. Anyway, today was my day of rest. And well-deserved, I think!


I found out something interesting today about insulin: Banting and Macleod, the guys who got the Nobel prize in 1923 for supposedly having discovered insulin, actually weren't the ones who discoverd it. A Romanian physiologist named Nicolaeu Paulescu had discovered it before them, and had published papers on the subject (and had even been granted a patent on his process for isolating insulin). Banting and Macleod merely confirmed work that had already been done by Paulescu. 

So why didn't Paulescu get the Nobel prize? Apparently because, regardless of his merits as a scientist, he was a scumbag of a man, and nobody felt like honoring him. Protests were launched against the "brutal inhumanity" of his anti-semitic views and other prejudices. As late as 2003, an attempt to honor him with a statue at a state hospital in Paris was canceled due to protests by Jewish groups. (He had it in for the masons, too, but I don't know if they were involved in the 2003 protests.) Anyway, there's a lesson here for ambitious scientists: stick to your lab bench, and leave the social bigotry to others, if you want to get the Nobel prize. (Nobel-prize-winning bigots have generally been prudent enough, as in the case of William Shockley, to wait until they already had the prize before letting the crazy out in public.)

So what is this hormone called insulin? It's a complicated protein with a structure that's shaped like this:

Although we usually think of insulin as a hormone with a single purpose (that of bringing down blood sugar levels, by stimulating muscle cells and some other cells to absorb glucose out of the bloodstream), insulin actually has many purposes, and many different effects on the human body. An article I read on the subject lists no fewer than 15 separate effects of insulin, including reduced sodium secretion by the kidneys. No wonder diabetes is such a complicated disease: diabetes revolves around insulin, and insulin does at least 15 different things.

The only reason Paulescu was even looking for insulin in the first place was because people were trying figure out what the Islets of Langerhans were for. The "islets" were little patches of tissue in the pancreas, discovered in 1869 by a medical student named Paul Langerhans. He had no idea what purpose the islets served (although his son Archibald later played a role in figuring it out). For a long time it was assumed that the pancreas secreted some substance that played a role in digestion.

In 1889, Oscar Minkowski and Joseph von Mering tried to find a clue to the mystery by removing the pancreas from a dog, to see what happened to its digestion. They couldn't see a difference in the dog's digestion, but they did notice that the dog's urine was suddenly attracting insects. Voila: the dog had become diabetic after losing its pancreas. Scientists then realized that the pancreas played a role in regulating blood sugar rather than digestion. They even found that liquefying pancreatic tissue and injecting it into diabetic animals brought their blood sugar down. Then, the hunt was on to isolate the substance in the pancreatic tissue that regulated blood sugar. That was where the Romanian anti-semite entered the story, and discoverd inulin --or pancrein, as he originally called it. The name insulin refers to the "islets" (insula in Latin) of Langerhans, whence the substance originates. Insulin was probably considered a better name simply because it hadn't been coined by the Romanian anti-semite. A further reminder that scientists with Nobel dreams should knock off the social commentary, particularly if it's of a hateful nature.


And I'll give the last word (for now) to Zach Weiner:


Thursday, December 29, 2011


Fasting glucose down to 80! Good recovery from Christmas.

Post-prandial glucose down to 91! (But that was such a low-carb dinner that it wasn't much of a challenge.)


The skies were cloudy today, which is about as much winter hardship as we've had to suffer this week.

I walked up to Lake Ilsanjo, but it wasn't such a hard climb that I was sure it qualified as a workout, so when I went to the gym later to do weights I figured I'd better throw in half an hour of sweating on the elliptical trainer just to make sure.


How I wish I didn't have to share Annadel State Park with the equestrians! Even when you don't run into any of them, there's no way you can forget that they're around.

Yuck!


And now let us discuss some people who make the equestrians seem considerate by comparison...

Most of the time, I try to keep a detached, ironical, faintly-amused attitude toward controversies over what is and isn't good for human health. For example, on the question of whether a vegetarian diet or a carnivorous one is better for us, or whether saturated fat does or doesn't promote atherosclerosis, I can't help noticing a contrast between the passionate certainty of people's opinions and the frustrating inadequacy of the evidence available to us. I don't know who will turn out to be right, when and if adequate evidence ever comes our way; in the meantime I will simply experiment as best I can, to figure out what seems to work best for me in the short term (in the hope that it will also be what works best for me in the long term).

Sometimes, however, a big fake "controversy" gets cooked up, over a matter which has already been settled -- because the evidence we have is very good, and it points toward a single conclusion. I find it hard to take a detached, ironical, faintly-amused attitude in cases like that. Especially if it is not a matter of people deciding what kind of diet they themselves want to eat, but rather a matter of people making decisions that affect the health of others in their community.

During my last visit to the doctor, I was asked if I wanted to be vaccinated for a combination of diseases. I don't remember what all the diseases were, but one of them was pertussis (sometimes called whooping cough -- a name which makes the disease sound like a bit of comical romp, when it's actually a nightmare). I hadn't liked what I'd heard about the disease, which is making a comeback and alarming public-health officials. I'm sure I'd been vaccinated for pertussis as a child, but the effect of a vaccination fades slowly over time, and if you haven't had a "booster" immunization within the past 10 years you're probably not resistant to the disease any more. Anyway, I of course said yes to the vaccinations.

But I realize that a lot of people would say no. There is a weird anti-vaccination movement going on in the U.S., and also in some other countries which one would have expected to be less susceptible to hysteria of this type. The crazy people ("crazy" is probably putting it nicely) who are leading this crusade claim that vaccination causes all sorts of problems that it has been shown not to cause (autism most famously), and they are having great success in scaring parents into replacing the miniscule (when not downright imaginary) risks of vaccination with the real and terrifying risk of bringing back epidemic diseases which used to devastate communitities in the days before vaccination was widespread.

The risk posed by the anti-vaccination movement is far from theoretical. We can put a human face on it.

Meet Dana Elizabeth McCaffrey, who was born in Australia in 2009...

...and died the following month, from pertussis. Although Dana, as a newborn, was too young to be vaccinated, the rest of the family already had been, so the parents assumed she was not in danger. What they didn't realize was that they lived in an area with an extremely low vaccination rate, thanks in part to the efforts of a powerful anti-vaccination movement in Australia, so herd immunity there was low, and many local people were carriers of the disease without knowing it.

It turns out that pertussis is a disease which is easily prevented but impossible to treat. Antibiotics make the patient less infectious to others, but don't make the patient's own condition any better. Doctors do not try to "cure" pertussis; they try to keep the patient alive long enough for the disease to run its course. Unfortunately, that course is a very long one (plan on coughing violently for six weeks at least), and only the lucky ones make it through the ordeal alive. Dana was not one of the lucky ones. Her parents have now become activists, trying to fight the anti-vaccination forces, so that other parents don't have to watch their children die.

We are all free to follow our own instincts in choosing sides in a controversy -- provided that it is a genuine controversy, which continues because the available data continues to be insufficient or contradictory. But when it comes to a bogus controversy, based on nothing, which continues simply because some people are emotionally or financially invested in it -- and that bogus controversy is killing people -- I think society has a right to push back, and push back hard.


Wednesday, December 28, 2011


Today, we had some more of our local version of winter: T-shirt weather.

Obviously I had to take the opportunity to go for a hike. I took a light jacket with me, in case the ocean breezes were too cold for me when I got over to the coastal side of the ridge. But there wasn't a breath of wind stirring.

Things could get ugly on Friday, though: a 30% chance of rain! If I don't post anything on Friday it will probably because I'm overwhelmed by having to cope with a little drizzle or something.


One of my goals for this site is to encourage people to take a more skeptical view of articles and studies on health. I don't always make that point explicitly -- often I merely state my doubts or questions about a particular research report or a health recommendation, without going into the reasons why I think uncritical acceptance of any claim is not a good idea. If the claim is worth anything, it will stand up to skeptical scrutiny, so we might as well expose it to some. (Skepticism doesn't mean simply assuming everyone's a liar, by the way -- it means being careful to consider the reasons why a particular report might be wrong.)

I'm going to give an example, but it doesn't have a lot to do with health. 

When I read that Cheeta the Chimpanzee -- one of the more convincing actors to appear in the original Tarzan movies -- had died on Christmas Eve, I immediately wondered if there wasn't something just a little fishy about the story. How long did chimpanzees live, anyway? That series of films (the ones with Johnny Weissmuller) began in 1932, for crying out loud, and the same chimp was supposedly used in them from the start, so Cheeta would have to be a real survivor to have been around as late as last Saturday.

Cheeta is pictured on the left below. (The far left, smartass!)

The original press acounts said that Cheeta (the most common spelling of the name in movie credits) died of kidney failure, at the Suncoast Primate Sanctuary in Palm Harbor, Florida. Debbie Cobb, who works there, was the proud owner of the creature. She claimed that Cheeta was 80 years old at the time of his demise. An early press report acknowledged that this advanced age made Cheeta "a rarity". But I checked, and found that animal experts are now saying it would make Cheeta not merely rare, but unprecedented. Chimps have a typical lifespan of 30 years in the wild, 50 years in captivity. No previous instance of a chimp surviving to 80 is known; the record-holder hasn't made it any farther than the low 70s.

Conveniently, 80 is about as young an age as Debbie Cobb could assign to her chimp, and still have her identification of this Cheeta as the Cheeta be chronologically possible. But is it biologically possible? That seems more to the point, if you ask me.

The above photo of Debbie Cobb's chimp certainly looks as if he's not in his first youth, but whether he looks 80 or not is hard for me to judge; chimp faces are not the sort of images over which my eye likes to linger, and I haven't studied the creatures in any depth. I just don't know my chimps.

Still, I have to assign a pretty low credibility to any animal story which hinges on the animal's having lived seven or eight years longer than any other member of that species has ever lived before. Perhaps Debbie Cobb believes her own story (she received Cheeta from her parents, and seems to have accepted their story of the animal's history without question).

By the way, this is not the first chimp to have been presented to the public as the original Cheeta. Perhaps, if you own a chimp (a dangerous sort of pet, by the way), you feel a need to justify keeping the thing -- and what would make a better justification than claiming that your chimp is not just a chimp, but a celebrity chimp? Surely that would make you feel more important than the average ape-fancier.

If someone is telling you a story which is hard to verify, and is implausible on a key point, and if they stand to benefit in some way from having the story accepted as true -- well, don't all of those things qualify as warning signs?

These cautions apply, I think, not just in the case of proud pet owners who want to claim an extraordinary history for their animals. It also applies in the case of people with a drug or a diet book to sell you.

A couple of more points about Cheeta...


Tuesday, December 27, 2011


Fasting glucose settling down to the 80s: good. Even my post-prandial test was in the 80s: excellent. That test followed a low-carb dinner (a Cobb salad, without bread or any other starchy extras), so a low result wasn't too surprising, but it's always to get that kind of non-surprise.


My readership is down a little this week, so far. Could it be that diabetes is something people don't want to read about on Christmas, or right after Christmas, and that it takes a while for people to get themselves ready to engage with that scary subject once more? Perhaps nobody will really want to hear about diabetes until 2012 begins on Sunday.

If so, I'll try to keep it light until then.


The weather in northern California still hasn't turned against us, so this afternoon I drove out to the coast and went for a walk along the cliffs near Goat Rock State Beach (south of Jenner and north of Bodega Bay). I hesitate to call it a "hike" because it wasn't long enough; I knew that I couldn't call it my workout for the day, and that I would have to go to the gym in the evening.

It's a rugged, "emergent" coastline, meaning that the movement of the tectonic plates is lifting the earth up here, not settling it down (hence the cliffs). There are monster rocks scattered around here and there.

Some of the larger rocks had climbers on them.

But most people just wanted to hike up to the hilltops for a better view.

Actually, those young guys were just showing off by choosing to go up the south side of the hill. The climb was a lot easier if you went up the north side.

In fact, it was easier to get up there than it looks.

And you sure could see a long way from the top.


Monday, December 26, 2011


Ah, the day after Christmas! In the U.K., it's known as Boxing Day, for reasons which I continue to regard as mysterious. The explanations I've heard seem pretty unlikely (not that that there's anything unusual about a British institution having a puzzling name and an unlikely explanation, but still!).

However, December 26 should probably be known, to diabetes patients everywhere, as Damage Assessment Day. This is the day when we try to analyze what all those parties and treats and big dinners did to us. Not too much in my case, apparently. A fasting test of 95 mg/dl isn't quite stellar, but for the day after Christmas it's not bad. And 100 after dinner is excellent.

Of course, my dinner followed a very long run. Wanting to compensate for the Christmas indulgences, and also wanting to take advantage of the good weather, I decided to do one of my longer standard routes through the state park.

We never have anything like a White Christmas in my part of California; it can actually be one of our greener times of year.

Fortunately, running felt good to me today. The weather was cool enough for long sleeves and long running pants to be appropriate, but having taken those precautions I was comfortable, and I felt energetic. I didn't have a hard time doing the run, and when I came to forks in the trail, I didn't have a hard time persuading myself to choose the direction that would add the most mileage to my route. (It ended up being 9.3 miles.)


The reason I had time for such a long run today is that my company shuts down during the week between Christmas and New Year's Day. Friday was pretty much a big Christmas party at work. There was a lot of music. Ours is primarily an engineering company, and both engineers and scientists are very often musicians as well -- perhaps because music is a kind of intuitive math (rhythm, melody, tonality, harmony, and dynamics are all entirely about ratios).

The brass band that we decided not to join forces with played separately from us in the office; our band played later. It went well. And it turned out that we didn't end up playing the "Alvin and the Chipmunks" Christmas song (the one about "I still want a HULA HOOP!") -- a threat which had been hanging heavy over me. It's hard to say where you cross the line, at Christmastime, between mild embarrassment and outright degradation, but for me the Chipmunks are the border-guards on that particular frontier. (And when their shift is over, Frosty and Rudolph take over for them.)


As I'd had a lot to eat at a family gathering on Friday and Saturday (and not a lot of exercise), I thought I'd better make sure I got out for a good run on Christmas morning. You might think I'd be alone out there, but lots of people were hiking, jogging, and cycling around the lake. Little kids checking out their new bikes. Adults trying to compensate for the pie they knew they'd be eating later in the day.

The weather was colder than it was today, but the winter sun breaking through the haze was beautiful.

These people were walking a dog that desperately wanted to get at the Snowy Egret in the grass. Even without a leash on I don't think he would have had a chance.

He probably would have liked to get at the cormorants in the treetops as well, but he had even less of a shot there.

Later in the day, a musician friend hosted a musical party at a rented house on the coast near Bodega Bay.

That's my idea of a good way to spend a winter afternoon: playing music in a warm and cozy beach house, with a view of the water.

Anyway, I seem to have got through the holiday season largely unscathed. But, like a lot of other people, I added some pounds and need to get rid of them now.


Friday, December 23, 2011

I'm taking a break until Monday. Have fun; just not too much!


Thursday, December 22, 2011


I was reading about a study of high blood pressure, which tracked the effect of blood pressure on long-term health, and concluded that people who get their blood pressure under control in middle age significantly reduce their risk of cardiovascular disease later in life. 

This line from the report intrigued me: "The patients who improved their BP may have changed their diet and exercise habits, or the difference may have just been regression to the mean." 

How's that again? The improvement in their blood pressure may have just been regression to the mean

"Regression to the mean" belongs to that category of concepts that are (1) important, (2) not widely understood, and (3) hard to explain. In other words, it's the kind of thing I like to write about.

The "mean" is, of course, the average of a set of values. When you take a series of measurements within a population, examining some characteristic that varies from one individual case to another (blood pressure in adults, for example), the measured values tend to bunch up fairly close to the mean. There are many values near the mean, right at the center of the range of all values, but there are few values far below the mean, and few values far above it.

For example, the average height for American adult males is 5'10". That is the mean. There are lots of men who differ from the mean by one inch (in either direction), but fewer who differ from it by four inches, and fewer still who differ from it by eight inches. The farther you get from the mean, the smaller the number of values in that range.  This is the famous "bell-curve" distribution.


When you look through a set of individual values one at a time, you will find that any extreme value (that is, a value far below the mean, or far above it, as in the yellow regions of the curve above) is usually going to be followed by one that isn't extreme at all (in the blue region). This natural tendency of extreme values to be followed by non-extreme ones is called "regression to the mean".  An extreme value is an unusual departure from the range where most of the values are clustered; regression to the mean is nothing more than the all-but-inevitable settling back into the more typical range of values. Because most values are not very far from the mean, any value that is far from the mean is probably going to be followed by one that isn't.

For example, if the first man you see on the street is 6'3" (that is, five inches taller than the mean), the next man you see after him is probably going to be closer to 5'10" than that -- simply because most men are closer to 5'10" than that. The idea here is not that the height of the first man somehow influences the height of the second. It's just that being close to the mean is likely, and being far from the mean is unlikely, so the odds are against your seeing two exceptionally tall men in a row.

So far, this all sounds reasonable and obvious -- and yet it conflicts with certain assumptions which we tend to make unconsciously. For example, many people tend to assume that an extreme value is likely to be followed not by a value closer to the mean, but by a value that is extreme in the opposite direction, just to "balance things out".

If the first man you see in the street is exceptionally tall, doesn't that mean the next man you see will probably be exceptionally short? (No; the next man you see will probably be closer to average height -- simply because most men are.)

If you are tossing a coin and you get "heads" several times in a row, doesn't that mean you are due for a long streak of "tails"? (No; that could happen, but it's not made any likelier by the recent streak of "heads"; the next several tosses will probably be a mixture of heads and tails, regardless of what has gone on before, simply because most sequences of coin-tosses are mixtures of heads and tails.)

This expectation that one extreme result will somehow trigger a result that's extreme in the other direction, so that things come out even in the end, is known as  the gambler's fallacy. It's a false expectation. The reality is that, in any circumstances, values close to the mean are likelier to to occur than values far from the mean. It never makes sense to bet on an unlikely event because you feel the unlikely event is "due". (If you insist on betting, bet on a likely outcome, not on an unlikely one that would feel appropriate to you if it happened.)

Of course, people also embrace the opposite fallacy: they see an extreme result, and they assume it establishes a new pattern which can now be expected to continue. A number is "lucky" because we bet on it before, and won. An athlete is invincible because he just had an unusually good season. When we invest heavily in such expectations, we are usually headed for a disappointment.

The trouble with regression to the mean is that we tend to forget it exists. Therefore, we don't plan for it. And when it happens, we don't even recognize it for what it is. We misinterpret it as something else.

For example, consider teachers who do one-on-one instruction (music teachers, flight instructors, and so on) -- a famously cruel breed. If you ask them to be more encouraging to their students and less angry, they may object that kindness is counterproductive, and they can prove it from their own experience. When a student performs unusually well and they praise him, he doesn't do as well next time. If a student performs unusually poorly and they jump all over him, he does better the next time. Conclusion: being nice yields bad results, and being nasty yields good results!

Of course, both of the teacher's examples are really just examples of regression to the mean. Students have good days and bad days, like anyone else;  their best-ever performance is an extreme result, and so is their worst-ever peformance. Because of regression to the mean, any extreme result is probably going to be followed by a more typical result. Therefore, we should expect a student to perform better after his worst day, and to perform worse after his best day -- regardless of how the teacher reacted.

Our misinterpretation of regression to the mean manifests itself in all sorts of odd ways, including the widespread belief that there is some kind of strange "curse" that afflicts people who win this or that achievement award, or appear on the cover of this or that magazine, or achieve some kind of extraordinary success. Obviously, achieving extraordinary success is an extreme outcome, not a typical one; we shouldn't expect such events to repeat themselves, and we shouldn't find it puzzling when they don't. The fact that an actress won an Academy Award probably means that she was finally lucky enough to be cast in a well-made movie, playing a well-written role that happened to be a good match to her strengths as a performer. That she never got that lucky again doesn't prove anything except that regression to the mean applies even to movie stars; it does not prove that winning an Oscar is bad for your career. But many people think there really is an Oscar Curse.

Okay, traveling back to my point of departure, what does it signify to say "The patients who improved their BP may have changed their diet and exercise habits, or the difference may have just been regression to the mean"?

The idea here is that anyone's blood pressure varies from one occasion to another. If you get a bad blood pressure measurement in the doctor's office, but it was an exceptionally bad day for you in regard to blood pressure, the principle of regression to the mean says that later blood pressure readings will be better than that one was. Someone comparing your initial result with later results will get the impression that you used to have a blood pressure problem but you fixed it somehow. The impression may well be false; maybe your blood pressure is no better and no worse than it used to be, but one day your doctor happened to measure it when it was atypically high, so all subsequent measurements seem to show an improvement.

It is very difficult to eliminate these problems of data interpretation that can be caused by regression to the mean. That is one of the reason that medical research papers have to be evaluated so carefully, and challenged so strongly. The cause of a problem, or the cause of an improvement in a problem, may seem obvious from what the researchers say about it. But we have to make sure that the problem, or the improvement in it, is real to begin with -- not a statistical artifact that we created by treating extreme values as if they weren't extreme.


Wednesday, December 21, 2011


Great -- a day with no embarrassingly unsuccessful Christmas-music rehearsals! So, I was free at lunchtime to get outside for a run. Really beautiful weather for it, once again. Now people are starting to worry that this very sunny and dry December may indicate we're heading into a drought year. Well, I'll worry about that after Christmas.

Of course, tomorrow there will be one last Christmas-music rehearsal, before our performance on Friday. So, no running in the daytime. It will be an evening at the gym for me. I'll try to reinforce the weight-training habit, which I have such a hard time establishing.


Today is the winter solstice -- the shortest daylight interval of the year. In my town, at least, we have slightly less than nine and a half hours of daylight today. People living farther north from the equator get an even shorter day than I'm having; in Scotland today, the sunlight lasted less than 7 hours. (In the southern hemisphere, of course, the situation is reversed: this is the longest day of the year in Australia, and residents of Sydney will see more than 14 hours of daylight.)  

It is often claimed that winter "officially" begins today. This odd choice of words suggests that the seasons are subject to government regulation, and had better fall in line and behave themselves. However, there is actually no sensible basis for saying that winter begins on the solstice. If "winter" means anything, it means the coldest quarter of the year. Whichever 13 weeks happen to be the coldest where you live, that stretch of time is what we call winter. Around here, and probably in most places, the winter season is usually in progress three or four weeks before the solstice occurs.

There are some interesting peculiarities involved in the astronomical cycles which make the seasons. One is that temperature changes lag behind changes in the amount of daylight. Given that today is the shortest day of the year, you might think today would represent the very depth of winter, and temperatures would rise from here on. So how come January is usually colder than December was?

Well, keep in mind that it takes a while before the days lengthen very much. Tomorrow will be only one second longer. Admittedly, the pace picks up after that, but even 30 days from now, there will only be 27 minutes more of daylight than we're experiencing today. And the cooling of the earth which occurs while the days are short is a process which tends to pick up momentum as it goes along; the trend is not easily reversed. For example, any snow that fell early in the winter has the effect of intensifying and prolonging the cold, in part because sunlight is reflected off the snow back into space, instead of being absorbed and warming the ground. It takes a few months for the days to lengthen enough to reverse this cooling trend, so January and early February are bound to be cold even though there is more sunlight time.

Another peculiarity worth noting is that, although December 21 is the shortest day of the year, it does not have the earliest sunset of the year. The sun set at 4:53 PM today in my town -- but it set 3 minutes earlier than that on December 1st. The reason today is 12 minutes shorter than December 1st, even with the sun setting 3 minutes later than it did then, is that the sun rose 15 minutes later today than it did on December 1st. (Changes in sunrise times and sunset times are asymmetrical; they do not proceed in lockstep.)

These changes in the length of the day are bound to affect us physiologically; perhaps the shortening of daylight acts as an appetite trigger, and this (rather than the holiday parties and Christmas cookies) accounts for the common tendency to gain weight at this time of year. At least, I've heard that it's a common tendency, and I'm doing my part.


I noted this morning, while reviewing Google search terms that resulted in referrals to this site, that one of them was "the best way to increase body hair for men".

It surprised me greatly that anyone seeking help in that area should be referred by the folks at Google to my site, as I didn't think I had emphasized that point in any of my previous blog posts. And I don't remember posting any self-portraits in this style:

Apparently my reputation for hairiness extends to wider realms than I had ever suspected, and a lot of men who are frustrated with their inability to get fuzzy are lining up to learn my secret. Well, I can give it to him in one word: almonds.

What are you looking at me like that for? Stupider health advice than that is offered to people every day! And people usually charge for it; at least this stupid advice is free.


Tuesday, December 20, 2011


I didn't get to run at lunch today (which is too bad, because it was beautiful weather for running), because of that extra Christmas-music rehearsal with the other band at work. I should have gone running instead; the rehearsal was a train-wreck.

The brass band plays "transposing" instruments and our band doesn't. I won't stay up to midnight trying to explain what that means, but one of the consequences of it is that we couldn't read their sheet music and they couldn't read ours. They seemed to play everything in a different key than we did, and it was futile to discuss that point, because they even meant something different by "we're in C" than we did. Our attempts at playing by ear did not smooth out the misunderstandings very well, either.

Embarassing! I guess the two bands will have to play separately on Friday. It's too bad, because they're good musicians, and the idea of having a big loud band kind of appealed to me. I have a soft spot for trombones. It's one of the great sounds, I think.

That "music is a universal language" trope can be pretty misleading. Music itself, purely as sound, may be a universal language, but the language in which music is described and written is another matter entirely. When "we're in C" means one thing to the trombonist and another to the flutist, things can really get ugly.


So I went to the gym after work to get my exercise in. Did some weight-training as part of it. I know weight-training is good for me -- good for glycemic control, and probably good for blood pressure too. But I sure have a hard time making myself do it. Still, it was a little easier tonight than it usually is; I hope it means I'm getting more comfortable with the idea.

The weight-lifting might possibly have contributed to the low post-prandial glucose result. But it was a pretty low-carb dinner, and that alone might have been enough.


"Increasing Heart Rate Over Time Linked With Heart Disease Mortality", says a long-term Norwegian study which tracked 13,499 men and 15,826 women over a dozen years. The people in the study began that period without cardiovascular disease. The idea was to see which of them developed cardiovascular disease later, and determine how those individuals differed from the individuals who didn't. It turns out that those whose resting heart rates went up over that time were at increased risk of heart disease mortality -- and the bigger the increase in heart rate, the greater the increase in risk.

"Compared with subjects who had a resting heart rate <70 beats per minute (bpm) at both measurements, those who had an increase in resting heart rate at baseline from <70 bpm to >85 bpm during the follow-up visit were nearly two times more likely to die from ischemic heart disease."

"The lack of an increase in resting heart rate could be a marker of good underlying health, including engagement in physical activity and a lack of smoking prevalence, although the researchers attempted to control for these factors in the multivariable analysis."

It sounds silly to express it this way, but the bottom line seems to be that we only get so many heartbeats per lifetime, and the faster your heart rate is, the sooner you use up your supply.

My resting heart rate is always below 70, and typically it's in the 50s, so I guess I'm keeping my risk down. It wasn't always that way -- my resting heart rate was certainly above 70 until I started my exercise program. So, there you have it: another tangible exercise benefit.


More medical news (from The Onion ):


Monday, December 19, 2011


The most dangerous time of the year has truly arrived! One of my coworkers (and running buddies) took Thursday and Friday off last week for her annual Christmas project of making top-quality chocolate truffles -- in massive quantities. Here is the sample tray she brought into the office this morning.

Oh my goodness. I can't talk about it right now.


At least we had a good run today. The weather has been ridiculously kind to us runners, so far this winter. It was sunny, clear, beautiful, and not cold. Here's the local weather forecast: 

You don't have a lot of excuses not to get outdoors and exercise when you've got weather like this. Even so, I know I'm not going to be able to do it tomorrow, because of a Christmas music rehearsal. It's going to be a night at the gym for me.

Our existing band at work (two fiddles, guitar, flute, horn, trumpet, string bass) is apparently going to join forces with another band (two trombones, trumpet, saxophone) that I didn't even realize was planning to play on Friday. We'll rehearse tomorrow at lunchtime to see if we can all play the same tunes and in the same keys. On Friday morning (the last work day before the Christmas break), we'll stroll around from building to building, playing a few Christmas songs in each location. We've never done it with this large an ensemble before. Could be really cool. Could be a total train-wreck. I guess I'll know more tomorrow!


Stunning news: "Testosterone Replacement Improves Male Metabolic Syndrome"... says a study funded by the people who want to sell you the testosterone. Apparently testosterone cures everything. It even makes the equity in your house go up! You'd be a fool not to take it!


I'm intrigued by the strange long-term consequences of seemingly trivial incidents.

In the early 1920s, a student at Dartmouth named Theodore Geisel was caught drinking alcohol in his room with his friends -- which, apparently, was a pretty serious scandal in those days. As a disciplinary measure, the dean of the college prohibited Geisel from continuing to participate in any of his extra-curricular activities -- including writing for the college humor magazine.

Unwilling to give up his work for the magazine, Geisel decided to write under a pen name, and chose one based on his German middle name. Thus was born "Dr. Seuss". (Seuss is actually pronounced "zoyce" rather than "suice", but the author resigned himself to the inevitable American rendering of it.) He did get a doctorate, by the way (in English Literature), so there's nothing shady about his use of the title.

If being translated into many languages is an indicator that an author has achieved great popularity, perhaps the indicator of extreme popularity is being translated even into languages that are dead (and deserve to be). Thus "Quomodo Invidiosulus Nomine Grinchus Christi Natalem Abrogaverit", a Latin version of "How the Grinch Stole Christmas", as translated by Jennifer Morrish Tunberg and Terence O. Tunberg. I guess it could be translated in other ways, too.

Exactly why the Tunbergs thought it might be a worthwhile project to translate Geisel's 1957 satire on holiday consumerism into Latin is hard to say -- but then, I felt the same way about a recording of Elvis Presley songs translated into Latin. If some people out there want to listen to a Finnish professor of literature known as Dr. Ammondt sing "Non Adamare Non Possum" (Can't Help Falling In Love), "Nunc Hic Aut Numquam" (It's Now Or Never), and "Nunc Distrahor" (All Shook Up), I guess it's their business.

Other fun facts about Dr. Theodore Seuss Geisel:

A not-so-fun fact about Geisel is that, while his wife was ill with cancer, he had an affair to with another woman, and this upset his wife so much that it was apparently a contributing factor to her suicide in 1967. Nowadays, of course, you can do that sort of thing and still run for president -- but we expect writers to adhere to a higher standard than that, do we not?


Friday, December 16, 2011


I knew it! As yesterday's schedule included no exercise, but did include a big office Christmas party in the afternoon (at which I of course had too much to eat), I thought it was pretty likely that I would get an elevated fasting test this morning, as I mentioned in yesterday's blog. Well, my fasting test was up today, all right. An additional factor probably played into that: because of a local power failure, my alarm did not wake me this morning. I overslept, and then woke up in a panic. The burst of adrenaline that I get when that happens seldom fails to boost my blood sugar at least a little bit.

Well, as I also said in yesterday's blog, I know what to do when my test results become elevated. I had a good hard run at lunchtime, and watched the carbs. My result after dinner was lower than my fasting test had been.

The Christmas-party season certainly doesn't make glycemic management any easier, but I'm not sure it is a useful strategy, in the long run, to try to practice such self-denial at Christmas parties that you end up feeling as if you didn't participate in the event. My doctor warned me about that: diabetes patients who learn to associate glycemic management with feeling miserably deprived are probably the likeliest diabetes patients to "burn out" and give up.

My feeling is that -- provided you are in good control most of the time -- you can afford to, and probably need to, let yourself get a little reckless with the Christmas treats once in a while. There are limits to this, though. You can't treat every day between late November and Christmas as an occasion for discarding all rules, because there are just too many opportunities offered to us to do that. Too many parties, too many holiday brunches, and too many plates of those Christmas cookies I wrote about so movingly last year. We have to be stoic and tough on most of the occasions when a coworker shows up with a plate of fudge.

But there have to be a few holiday gatherings per year at which we're not so careful. Even at those, however, we shouldn't just chomp away at everything within reach. I guess what I'm recommending is anarchy, but contained anarchy that happens only once in a  while.


241 years old today:

Ludwig van Beethoven, pianist and composer, born December 16, 1770 in Bonn, Germany.

His various health problems throughout most of his adult life (including, of all ironic things, gradually worsening deafness) have led biographers to speculate endlessly about what the underlying cause of all these problems was. Although, these days, we tend to think "diabetes" when someone's health is deteriorating in multiple unrelated ways, that doesn't seem to have been the case with Beethoven. One of the reasons the situation is mysterious is that Beethoven's doctor, when he fell ill himself and mistakenly thought he was dying, ordered an assistant to burn Beethoven's medical records, lest they fall into the wrong hands. The doctor recovered, but never explained why he had thought it so urgent for the composer's medical records to be destroyed. A veneral disease is the obvious explanation (perhaps the only one that makes any sense), but few biographers can bear to come out and say it, committed as they are to the notion that the artists who could write this music could not have had human needs.

Well, we're flawed creatures, we humans. We even eat Christmas cookies sometimes.


Thursday, December 15, 2011


Rest day -- and office Christmas-party day! A dangerous combination, and perhaps my test results tomorrow won't be stellar. Well, if they're not good, I know what I need to do to improve them.

But I'm afraid I have no time to add more commentary today. I hope I'll have more to say tomorrow. 


Wednesday, December 14, 2011


Well, well, well. After lunch today: another really low post-prandial result. I'm on a roll. Or on a non-roll, as you might say: my lunch included no bakery products, just chicken and non-starchy vegetables. And I ate lunch right after I got back from a good run. So, the explanation for the low post-prandial result after lunch today is probably the same as the explanation for the similar result after dinner last night: hard workout + low-carb meal = lower-than-usual post-prandial glucose.

How much lower than usual, though? A result of 76 after eating lunch (even a low-carb lunch) is a lower result than I ever expect to see. I expect a "low" post-prandial result to be in the 90 to 100 range, not in the 70 to 80 range. But I do go through periods like this from time to time, when I get unusually low results for a few days; these streaks usually start for no obvious reason and end for no obvious reason. I have to assume that the cause is some short-term fluctuation in my insulin sensitivity, in my insulin productivity, or in the glucagon-regulated glucose output of my liver. I don't know why such fluctuations occur, and I strongly suspect that nobody else does, either.


My work site consists mainly of four large 2-story office buildings, descending like a staircase down the side of a steep hill. The bottom floor of each building is even with, and connects to, the top floor of the next building down. My building is at the bottom of the hill. Over the course of the day, I need to get around a lot, from building to building and from floor to floor, so I end up climbing a lot of stairs -- indoors and outdoors.

Now, it is actually possible to get around in all these buildings by using elevators. It is never the most direct route to anywhere, because the elevators are tucked away in odd and inconvenient corners of the buildings. It isn't the most dignified way to get around, either, because these are very definitely freight elevators, and the ugly interiors of these conveyances send a strong message that they were not designed to make human passengers feel welcome in there. But, if necessary, a handicapped person would be able to get around at my workplace without using stairs, even though the able-bodied employees seldom do that. We may hate the stairs, but the stairs are what we use.

However, in many workplaces that exist within buildings with multiple floors, the choice between elevators and stairs does not so heavily favor the latter. In those workplaces, people usually avoid the stairs. Often they don't admit that they avoid the stairs because they hate working that hard and getting embarrassingly out of breath. No, they say the real reason they choose the elevator is that they don't have time to take the stairs.

However, the excuse that using the stairs would take too long is highly questionable. A Canadian study found that doctors working in a six-floor hospital who used the stairs instead of the elevator saved themselves about 15 minutes a day. This was so because it took the doctors about 13 seconds to climb one floor; using (and waiting for) the elevator took 37 seconds per floor. In the study, doctors were timed as they made a set of 14 climbs between various floors, distant or nearby. Staircase-users spent 10 minutes doing all those climbs, while elevator-users spent 20 to 25 minutes.

Was there an advantage besides the time saved?

"As for the health effects of stair-climbing, not much research has been done. But in a study published last year, Dr. Philippe Meyer at the University Hospital of Geneva, Switzerland, found that encouraging sedentary workers at his hospital to take the stairs had some tangible benefits. After 12 weeks, those workers were climbing an average of almost 21 flights per day, versus 4.5 before the stair-climbing campaign. And with those extra steps came small improvements in waist size, body fat, blood pressure and LDL cholesterol."

I certainly wouldn't count the stair-climbing I do every day at work as the equivalent of a workout; it can't be as significant medically as my daily 4+ mile runs. But for people who don't exercise in any other way, maybe using stairs makes a difference. Perhaps, in the days when most apartments and office buildings had a staircase but no elevator, people had healthier hearts for that reason.

Of course, some people use stairs specifically as an exercise tool. It can be done, and it's considered an especially effective form of aerobic exercise.

What a shame that it is also an especially miserable form of aerobic exercise. There's nothing like the sight of a tall, steep staircase to deprive you of the will to continue.


More breaking medical news: according to The Onion, researchers have found that getting smacked in the face with a tree branch is very unpleasant at first, but people soon get over it.

Well, I guess worse things than that have happened to me in a clinical setting.


Tuesday, December 13, 2011


Why the low post-prandial result this evening? Probably because I had a low-carb dinner, and I ate that dinner right after getting home from a fairly hard gym workout. The gym workout included weight-training, which can boost your insulin sensitivity a lot; I assume that's what happened here.


In yesterday's blog posts about sugary children's cereals, I spent a lot of time poking fun at the people involved in that sleazy business, and very little time talking about why it matters -- or even if it matters -- that children's cereals are full of sugar. 

Is sugar really so evil? 

Well, sugar certainly has its drawbacks. I imagine that feeding children large doses of sugar is not especially helpful in preventing tooth decay (or other infections in the mouth and throat). And from the standpoint of meeting nutritional needs (especially the need for such things as fiber and vitamins), it isn't unfair to deride sugar as a source of "empty calories", as compared to whole grains. And this business of adding sugar to everything seems to have the effect of sneaking a lot of extra calories into commonplace foods -- so that people gain weight on a diet which does not seem especially indulgent.

Purely from the standpoint of diabetes management, however, the distinction between one form of carbohydrate and another is not as large as we would like to think.

We often make too much of the distinction between sugar and "complex" carbohydrates (that is, starch). Replacing a bowl of Honey Smacks with a bowl of oatmeal (which was recommended by the authors of the EWG report I discussed yesterday) may reduce sugar intake, but if it increases starch intake at the same time, how much have things actually improved? A substitution of this sort isn't likely to bring down the carbohydrate total, and might very well increase it. Also, preparing oatmeal takes time and effort (especially if it's good oatmeal -- meaning steel-cut oats, not that quick-cooking mucilage), so if you bother to make the stuff at all, you're not likely to make only a tiny serving of it.

Admittedly, starch is digested more slowly than sugar, and the starch in whole grains is digested a little more slowly than starch in a more refined form. But how big is that difference of timing? How much longer do whole grains take to give you a glucose peak -- 60 minutes instead of 20 to 30? Any generous serving of any kind of carbohydrate is going to give you a substantial glucose peak somewhere within the following hour, and I'm not sure it matters all that much how far into the hour the maximum occurs.

To be sure, glycemic response varies a lot from person to person, and some people may get a significantly lower glucose peak from refined grains than from sugar -- or a significantly lower peak from whole grains than from refined grains. However, such things must be proved, not assumed. I'm afraid that a lot of people simply take it for granted that a bowl of Honey Smacks is bad for their blood sugar, but a bowl of oatmeal is not. If you have data from your glucose meter to confirm that, fine. If you don't, you have no grounds for believing it -- no matter what articles you may have read on the subject. 


Monday, December 12, 2011


An organization called the Environmental Working Group (a non-profit organization which keeps track of environmental risks to health) recently decided to take a break from investigating industrial pollution sources, and began looking into a (slightly) different subject: sugar in children's breakfast cereal.

But before we discuss the EWG's findings, what exactly do we mean by the phrase "children's breakfast cereal"? The meaning seems, at first glance, to be perfectly obvious -- and yet it's hard to define with any precision. We are tempted to take the same shortcut that judges have used in defining pornography: we know it when we see it.

If you stroll down the cereal aisle of your local grocery, you won't be in much doubt about which cereal boxes are designed to catch the eye of a pre-adolescent, and which cereals are designed to please an immature palate. But can your subjective reaction be translated into objective criteria? Exactly what is it that tells us Cap'n Crunch is probably not intended for the sophisticated adult consumer?

I think there are three easily-recognized markers of a children's cereal:

  1. Is the box so garishly colored that it makes you want to put sunglasses on?
  2. Is the box decorated with an insanely-grinning cartoon character who seems to be bursting out of the box in 3D?
  3. Does the cereal contain so much sugar that it is bound to taste like a bowl of candy?

I would argue that a cereal which satisfies those criteria is unquestionably a children's cereal -- but this point can be disputed, as we shall see.

Back to the EWG. They ranked children's cereals by their sugar content, and found that 44 of them provide more sugar, in a one-cup serving, than 3 "Chips Ahoy!" chocolate-chip cookies. The worst cereal on their list: Kellogg's Honey Smacks, which is nearly 56% sugar by weight. A cup of this cereal contains more sugar than a Hostess Twinkie.

Not so fast, says Kellogg!

The folks at Kellogg feel the EW's report is unfair to them (perhaps because 5 of the 10 most sugary cereals on the list are made by Kellogg?). Regarding Honey Smacks, three important facts need to be kept in mind -- facts which you would never know from the EWG report:

  1. The recommended serving size for Honey Smacks cereal is 3/4 of a cup, not a full cup -- so a proper serving of this cereal would not have quite as much sugar as a Twinkie.
  2. "Kellogg has reduced the sugar across our U.S. kids' cereal by approximately 16%" (say Lisa Sutherland, vice president of nutrition at Kellogg North America).
  3. Honey Smacks cereal is not marketed to children, so this particular cereal is not relevant to any discussion of sugar in children's cereals.

My thoughts on these points...

This business of "recommended serving size" is a legal loophole which should long ago have been closed. We allow food makers to define the serving size of their product -- ignoring the question of how much of it a consumer would actually eat, and instead making up a hypothetical serving so small that it yields non-threatening numbers for the nutritional label. Most consumers don't even notice the serving size when they look at a nutritional label; they notice the calorie count or the carb count, and they assume that these numbers reflect the sort of serving that most people would eat. But most people do not pour 3/4 of a cup of cereal into a bowl and say, "Okay, that's about enough!". Have you ever even seen what a cereal bowl looks like with only 3/4 of a cup of cereal in it? I wasn't sure I had, either, so I conducted the necessary experiment. It looks likes this:

I assume you never pour more cereal than that into a bowl, do you? Hardly anyone ever does, I'm sure...

Regarding Kellogg's second point: if they reduced the sugar in their children's cereals by 16% and still ended up occupying 5 of the top 10 slots on the most-sugary-cereals list, they still have some reducing to do.

But what really intrigues me is Kellogg's third point -- the assertion that Honey Smacks cereal is "not marketed to children". This claim seems to strain credulity just a little bit. A cereal that is more than half sugar by weight? A cereal that's sold in a box that looks like this? 

If Honey Smacks is not aimed at kids, what exactly is the target market for this product? Health-conscious gourmets who go to the farmer's market every Saturday to pick up a basket of organic vegetables? (Probably not.) People in their sixties who are under doctor's orders to get more fiber into their diet? (Probably not them, either.) Art students too heavily influenced by Andy Warhol's junk-is-cool aesthetic? (Not a large enough demographic for Kellogg's to spend a lot of money chasing it.) Call me paranoid, but it sure looks to me as if the target market for Honey Smacks is young children.

Apparently the people who make this stuff are competing with the people who make soft-drinks to see who can cram the most sugar into the nation's children. Exactly how big a factor this is in the rising prevalence of childhood obesity and early-onset Type 2 diabetes is hard to evaluate, but I can't believe it plays no role.


A little bit of winter-like weather finally arrived on Sunday -- cloudy, foggy, and cold by local standards. Cold enough for me to put on long sleeves and long pants when I went trail-running, anyway.

I thought the weather might discourage people from exercising in the state park, but there were squads of them.

It was almost as if the gray skies made more people want to get outdoors and exercise, after so much sunshine recently.

Lots of wild turkeys were about, too.

Exercising outdoors in the winter, provided you're dressed properly for it, can feel more refreshing than exercising in warmer weather.

And, of course, the hot shower afterwards is great.


Friday, December 9, 2011


I came into the office this morning, checked my e-mail, and found a message at the top of the queue, from a woman in the marketing department. The subject line: More cupcakes to taste today.

The message read: Today we have peppermint chocolate and maple bacon (yes... BACON).  Stop by to try one... or two... or more.

(Well, I did say she was in the marketing department.)

This is the way we live now. There is no getting away from this kind of thing, even when it isn't Christmastime. If the historians of the distant future ever dig through the ruins of my office and find that message, I guess it will tell them most of what they need to know about our civilization (and its fall).

I stuck to my guns, though; I ended up not having the cupcakes. Not the peppermint chocolate, and not the maple bacon (yes...BACON), either.


Which has a more harmful impact on your health: gaining weight, or losing fitness?

Before going even one step further with this discussion, let us admit that it isn't at all easy to separate these two issues from one another, and treat them as if they were unconnected. People who gain weight typically lose fitness. People who lose fitness typically gain weight. Therefore, when a study claims to show that one of these things matters and the other doesn't, we need to be cautious about accepting the claim as credible.

But anyway, here's what researchers at the University of South Carolina found: 

"Fitness appears to trump weight loss when it comes to reducing all-cause and cardiovascular mortality. Data from a large longitudinal study show that maintaining and improving physical-fitness levels were associated with lower risks of all-cause and cardiovascular disease mortality, whereas changes in body-mass index (BMI) were not."

The researchers found, during a follow-up period of 11 years, that men who lost physical fitness had a significantly higher risk of cardiovascular death -- regardless of what was going on with their weight. Dr. Duck-chul Lee and his research colleagues stated:

"To date, extensive attention has been given to weight loss. However, the long-term effect of fitness change, primarily resulting from increasing physical activity, is likely to be at least as important as weight loss for reducing premature mortality. Increased attention needs to be placed on strategies to maintain or improve fitness." Men who became or remained fit had a significant 47% and 48% lower risk of all-cause mortality, respectively, compared with men who remained unfit. Similarly, men who became or remained fit had a 41% and 44% lower risk of cardiovascular disease mortality, respectively."

In contrast, changes in BMI status were not associated with mortality after adjustment for potential confounders, including changes in fitness. For example, becoming normal weight did not appear to have any benefit in terms of reducing the risk of all-cause and cardiovascular death, while becoming overweight did not appear to increase the risk of mortality. Lee noted that men who increased their physical-activity levels were significantly more likely to maintain their cardiorespiratory-fitness levels, adding that 80% of men who maintained their fitness levels increased their physical-activity levels.

Well, the bottom line seems to be that losing fitness is more harmful to your health than gaining weight. But any conclusions we draw from this study are contingent on one thing: have the researchers, or have they not, successfully separated the effects of weight from the effects of fitness? Or have they muddled up a bunch of data and drawn iffy conclusions from it?

As an advocate of exercise for health (and as a guy who is better at maintaining his exercise regimen than he is at maintaining his weight), I am naturally prejudiced in favor of this study's conclusions. But I try to retain my skeptical attitude about studies, even when the studies are saying something I'm inclined to agree with.


Thursday, December 8, 2011


Usually I eat lunch right after I run, and I'm sure that the rather low post-prandial results I often get after lunch reflect the short-term boost in insulin sensitivity which tends to be triggered by a workout. 

Today, though, my schedule was thrown off by a lunchtime meeting, and I didn't get a chance to run until later in the afternooon. I was curious to see how much higher my after-lunch test result might be today, with no pre-lunch workout (and no workout yesterday, either). The answer: not much higher. I'm often below 100 after lunch, if I run before; but 119 is within the range of normal glycemic response for a non-diabetic individual, so I have nothing to regret. And I hadn't "cheated", if that is the word, by eating an unusually low-carb lunch. (I had a sandwich, in fact, so it was a fair test.)

For contrast, I also checked after dinner, and got 100. But of course that was after my afternoon run (a pretty hard run), and it was a low-carb dinner of fish and vegetables, so there's not much doubt about why it was lower.

If the difference in my glycemic response between a meal without bread, following a workout, and a meal with bread, following no workout, is no worse than 20 points (on this occasion, anyway), then it looks as if my endocrine system is functioning pretty robustly -- especailly for a guy who was diagnosed with diabetes a decade ago. This, at least in my interpretation, is evidence that we don't have to accept the usual doom-and-gloom view that diabetes always gets worse and diabetes patients always have to sit there helplessly, watching the tragedy unfold.

I'm not claiming to be invincible, nor am I claiming that it's always easy for me to keep things under control. But the fact that I can at least get things back under control, at this late date, suggests that the concept of diabetes as an inescapable downhill slide may be a bit exaggerated.


"Women on Night Shift Have Increased Risk for Type 2 Diabetes" says the headline. Just women? Well, maybe not, but the study only looked at women. Specifically, it looked at nurses, so the headline could have been "Nurses on Night Shift Have Increased Risk for Type 2 Diabetes". More specifically, it looked at nurses on a rotating shift that included night duty part of the time, so the headline could have been "Nurses Rotating On and Off Night Shift Have Increased Risk for Type 2 Diabetes".

"The study, led by An Pan, PhD, a research fellow in the Department of Nutrition at the Harvard School of Public Health, Boston, Massachusetts, found that women who had worked rotating night shifts at least 3 times per month for 1 to 2 years had a 5% increased risk of developing type 2 diabetes. The risk rose the longer a woman had done rotating night-shift work, from 20% after 3 to 9 years to almost 60% for 20 or more years."

Over 100,000 women were included in the study, so there is a lot of data to examine. It shows, among other things, that night-shift work was also associated with weight gain. Other studies have found a link between night-shift work and metabolic syndrome, obesity, and "glucose dysregulation" (which appears to mean "any abnormality in regulation of glucose levels, even if we're not quite willing to call it diabetes").

"Dr. Pan and colleagues theorized that the link between night shifts and diabetes could be the result of disrupted circadian clocks, which help regulate the gastrointestinal and endocrine systems. Sleep disruption is also known to contribute to insulin resistance, increases in appetite, and weight gain, and night-shift work has previously been linked to negative changes in health behaviors including smoking, irregular meals, and lack of exercise."

Gosh, do you think there are enough variables in this mix? Obesity, sleep disruption, insulin resistance, changes in appetite, irregular meals, lack of exercise... There's a lot going on here. And then, of course, there are the variables that Dr. Pan and colleagues haven't thought of. Could it be that there are other differences between nurses who take rotating night-shift assignments and nurses who don't -- differences of which Dr. Pan and colleagues are unaware? If so, could it be that these differences have a potential impact on health?

These things can get really complicated. (By "these things" I guess I mean "issues involving human beings".)


The trouble with doing research on human health is that we can't really do any controlled experiments. We don't breed people in captivity (well, not exactly) and we don't raise them from birth to death in cages. Human participants in any kind of long-term health study are still going to continue living their lives while the study is going on, and most of the time they are not going to be under the control, or even the surveillance, of the researchers. Who knows what might be going on, unobserved, in their daily lives? And even if it's a short-term study, in which the participants spend a few hours in a clinic under observation, who knows what might have gone on, unreported, in their personal histories? Observing them in the clinic tells you nothing about their past. There might be things you need to know about them, and don't.

All this uncertainty really gets in the way of setting up experimental controls. Ideally, you divide the participants into a study group and a control group, and arrange it so that there is just one difference between the two (the study group gets the experimental drug and the control group doesn't, or whatever). If there is only one difference between the two groups, then you have a solid basis for comparing their results. But if there are actually multiple differences between them -- or if they don't actually differ, in the one crucial way that you think they differ -- then who knows what the results mean? If we don't know about all the similarities and differences between the study group and the control group, then we might be very much misled by the data we collect. And how can we know about all of those similarities and differences, considering how complicated people's lives are, and how incompletely they share information about themselves?

Sometimes it isn't even possible to set up a control group in the first place. I once read a sad tale about a British researcher who wanted to study the long-term effect of pornography on the male personality (to see if it actually caused any of the psychological and behavioral problems which critics, especially feminist critics, have been attributing to it). To begin the study, he needed to assemble a study group of men who looked at pornography, and a control group of men who didn't. The trouble was, he couldn't find any men who didn't, so he had to abandon the study before it began.

If people were being raised in cages like lab rats, such things could be strictly regulated, but that's not how it is. Therefore, most health studies are not very strictly controlled. They compare two groups of people whose similarities and differences are not thoroughly understood. To take an extreme example: a scientist who thinks a particular food is bad for us compares disease statistics in populations before and after the historical date when that food was introduced into the human diet. If a disease became more prevalent after the food was introduced, then that food promotes that particular disease -- right? But accepting this as proof requires us to assume that the world did not change in any other way between the two time periods, and therefore the two populations being compared are alike in every way except for a single change in diet -- which is a lot to assume. Not all health studies are quite that uncontrolled, but the results of any study that doesn't have excellent controls can only be suggestive, never probative. If you want proof, you'll have to look elsewhere.

Nevertheless, the results of a study are usually announced as if a point had been proved. Whenever this happens, we need to think about all the things we don't know for sure about possible similarities and differences between the study group and control group -- especially when different studies support opposing conclusions about the same issue!


Wednesday, December 7, 2011


I seldom do a post-prandial test after breakfast. The main reason is that the timing isn't good, especially on a weekday. At the point I would need to test, I'm usually in the middle of a morning meeting at work. Let's face it, I'm not going to walk out of a meeting to go do that. And I'm certainly not going to do a blood test right there at the conference table, in front of everybody. There's enough bloodshed going on at business meetings as it is.

But, as luck would have it, the morning meeting was canceled today. I figured I might as well take the opportunity to do an after-breakfast test, to make sure breakfast doesn't hit my system any harder than lunch and dinner do. Apparently it doesn't, if I only went up to 101. And it wasn't an exceptionally low-carb breakfast. (I had toast, after all, which is what people with diabetes call "taking a walk on the wild side".)

As I said on Monday, in my discussion of getting by on a limited supply of test strips, the tests to skip (if you have to skip some) are the ones that are usually the least problematic. In the past, I have found that I tolerate carbs better in the morning than I do later in the day, so I am less concerned about post-prandial spikes after breakfast than after lunch or dinner. But, as I also pointed out in the same discussion, patterns of that sort can change over time, so you do sometimes need to do the sort of tests you would normally skip on the grounds of predictability -- just to make sure they haven't become less predictable. Many people with Type 2 report that their tolerance for carbs is at its worst after breakfast; what if I have become one of those people without realizing it? It's worth checking into, even if you don't get the opportunity to check into it every day.


It occurred to me today -- perhaps a little later than it would have occurred to a more alert person -- that I haven't had a day off exercise in more than a week. So today is it!


The people who say that Type 2 diabetes has nothing to do with obesity have never been very good explaining why Type 2 diabetes is so often alleviated by weight loss. They have even more explaining to do now: a new study finds that obese Type 2 patients who lose weight experience an improvement in the functioning of pancreatic beta cells (the cells which produce your insulin supply -- and which, supposedly, go into irreversible decline once you are diagnosed with diabetes).

Apparently that decline is not quite so irreversible as we thought.


Tuesday, December 6, 2011


Another one of those odd days when my fasting test is higher than my post-prandial test. Oh well -- neither result is outside the normal range as it is usually defined. 

We're continuing to have a beautiful winter -- lots of sunny days, and weather that's warm enough for running in shorts and a short-sleeved shirt. May it continue!

Google referred two people to this site who had entered the search string "running without clothes". Alas, I can't  quite live up to their expectations. It's not that beautiful a winter.  


A reader who has a problem with high glucose after exercise (especially after intense exercise, such as a footrace) has been wondering what might be the explanation for this. It's an issue I haven't payed attention to until now, for the selfish reason that I haven't been having a problem with it myself. But I know that a lot of people with diabetes do have a problem with it, so if I'm going to be recommending exercise to people with diabetes, the least I can do is find out what's up with exercise-induced hyperglycemia, and see if there's anything people can do about it.

So, I looked into the matter, in my usual shallow yet energetic way. And, in the process, I not only discovered an explanation for the problem which hadn't occurred to me, I also discovered some possible reasons why other people are having this problem and I'm not.

To begin with, we need to differentiate between "moderate" exercise and "intense" exercise -- because, as it turns out, glucose regulation operates very differently during those two activities. But first: how exactly do we define the difference between moderate and intense exercise?

Well, if we are exercise physiologists, we define moderate exercise as <60% of VO2max and intense exercise as >80% of VO2max. VO2max refers to the rate of oxygen usage at the highest possible exercise intensity. It is measured, in a clinical setting, by monitoring your respiration and blood-gases while you "exercise to exhaustion" on an accelerating treadmill. Clearly this is not something you can do at home, and most of us are never going to have it done in a clinic, either. So, we can only make a very rough, subjective estimate of how close we are to 100% of V02max, based on how much physical stress a workout is causing us. (Here's how I see it: at 60% of V02max you are merely wanting to die; at 80% of VO2max you are expecting to.)

Anyway, it turns out that glucose regulation operates rather differently in wanting-to-die exercise than it does in expecting-to-die exercise. The reason for this relates to the way the body burns chemical fuel to sustain physical activity. Fat contains more than twice as much energy per unit weight as sugar, and because of this efficiency, fat is the body's preferred fuel most of the time. But fat has a drawback: it doesn't burn as fast as sugar. When the rate of energy expenditure is very high, the body simply isn't can't burn fat fast enough to obtain the amount of energy it needs. So, as exercise intensity increases, the body has to start burning more and more sugar. At a high level of exercise intensity, the muscles are being fueled almost entirely by sugar-burning, not fat-burning (fat is still burned, but it is used mainly to sustain basic life functions -- such as keeping the brain operating). At a lower level of exercise intensity, things are not so lopsided; sugar-burning is not so dominant. During any kind of exercise, the endocrine system adjusts the proportions of the regulatory hormones it is releasing, to match the intensity level of the exercise.

So what does this mean in practical terms? Well, during moderate exercise, glucose production (that is, conversion of glycogen stored in the liver and muscles to glucose) increases -- but only by about a factor of 2. Glucose utilization increases by roughly the same amount, or a little more; therefore, blood glucose typically remains constant or decreases during moderate exercise, and declines after it. This is what I usually experience (as you know, moderation is practically my middle name).

During intense exercise, on the other hand, glucose production increases much more dramatically -- by a factor of 7 or 8. However, glucose utilization only increases by a factor of 3 or 4! Obviously, this imbalance between production of glucose and usage of glucose leads to increasing hyperglycemia as the exercise goes on. And then, when the exercise stops, glucose utilization drops before glucose production does; naturally, this aggravates the hyperglycemia.

In normal individuals, the pancreas responds to this situation by releasing an unusually large amount of insulin over the next 40 minutes or more, to bring the hyperglycemia under control quickly. But this corrective response may not be possible in individuals with diabetes, because in their case the pancreas may not be able to make enough extra insulin for the task. This is certainly true of Type 1 patients, and is often true of Type 2 patients as well. The bottom line is that, although intense exercise can trigger a temporary hyperglycemic episode even in non-diabetic athletes, the hyperglycemia tends to be more severe and more prolonged in athletes with diabetes.

This explanation of exercise-induced hyperglycemia suggests three possible reasons why I haven't been having a problem with exercise-induced hyperglycemia:

  1. I tend to exercise at a moderate pace. What I call running is probably called "jogging" (if not "tottering") by more serious runners. Maybe I'm not pushing myself quite hard enough to induce hyperglycemic episodes.
  2. Insulin productivity has not been compromised as much in my case as it has been in many cases of Type 2 diabetes, so my endocrine system is able to suppress hyperglycemia quickly -- so quickly that I never become aware of it.
  3. My endocrine system reacts less strongly to exercise than some patients' endocrine systems do -- so there isn't as much hyperglycemia to be suppressed when the workout is over.

Perhaps it's a combination of all three. But the only one of those factors that is under my control is exercise intensity. Therefore, if I were having a problem with exercise-induced hyperglycemia, I guess I would have to experiment with lower-intensity workouts to see if that made a difference. However, I think I would also sacrifice a bunch of test strips, and do a lot of testing after (and even during) workouts, to see if I could get a clearer idea of what was happening, and how consistently it was happening. If it wasn't happening during most of my workouts, I might decide that I could ignore the issue and still come out ahead. If it turned out that it was happening all the time, I would need to experiment (with lower-intensity workouts, or with whatever else seemed worth trying) to see if I could find a solution to the problem.

If I couldn't find a solution to the problem, I'd have to do some serious cost/benefit analysis, to try to figure out what exactly the tradeoff is between the raising of blood glucose during a workout and the lowering of blood glucose which tends to occur, the rest of the time, in people who work out regularly. But I doubt the cost/benefit analysis would come down in favor of giving up exercise.

One final point: you may be wondering why, during intense exercise, the body increases glucose production more than it increase glucose usage, thus creating an imbalance which promotes hyperglycemia. Isn't mother nature making a pretty dumb mistake here? My guess is that mother nature is playing it safe. During intense exercise, when the body is burning glucose at a very rapid rate, there is a heightened danger of running out of the stuff. So, the body ramps up glucose production more steeply than necessary, to ensure that there is a safety margin. (Going too high has its drawbacks, but going too low represents a far greater immediate threat to survival -- better not risk it!)


Monday, December 5, 2011


With health insurers cutting back severely on reimbursement for glucose test strips, and the economy in a state of near-collapse, a lot of diabetes patients are finding that they can't afford to test as often as they once did. Naturally, they're wondering what is the best way to get by on a less-frequent test schedule. I've been asked to share my own thoughts on this issue; I'll share them here... but I can't claim that this is anything more than my personal opinion. 

First of all, I hope this is not a problem affecting people who take insulin. It shouldn't be. Taking insulin without testing is dangerous (in a study of adverse drug reactions requiring hospitalization of elderly patients, just four drugs accounted for 67% of such incidents, and insulin was one of those four). Seemingly it is impossible to argue that people taking insulin don't need to test often; my impression is that patients who take insulin are not having their test strips taken away from them, and I hope my impression is correct. 

Unfortunately, the official thinking these days about patients who don't take insulin is that they don't need to test often, so there is no need to reimburse them for test strip purchases. The justification for this view is research showing that, for the average Type 2 patient, frequent testing doesn't improve glycemic control enough to be worth the trouble (probably because a lot of Type 2 patients fail to do anything with the test data they collect except get depressed about it). 

I hate to see that statistical chimera known as the Average Patient become the driving force in medical policy, even though I can understand why it happens (from an accountant's perspective it appears to make sense).

I say to hell with the Average Patient. Diabetes management consists mainly of trying not to be anything like the Average Patient. I would rather see testing evaluated in terms of how much it can benefit people who do it right, not in terms of how much it usually benefits people who do it wrong. Are we really going to adopt the principle that, if people typically fail at some task, that means it's not a good thing to do, and should be discouraged? We need to start discouraging marriage, then. With regard to glucose testing (if not marriage), maybe our focus should be on helping people do it better, rather than on advising people not to try it. 

Of course, this raises the question of what "doing it better" means in terms of glucose testing. What makes glucose testing useful, and how can we preserve its usefulness even when we can't test often? 

The whole point of testing is to find out enough about our glycemic response to be able to steer it in the direction of a normal glycemic response. By "glycemic response" I mean variations in blood glucose levels in reaction to something else. That something else might be a particular food (or drink); it might be exercise; it might be a long fast; it might be a midnight snack; it might be a sleepless night; it might be a cold virus; it might be emotional stress.

People who test regularly over a long period learn to recognize their own personal glycemic patterns, which might differ markedly from another person's. Some of us get our highest glycemic spike after the morning meal, while others get an unusually low one at that time, and are much higher after dinner. Some people go high after a workout, and some go low. Some people get an exaggerated spike in response to a particular food, or even a particular ingredient in food (for example, some people report that high-fructose corn syrup spikes them a lot higher than an equivalent amount of table sugar). Some people say that brown rice doesn't spike them the way white rice does, while others see no difference between the two. 

Because individuals vary so much in terms of glycemic response, we need to test under various conditions, to see how those conditions affect us. And we can't just do that for a month or two, and conclude that we now know everything we need to know about our glycemic response. Glycemic response tends to change over time, especially in people with diabetes. We need to keep testing over the course of our lives, to make sure that the conclusions we drew from testing last year are still true this year. That doesn't mean we have to test several times a day, but it does mean we need to test on some kind of routine basis, and under conditions different enough to help us understand how we respond under varying circumstances. 

Okay, then, let us assume that you can only afford to test once a day, or once every other day, or even less often than that. How do you get the most value out of whatever testing you can afford to do? What sort of testing should you emphasize? Fasting tests? Post-prandial tests?

Here are my thoughts on on how to prioritize testing:

I'm not sure how helpful any of this is, to people trying to get by with fewer tests, but that's what I was able to come up with.


I wasn't able to do a blog post on Friday because, after work, I went to a fantastic party at the home of some neighbors of mine, a family from Ireland. They know all the good Irish musicians in the area, so when they have a party there's always a great jam session in the living-room. Their parties are always good, but the musical quality was exceptional at this one. There was a very high density of good musicians per square foot.

For me, time really flies when I'm playing at a good session. I looked at my watch to see if it was 10 o'clock yet, and it was past midnight.

I had a lot to eat and drink (which made getting up the next morning and doing a hilly 5.3-mile run just a bit challenging), but I tried my best to make up for it during the rest of the weekend. I ran an 8-miler on Sunday. My theory of party-attendance for diabetes patients is: you can still go to parties, but you have to pay for it later. 

I was a little surprised by the low post-prandial result after lunch today, so I did another post-pranidal test after dinner just to see if anything unusual was going on. The second post-prandial test was low, too, but not as low as the one after lunch. Anyway, these aren't weird results; they''re just better-than-expected results. I guess I should just be happy with them and shut up.


Thursday, December 1, 2011


I was relieved to find my fasting test lower today. Maybe it was worth it to refuse those cupcakes yesterday.


Ah, another seasonal ritual: today at lunchtime we held the first rehearsal for the Christmas-music strolling players at my workplace.

On the last day before the Christmas holiday, we wander around from building to building and from floor to floor, playing a few Christmas tunes at each location and then moving on. We've been doing this since 1996, but with a changing roster of musicians (mainly owing to layoffs, I'm afraid -- a grim fact which always gives me very mixed emotions about the whole thing). It's not the most serious and well-prepared performance any of us ever do, but we like to fit in weekly rehearsals during December to get ready for it.

I think we'll have seven musicians this year: two fiddles, two guitars, a flute, a trumpet, and a French horn. Only five of us made it to the rehearsal, but it went quite well, especially considering that none of us had played any of these tunes since last year at this time. That's the trouble with Christmas music: you're always rusty on it, because you've always gone at least 11 months without playing it.

Oh, and about the word "fiddle": any use of this word makes people wonder what the difference is between a fiddle and a violin. There isn't any; it's the same instrument. Fiddle is the English word for it (and has been, for as long as there has been an English language). Violin (from the Italian violino) is professional jargon: it's a classical musician's way of saying fiddle, just as cranium is a doctor's way of saying skull. The word violin is used mainly in a classical context. Saying violin when you're talking about folk music is like saying cranium when you're talking about Halloween decorations: it's not incorrect, but it sure doesn't sound right.

Rehearsing at lunchtime meant that I couldn't go for a run -- which is just as well, as the wind today would probably have knocked me over if I'd tried it (I was relieved to come home and find that no redwood tree had crashed through my roof). I went to the gym after work instead, did some weight-lifting, and then used the stair climber.

I like to sit in the hot tub at the gym after I do any weight-lifting, for a little therapeutic heat. But it's outdoors, so I had to walk outside in the cold wind to get to it. It was worth it. It was a clear night, and as I soaked in the hot water I was able to look up and see Cassiopeia, Perseus, Taurus, the Pleiades, and the bright, cream-colored planet Jupiter. I like to think that this sort of thing gives me a sense of perspective on my petty little problems. It doesn't, really, but I didn't say that it does, I just said that I like to think it does.


Here's a bit of weird medical news:

Michael Jackson Doctor Sentenced to 4 Years in Jail

LOS ANGELES (Reuters) Nov 29 -- Michael Jackson's personal physician, Dr. Conrad Murray, on Tuesday was sentenced to four years in jail and denied probation for his conviction on a charge of involuntary manslaughter in the pop star's death... Dr. Murray, 58, dressed in a gray suit and purple paisley tie, sat emotionless throughout much of his sentencing in the trial that captured the world's attention. Just before he was led out of the courtroom, he blew kisses at a woman who shouted "we love you" to the convicted killer.

At first I thought the woman's reaction a trifle strange. But then I thought: if child molesters with cronies in the Penn State football program can have a cheering section, why not doctors who kill their patients? 

In our modern world, it's pretty much impossible to do anything so bad that you don't get to have any fans.


Older Posts:

November 2011
October 2011
September 2011
August 2011
July 2011
June 2011
May 2011
April 2011
March 2011
February 2011
January 2011
December 2010
November 2010
October 2010
September 2010
August 2010
July 2010
June 2010
May 2010
April 2010
March 2010
February 2010
January 2010
December 2009
November 2009
October 2009
September 2009
August 2009
July 2009
June 2009
May 2009
April 2009
March 2009
February 2009
January 2009
December 2008
November 2008
October 2008
September 2008
August 2008
July 2008
June 2008
May 2008
April 2008
March 2008
Jan/Feb 2008