3rd Thursday Update

November 19, 2015

 


Fall color, California-style

It's starting to get cold at night. Or perhaps I should say it's getting "cold". Scare-quotes are needed when Californians discuss low temperatures. I'm only talking about the low forties here (or, for those of you living in modern societies, five to seven degrees). It doesn't get cold enough here to produce the sort of colorful foliage you see in less temperate climes. Still, we do get patches of red here and there, and we're getting them now. Who knows? We might even get some precipitation. Snow isn't really a possibility, but rain is. Certainly we need some.

Regardless of what might be happening to the temperature at night, the daytime weather is remaining obstinately sunny and mild, so there's no excuse for not going outside to exercise. I went running this afternoon, and there was nothing remotely wintry about the experience.

However, preparations for Christmas music at work continue. We were asked to choose a band name for our ensemble of musicians, and I threw it open to suggestions from the participating players. One of them suggested that we call ourselves "The Keytones" (this refers indirectly to the name of our company), and others liked it. I had to point out gently that it might be better not to give ourselves a name that sounds like something people test their urine for.

 


Eat your insulin!

When insulin therapy for diabetes was first made practical in 1922, diabetes patients were deliriously happy and grateful for this life-saving treatment of a formerly untreatable disease.

However, that attitude wore off over the years. The trouble with insulin was that you had to inject it -- and a lot of people really, really hate needles. For generations now, diabetes patients have been asking why you can't take insulin in the form of a pill. Almost every other medication can be taken in the form of a pill -- why not insulin, for crying out loud?

The answer is that insulin isn't a medicine, at least not in the sense we usually think of some laboratory-created chemical as a medication. Insulin is a naturally-occurring hormone (although it doesn't occur naturally in the bodies of people with Type 1 diabetes, and not enough of it occurs naturally in the bodies of some people with Type 2).

Insulin is a "peptide" hormone, meaning that it's a protein chain, made up of amino acids linked together. And the trouble with a protein chain is that the human digestive system is designed to break down proteins into their constituent amino acids. Even though proteins are all assembled from a limited set of amino acids (a kind of chemical alphabet for creating protein "words"), the possible variations on a protein are endless, and each species forms its own unique versions of the various protein configurations. When you eat chicken, beef, or beans, your body doesn't just take in the proteins in those foods in whatever form the chicken, cow, or bean plant created them. Your digestive tract breaks down proteins into their basic amino-acid building blocks, so that the body can start building human proteins out of them.

If you take insulin orally, it's not going to remain insulin for very long. As soon as it hits the digestive acids in your stomach, it will start breaking down. And if any of it gets through the stomach without being broken down, digestive enzymes in the small intestine will finish the job. Insulin taken orally is just not going to enter your bloodstream as insulin. If you want that to happen, you have to bypass the digestive tract entirely, and inject the stuff directly into your bloodstream.

For many years this has been considered an unsolvable problem. Diabetes patients who needed to take insulin would simply have to stop complaining about it, and take the shots, because there wasn't any way that insulin could be taken orally.

However, researchers have never entirely given up on the dream of finding a way to administer insulin orally -- and for good reason. Patient resistance to taking insulin shots has proved to be very high -- and this is particularly true for Type 2 patients. If you have Type 1, there is really no way to avoid taking insulin shots, because your pancreas has stopped making the stuff. If you have Type 2, your body is still making insulin, but the amount it's making (whether normal, or abnormally low, or even abnormally high) is insufficient, owing to diminished sensitivity to insulin. Type 2 patients have, at least for a while, the option of controlling their blood sugar without insulin injections, either by exercising (my solution) or by taking drugs which address the problem in other ways.

Because Type 2 patients are not reduced to a single option, they tend to avoid insulin as an option as long as they possibly can, even if whatever else they are doing to control their blood sugar is clearly not working well enough.

Insulin therapy is seen, at least by Type 2 patients, as the option of last resort -- and also as an admission of defeat. If you agree to start taking insulin shots, you are confessing that you failed as a diabetes patient. (And some doctors, or so I am told, use insulin therapy as a threat: "you'd better do what I'm telling you to do, or else I'll put you on insulin".) Because insulin therapy is carrying such a heavy symbolic weight, patients are primed to look for ways to avoid it -- at all costs. Many patients delay taking insulin until years after they are past the point where it appears to be needed. And a lot of this avoidance behavior is based on the practical detail that insulin has to be injected. If you could take it as a pill, people wouldn't be so alarmed by it -- hence the perceived need for a way to take insulin orally.

Well, it looks as if his Holy Grail of pharmaceutical product-development has finally been found. Researchers at the University of California at Santa Barbara report that they have found a way to embed insulin-secreting patches into a capsule.

The capsule is coated in a way which allows it to survive the digestive acids in the stomach and pass through to the small intestine. Once inside the small intestine, however, the capsule breaks open and release the patches, which are "mucoadhesive", meaning that they tend to cling to mucous membranes -- such as the lining of the small intestine. The patches resist breakdown by digestive enzymes, and they adhere to the intestinal wall. While there, they release insulin into the lining of the intestine, so that it can secrete the stuff into the bloodstream.

I believe that the coin in the photo above is there only to make us aware of the small scale of the pictured items. I'm sure money plays a role in all this, but not in that denomination.

This new method of administering insulin would also work for other proteins, not just insulin, so capsules of this type could also be used for oral administration of other things which must be injected currently -- such as vaccines, antibodies, and growth hormone.

Steroids, too! Disposing of needles discreetly will be one less thing for athletes to have to worry about. (The UCSB researchers aren't making that point; I just thought it might be worth mentioning.)

The new insulin capsules are still being tested to ensure their effectiveness and safety; I don't know when they are likely to become available to diabetes patients. I imagine they'll be popular when they do show up on the market.

I'm curious to know how they will affect the way people feel about insulin. Will insulin shed its current stigma as a therapy of last resort, which Type 2 patients begin taking because they have "failed"? Will insulin come to be seen as just another pill, a run-of-the-mill pharmaceutical which carries no special symbolic burden?

Because insulin is something the human body is supposed to produce anyway, I think that (with needles taken out of the equation) people will see it as "natural" and therefore no big deal. Well, it is natural to have insulin in the bloodstream. However, it is only natural to have a certain amount of it.

For patients with Type 1, the goal of insulin therapy is to give them only as much insulin as would normally be present in a healthy person. Patients with Type 2, who are insulin-resistant, are often given a good deal more insulin than would normally be present in a healthy person. Hyperinsulinemia (an abnormally high concentration of insulin in the blood -- which can be produced either by insulin treatment, or by the body itself, as a way of compensating for insulin resistance) is a problematic thing. It can bring blood sugar under control, but often at the cost of triggering inflammation of the blood vessels, which tends to promote atherosclerosis and, ultimately, coronary heart disease.

In other words, I can see a potential downside to any development which de-stigmatizes insulin, and makes Type 2 patients feel more comfortable about taking large doses of it. Fear of needles is not the only reason for a Type 2 patient to want to avoid insulin therapy. Avoiding a therapy which is itself a risk factor for heart attack simply makes sense. Of course, it only makes sense if you can find some other way to keep your blood sugar under control. My preference, for as long as I can keep making it work, is not to take insulin, but to improve my sensitivity to the stuff (by means of exercise) so that I don't need to take it, either as an injection or as a pill.

But if I do ever need to take insulin, I guess I'll be glad to know that, by then, it will probably be available in pill form!

 


Mice in the news!

I don't know why, but it's mice everywhere I look today, at least on the internet:

From The Onion:

From Science Daily:

From Dilbert:

 


2nd Thursday Update

November 12, 2015

 


Causes of death: what's popular?

By and large, people have an accurate idea of whether or not they're going to die some day (spoiler alert: they are), but they seldom have an accurate idea of what is likely to make that happen. They tend to worry more about unlikely threats than likely ones. And even if they do recognize that they are less likely to be killed by a malfunctioning aircraft than by a malfunctioning coronary artery, they usually don't know how much less likely that is. The disproportion between common and uncommon causes of death is greater than we tend to assume.

The National Health Service in the UK decided to clear the situation up, by issuing charts comparing the most common causes of death to the less common ones. (Although these reflect deaths in the UK, the numbers are reportedly similar in the US.)

The big three (heart and circulatory disorders, cancer, and respiratory disorders) dominate the chart pretty strongly: most of us are going to die from one of those. Of course, each of these encompasses a variety of specific problems. The top category combines heart attacks, congestive heart failure, peripheral artery disease, carotid artery disease, stroke, and various other things that can go badly wrong with the circulatory system. Still, those things are clearly related, and they're clearly a lot likelier to be the death of us than an accident or infection would be.

Although the green disk representing diabetes looks comfortingly small, I feel bound to point out that this is a little misleading. Diabetes is not very often cited as a direct cause of death, but it is often a contributing factor to circulatory disorders which are cited as a direct cause of death. (The most dangerous aspect of diabetes is its tendency to promote heart disease -- which is why I think the disease is best controlled by means of exercise, which reduces blood sugar and prevents heart disease at the same time.)

However, the NHS went beyond merely ranking the various causes of death. They also ranked the risk factors leading to those causes of death:

Here the contrasts are not quite so extreme -- there is a more gradual fall-off from the top three risks. High blood pressure, smoking, and high cholesterol are the top risks, but obesity, low consumption of fruit and vegetables, and physical inactivity are not too far behind.

At least the top three risk factors are comparatively easy problems to solve. High blood pressure is controllable with a wide variety of medications (and better controllable with exercise, at least in my experience). Smoking is... not even a topic for discussion in this day and age, or so I would have thought (I'm not surprised that it's so harmful, but very surprised that enough people are still doing it for it to have a big impact on mortality statistics). High cholesterol is also controllable with medications (and better controllable with exercise, at least in my experience).

I'm not claiming we can become immortal just by reducing risk factors, but it seems as if the risk factors outlined here provide us with a pretty good guide to staying healthier longer. Somebody's going to die before their time, but it doesn't have to be us.

 


More risk-assessment!

While we're on the subject of unpleasant things that might happen to us, but don't have to, a pair of researchers from the University of Nottingham in the UK (and where else could you be from, if you are going to claim that your surname is Hippisley-Cox?) have been working on developing web-based risk calculators. These online applications are designed to help doctors find out how much at risk their patients are for diabetes -- or, if they already have diabetes, how much at risk they are for the nastier consequences of diabetes.

I should warn you that their online apps will require most people, especially most Americans, to do some unit conversions. This isn't just a matter of using metric-system units for personal characteristics such as height and weight; they even expect hemoglobin A1c results to be given in mmol/mol rather than as a percentage (at least I found a converter tool for that one). They also expect you to be able to specify your cholesterol/HDL ratio, so see if you can figure that out from your latest lab report.

Their calculator for your risk of becoming diabetic is declared to be "only valid if you do not already have a diagnosis of diabetes", but I couldn't resist the temptation to use it, and see what my risk was (it turned out to be 16%, which I found gratifyingly small under the circumstances).

However, the more relevant calculator in my case was the one they developed to determine the risk of amputation and blindness. Those are the two potential complications of diabetes that are probably most feared, so I might as well find out what my risk was.

I ran into a surprising obstacle: after I had converted my most recent A1c result to their preferred unit of mmol/mol, and entered that value in their form, it rejected my value, because it was too low. They had defined a minimum value for the A1c result, based on what they assumed was the possible range for a diabetes patient, and my result was a little lower than what they had assumed the bottom of the range could be. So, just to get around that obstacle, I entered the minimum value that their form accepted. (Maybe that's the sort of problem you want to run into.)

It's often a problem with risk-calculation tools that they have gloomy assumptions built into them. Years ago I was asked to use and evaluate an online risk-calculation tool. It told me that I was at high risk for diabetes complications because I "wasn't taking my medications". The possibility that a diabetes patient might not be taking medications because his doctor hadn't found it necessary to prescribe any had not even occurred to the designers of the tool. (They were slightly apologetic when I complained about this, but they made it clear that they considered my case so freakish that their design need not take into account the possibility that anyone besides me might be maintaining normal glycemic control without using meds.)

Anyway: after slightly overstating my A1c results, just to get the app to work, here's the result I got:

Okay: the probability that I won't have to face those problems during the decade to come is 98% (which, considering that I will be 68 years old at the end of that decade, probably puts me a mile or two ahead of the average diabetes patient).

And even that low risk was calculated from an A1c result which was artificially inflated just to get the online app to believe I was a diabetes patient.

If you use this risk calculator, and the result you get doesn't make you happy, at least you know what you need to do: more exercise, less starch, and see if you can get yourself (over the long term) into a situation where the risk calculator gives you more reassuring answers.

 


1st Thursday Update

November 5, 2015

 


Ho-hum!

Just what we need: another study claiming that there is a connection "between strict control of blood sugar and increased mortality", because of hypoglycemic episodes causing cardiovascular problems.

What they really mean, and should have said, is that there is a connection between using drugs to achieve strict control of blood sugar and increased mortality due to cardiovascular problems.

I would venture to guess that my numbers today (80 mg/dl fasting, and only 111 post-prandial) were better than anything achieved by the people in their study, and I'm 14 years into un-medicated diabetes management, using exercise as a substitute for drugs. I don't have any kind of serious problem with hypoglycemia, and no heart attacks so far.

Having said that, I guess I'd better not have a heart attack now, and lose the argument!

 


Homeopathy, placebos, and the music of John Cage

This is the time of year when I start organizing musicians at my workplace to play Christmas tunes at the office, on the last working day before the holiday break. While preparing a message to them (to see who's participating this year, and schedule the rehearsals), I went looking for a Christmas-themed cartoon to use as an illustration. I found this:

In case you don't get the point of this cartoon, let me explain. John Cage (1912-1992) was an experimental composer -- the sort of artist who thinks an artist's job is to "raise questions" about what is or isn't art. Cage's most famous work (entitled 4'33") consisted of a pianist sitting down at the keyboard... and then not playing the instrument, for 4 minutes and 33 seconds. The score is beautiful in its simplicity.

You see, the "music" consists of the ambient sounds in the concert hall, including the rain on the roof, the rustle of bodies shifting in upholstered chairs, and the murmurs of puzzled audience members asking each other why the pianist isn't playing. (It doesn't actually have to be a piano, by the way, although a piano was used at the premiere. Cage authorized any instrument or ensemble to play the piece, provided they don't actually play.)

Apparently 4'33" has been performed (and even recorded) by many musicians, including Frank Zappa. Well, you could argue that there is nobody who hasn't performed it.

If you've ever sat quietly for 4 minutes and 33 seconds, seemingly you, too, have performed the piece. However, I'm not sure it counts as a performance unless you have an unplayed musical instrument in front of you. And maybe the disappointed audience is an essential ingredient in the performance, too: does 4 minutes and 33 seconds of not playing an instrument qualify as a performance of 4'33", if nobody is being irritated by it? I guess that's one of "questions" that the piece "raises". Here are some others that occur to me:

  • If 4'33" is performed in the forest and no one is there to be unable to hear it, does it still make no sound?
  • Is amplification allowed, or is an authentic performance necessarily an acoustic performance?
  • Is it allowable for non-musicians to perform the piece, or is it necessary that we know they could play, so that it seems significant when they don't?
  • Are performers permitted to create polyphony, by starting three performances a few seconds apart (as in "singing in the round")?
  • Was Cage tolerant of performers who took liberties with the tempo of the piece, perhaps stretching it out to a leisurely 5'12", or whipping it up to a frantic 2'37"?
  • Does the piece "travel" well, or is it the sort of thing that goes over big in New York but flops in Chicago?
  • Are there distinct regional preferences in performances of the piece -- do some people complain that German performances are too analytical, and British ones too sentimental?
  • Does the Cage estate claim copyright ownership of the piece, and if so, how vigorously is the copyright enforced? Is the copyright infringed whenever any musician spends 4 minutes and 33 seconds not playing?
  • Has Woody Allen's repeated appropriation of the piece (in silent opening-title sequences for his movies) ever been litigated in court, or has this been handled quietly in out-of-court settlements?
  • Have recordings of the piece been improved or degraded by recording techniques which eliminate tape hiss, such as Dolby noise reduction and digitizing?

By the way, Cage eventually gave 4'33" an even more abstract sequel, called 0'0". I know it sounds like I'm making that up, but I'm not. If you ask me, John Cage is an artist for people who think Andy Warhol worked too hard. (Warhol signed his name to all sorts of prints and objects not made by himself, but at least the people who bought them had something tangible to take home.)

You might think there would be no equivalent to John Cage's music in medicine, but there is in fact a popular form of medicine which takes the same approach to patient care as Cage took to musical performance: the practitioner goes through the motions of administering a medication, but doesn't.

This practice is known as homeopathy, and it involves treating the patient with a drug from which the drug has been removed. That is, you start with a drug, dilute it in a great deal of water, dilute the result with still more water, and keep repeating the dilution over and over again -- usually dozens of times -- until what you have is pure water (usually with a generous addition of alcohol, as a "preservative").

Why all the dilutions? (I am going to use the word dilution, even though homeopaths dislike the term -- they like to call dilution "potentization", and they refer to the agitation of the samples as "succussion".) Anyway, why do this? Well, it's a long story.

Homeopathy was invented by Samuel Hahnemann, a German doctor (or ex-doctor) who had given up his practice in 1784, frustrated by the primitive medical science of the time (blood-letting and the like). I can understand why he wasn't satisfied with the state of medical knowledge in 1784, but his solution to the problem was to start making stuff up.

Well, actually, he thought he was following the scientific method. He had read that quinine, a malaria drug derived from the bark of the cinchona tree, produced malaria-like symptoms in healthy people. He tried it on himself, experiencing symptoms which seemed sufficiently malaria-like. (To put this amazing coincidence in perspective, bear in mind that malaria causes the same "flu-like" symptoms as most infectious illnesses do: fever, chills, fatigue, nausea, and so on.) Anyway, Hahnemann decided on this very flimsy basis that any disease can be treated by substances which produce the symptoms of that disease. However, he thought the substance would be an effective treatment only if it was diluted enough that it didn't produce those symptoms. In short: the way to find an effective cure for any disease is to find a toxic substance which produces symptoms of that disease, and then dilute it to the point that it produces no symptoms. He began publishing essays on this system of therapy in the 1790s, but didn't start calling it homeopathy until 1807. There is no indication that Hahnemann had any basis in evidence for these ideas; he simply decided that they sounded right to him, and published them.

Once Hahnemann got the dilution idea into his head, he went a little crazy with it, recommending levels of dilution which made it pretty much impossible that even one molecule of the original substance remained in the water. In a way, that was good, since some of the substances used in early homeopathic medicine were poisonous or disgusting. But what was Hahnemann thinking? Apparently he thought that water could somehow "remember" the characteristics of a substance formerly dissolved in it, so the dilutions got rid of the toxic aspect of the substance but the water retained the therapeutic aspect of it. Modern homeopathy holds that water molecules form clusters replicating the shape of the (now absent) active ingredient; all those dilutions (and the "succussion" of the containers, which involves banging them with a stick in just the right way and just the right number of times) somehow trains the water to memorize the shapes of the molecules of the active ingredient, and retain those shapes even after dilution has eliminated the active ingredient entirely.

After all this preparation is complete, the water is etched with a persistent chemical image of the absent drug (made more permanent by the alcohol preservative). This trained water is now more effective than the drug itself would be. (It is a principle of homeopathy that the more you dilute a medicine, the more effective it is -- thus the old joke about somebody overdosing on his homeopathy medication by forgetting to take it.) If this concept strikes you as over-imaginative, I disagree: it is not imaginative enough. An imaginative person would think hard enough about it to realize it's silly.

Water is a rather well-studied compound. When it isn't frozen, it consists of molecules that are moving freely, not locked into a crystalline matrix; therefore, the water molecules can't preserve drug-shaped gaps amongst themselves. This seems to rule out any possibility of water acting as if it "remembered" what it used to be mixed with -- which is probably just as well! The hydrogen and oxygen atoms water is made from are essentially permanent, and over the billions of years they have existed on earth, they have been mixed with everything. Would you really want the water you drink to be retain features of every substance it has ever been exposed to? I wouldn't. I don't know where that water's been. Or rather, I do know where that water's been, but it doesn't bear thinking about.

The pure water which falls to earth as rain all found its way into the sky in the first place through the evaporation process, and it came from every possible source -- not just from the surface of alpine lakes. Think of every disgusting substance you'd never want to touch: if it's capable of drying out, then it has contributed water to the atmosphere, and after a while the atmosphere gives it back. You'd better hope the water you drink doesn't retain a lasting influence from what it's been mixed with. Atoms and molecules are so tiny, so numerous, so persistent, and so mobile that it's a little creepy to do the math and figure out how easily they connect us to things distant in space and time. (Some of the atoms you inhaled today were farted into the atmosphere by Emperor Nero; would you want them to be retaining the influence of that?)

Anyway, the idea that the molecular profile of a drug can be stamped onto water, and the water used as a more effective version of that drug, is pretty ridiculous. And even if we decide to overlook the implausibility of the theory involved, there is the slight problem that homeopathic remedies don't work better than a placebo. (Which makes sense, given that a homeopathic remedy is a placebo.) Personal testimonials from people who say homeopathic remedies helped them are just further confirmation of the placebo effect: thinking you're doing something about feeling bad makes you feel better.

Of course, from the point of view of anyone conducting a clinical trial, the placebo effect is only an obstacle -- an irritating distraction. And most people think of it as a purely mental phenomenon -- a case of people fooling themselves into thinking they feel a little better, when they really don't, simply because they want to think they feel better. It's not that simple, though: the mind affects the body, and the placebo effect can sometimes produce improvements that are objectively measurable. Feeling better mentally sets things in motion within the body that have physical effects. So, there is a bit of an ethical dilemma here. Since we know that the placebo effect exists, and that it can actually be beneficial, is it okay to "treat" people with a fake cure which only helps them because they assume it will?

I guess in some situations that's okay, but there are a few requirements this placebo-based "treatment" would have to meet before I'd say it passes the sniff test:

If you want to sell grass clippings as "Malaysian Dragon Fern" at an affordable price, and claim that this wonderful ancient remedy reduces the duration of a cold, I guess I'm not going to be too upset with you. But if you sell it at an outrageous price and claim it shrinks tumors, then you really need to die soon.

Unfortunately, I'm not in a position to benefit personally by the placebo effect -- it doesn't work for people who know how it works. I feel about the placebo effect the way Arthur C. Clarke felt about horoscopes: when an interviewer asked him if he believed in astrology, he said that he couldn't, "because I'm a Saggitarius and we're skeptical".

 

 


Older Posts:

September 2015
August 2015
July 2015
June 2015
May 2015
April 2015
March 2015
February 2015
January 2015
December 2014
November 2014
October 2014
September 2014
August 2014
July 2014
June 2014
May 2014
April 2014
March 2014
February 2014
January 2014
December 2013
November 2013
October 2013
September 2013
August 2013
July 2013
June 2013
May 2013
April 2013
March 2013
February 2013
January 2013
December 2012
November 2012
October 2012
September 2012
July 2012
June 2012
May 2012
April 2012
March 2012
February 2012
January 2012
December 2011
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