Map of life expectancy at birth from Global Education Project.

Monday, September 26, 2016

Social Psychology vs. Parapsychology

Hard to say nowadays which is less credible as science. Here's a lengthy post by Andrew Gelman about the so-called "replication crisis," which is a fancy way of saying that the entire field of social psychology is looking like mostly bunkum.

The defensiveness of its practitioners is to be expected, but let's check our own walls for glass before throwing stones. The main problems in the field of social psychology are not ones to which other social sciences or for that matter biomedical research are immune. It's a bit hard to explain if you haven't taken much in the way of statistics or research methods generally, but the keystone issue is the worship and misunderstanding of the concept of "statistical significance."

If I compare two samples from a given population, with an equal (or at least known) probability of being selected at random, and some with characteristic A are more likely to have characteristic X than are those without A, I want to know how likely this is to just be a coincidence. If the observed difference is expected to occur less than 5% of the time when there isn't really a difference in the total population, we say the "p value" is less than .05 and we declare the observation "statistically significant" which is presumed to be more or less synonymous with "true." If the probability is 6% we declare the observation "not statistically significant" which is presumed to be synonymous with "false."

This is so wrong for so many reasons it makes one feel foolish to point them out. One is that the p value depends on sample size as much as it does on the magnitude of the effect. If my sample is too small, I will be likely to get an insignificant p value even if a meaningfully large effect exists. If the sample is large, I will likely get a "significant" p value for an meaninglessly small effect. A bigger problem is that if I make multiple comparisons I will likely find a "significant" value in there just by chance, because you have to multiply the values by the number of comparisons. Cherry picking the ones that are "significant" is basically fraudulent, although it seems most people who do it don't know that.

Other problems are that in social psychology, dependent variables are typically quite subjectively measured and it may be difficult to detect observer bias; independent variables may be associated with other, unmeasured variables that are actually responsible for any effect; there are all sorts of rationales for excluding cases selectively after the fact; and samples are rarely representative of any broader group than (quite typically) undergraduates at a selective university from which they are drawn -- who, by the way, are quite likely to divine the research question and consciously or unconsciously alter their behavior in response.

Gellman points to all sorts of other design flaws but the overall lesson is that it's just too easy to find what you are looking for. These studies get a lot of press because they seem relatable and often directly relevant to our own lives and supposed behavioral predispositions and those of the people around us. But they're largely gahrbahzh. So sad.

Thursday, September 22, 2016

Too many emergencies

Yes, the Trump Foundation and the border wall and birtherism and all of the dreck that spews from Ronald T. Dump should get news coverage.

However, there is shit happening in the world that is like, really, really important and probably ought to be discussed by the candidates. You know, that little climate change problem, nucular weapons, mass extinction . . .

There is also this. When antibiotics stop working, it's not just those poor dusky-hued people in distant lands currently dying of tuberculosis who will be shit out of luck. It's you. Not only might you die from an infected scratch on your hand, common surgical procedures will be far too risky, women will die in childbirth and children will die from strep throats. Gonorrhea and syphilis will be incurable.

Of course, avoiding this fate requires paying a few cents more a pound for pork, beef and chicken, so we can't possibly afford it.

What are the chances it will come up in the debate on Monday?

Monday, September 19, 2016

San Quintin

Many years ago, I visited the small town of San Quintin in Baja California Norte with my then-girlfriend. There is, or was, a small resort there catering to United Statesian surfer bums and retirees. Pensioners would live there because it's cheap and the weather is always sunny and warm. They never bothered to learn Spanish and would order the staff around arrogantly. (Of course.)

We visited the town which consisted of a few cinder block shacks, a Pemex station, and a fly-infested general store. The people had a few wilting cabbages and starving cows. One guy owned the whole place, he had a farm down the road where the women and old men who hadn't gone north labored in the dust to fill trucks with tomatoes for the journey up the peninsular highway. He had all the water rights so the peasants couldn't irrigate their pathetic gardens. There was a big field with a fence around it and a sign saying "no cazar." (No hunting.) There was a guy sitting on a horse with a rifle to make sure nobody did. Mostly though there was nothing to hunt. More than a mile or so from the ocean was high desert, just coarse sand and the meanest cacti you ever saw.

Oh yeah. The farm had pumped out so much groundwater that salt water was intruding into the aquifer and the water in the resort was spoiled, so they had to use bottled water. It hardly ever rains there. It looks like that's about to change:

The storm is taking dead aim at San Quintin. When we were there I found some old cattle bones sticking out of the sand. The people told me the beast had drowned in a storm some years ago. So, once every thirty years or so they get a tropical storm. I imagine the desert blooms.

Friday, September 16, 2016 (wonky)

Krugman does the "wonky" trigger warning on his blog so I figured I should too. A major problem with biomedical research is what's called "publication bias." This takes a couple of forms. One is that negative findings -- i.e. "A does not cause B" or "is not associated with B" -- are unlikely to be published. Journal editors and reviewers just don't think they're exciting. However, if there are three trials that show that A does not cause B and one that does, and the last one is the only one that gets published, we wind up with a false view of reality.

A second form is that -- oh, did I ever mention that drug companies are evil? They choose not to publish studies that are unfavorable to their products. Same result, we come to believe that drugs are much more effective than they really are.

Another problem is post hoc analysis. If your initial hypothesis isn't borne out, dredge through the data to find some sort of significant association, say with a sub-group or a variable that you originally intended as a covariate rather than an outcome. The problem with this is that the p values are spurious, because if you make a large number of comparisons some associations will appear significant just by chance, when nothing is really going on.

Recognizing these problems, the FDA requires that all clinical trials for drugs be registered in advance, so that a) we'll know what the original hypotheses and protocols were and b)  we'll know about trials that aren't published. The penalty for not reporting your results within a year is supposed to be $10,000 a day.

Surprise! It isn't happening, and the law is not being enforced. We know that results aren't being reported in many cases, but we don't know how many trials aren't even being registered in the first place. Things appear to be getting a bit better, but there are still a lot of drugs being prescribed that probably should not be. This is a huge scandal that is almost completely ignored while we obsess over Hillary's e-mails.

Monday, September 12, 2016

Science for Sale

As my 2 1/2 long time readers have probably noticed, nothing frosts my pumpkin more than corrupt science. Well the gourd has a thick layer of ice right now. Stanton Glantz - a major figure in exposing corruption of science by the tobacco industry, has turned his guns on the sugar industry. They too were paying scientists to say that sugar is good for you.

You may recall that we had a consensus for a couple of decades that the way to avoid heart disease was a low fat diet. It's taken a couple of decades more to eradicate that falsehood and it's still clinging to life. Food manufacturers touted their "low fat" products as healthful while they were full of toxic sugar. What we're finally getting around to understanding is that dietary fat -- other than transfats -- and dietary cholesterol do not cause atherosclerosis. But sugar does, and it also causes a glycemic spike in the blood which contributes to diabetes. Plus it makes you fat. The entire edifice of nutrition science from the late sixties right into the nineties was a fraud erected by the sugar industry and its scientific prostitutes in academia.

That's really evil.

Thursday, September 08, 2016

Celebrity Jeopardy

So one way I dispose of 1/2 hour of unneeded consciousness from time to time is by watching Jeopardy! For those of you who aren't familiar with the program, they have special tournaments for categories such as high school students, college students, teachers and what not. They also do a celebrity tournament every year which features people who for one reason or another are famous. As it turns out famous people tend not to be very smart so they make the questions really easy.

This year many of the players as it turns out are "journalists," among them Chuck Todd and Anderson Cooper. Apparently the only reason they agreed to go on the show is because of the Dunning-Kruger effect. As it turns out they are both astonishingly ignorant of the most basic facts of history (e.g., Cooper thought that Alexander the Great presided over the golden age of Athens, and also couldn't figure out that "tank" is a word that means both an aquarium and a war fighting machine) and incapable of the simplest deductive thought. They revealed themselves to be total airheads. Louis C.K., on the show with Jonathan Capehart and Kate Bolduan -- who are maybe a small notch above Cooper and Todd but not much when it comes to gray matter -- destroyed them both. I will venture to say that comedians in general are probably much smarter and better informed than journalists.

The way you get to be a talking head on TV is by being pretty and having good diction. Being an idiot who will read whatever drivel and propaganda your corporate overlords put in front of you is how you rise to the top and make millions of dollars.

Tuesday, September 06, 2016

Rapid Testing

Reading between the lines -- and not very far between them -- it appears the whole Theranos corporation thing was never anything but a scam, although as with most cons it is likely that CEO Elizabeth Holmes on some level believed her own bullshit.

For those who don't know, this was a silicon valley start up that promised to do a whole suite of medical tests from a drop or two of blood. No more getting the needle and giving up vials. The company attracted billions in venture capital and Holmes was briefly recognized as the world's wealthiest female entrepreneur. Turns out, the technology doesn't work and the billions have melted away like the snows of March.

What you may not know is that this probably didn't seem preposterous to investors because in fact you can get some reliable test results from a fingerstick, in just a few minutes. Many large practice now have on-site rapid testing for blood lipids -- i.e. cholesterol -- and what is called the HbA1c,* which is an indicator of what your blood sugar has been over the past few weeks and is the standard for monitoring diabetes control.

When practices have these things, it makes diabetes care better, easier and cheaper. You come in for your visit, the medical assistant pricks your finger and gets a drop of blood for the HbA1c and a small pipette for the lipid test. Then she (yeah, usually it's she, that's the real world) goes down the hall, sticks them in a machine, and in a few minutes has the numbers to give your doc before the visit. The alternative is for the doctor to give you a test order and for you to make a second visit to a lab, in which case the doctor won't have your test until after the visit when obviously it's much less useful. Plus which you might not bother to get the test at all.

Unfortunately, other than simple blood glucose that's all we've got right now. Other tests require more blood and can't all be run by one machine either. It is probably impossible even in principle. But what the Theranos story tells us is that with sufficient audacity and a convincing act, even smart rich people can be conned. Viz. Bernie Madoff and Ken Lay, among too many others to name.

* Stands for hemoglobin A1c, and the way it works is that the glucose in your blood gets attached to the hemoglobin in your red blood cells. The more glucose, the more of it is attached. This is also called "glycolated hemoglobin."

Tuesday, August 30, 2016


Clostridium dificile is a bacterium which is an opportunistic pathogen in humans. Specifically, it infects the intestines and causes severe diarrhea, which can be fatal but is always indescribably unpleasant. Much C. dificile is now antibiotic  resistant as well.

For the medical industry, additional bad news is that it is mostly iatrogenic. While community acquisition occurs, the most likely way to get C. dficile disease is to be in the hospital, wherein it lurks, and to receive a massive antibiotic bomb that wipes out our normal intestinal microbia. C. dificile then moves into the abandoned territory.

What to do? How about restore the intestinal microbia. How to do that? Well, don't think about it too hard. What you need is a shit enema. So that's what they are doing, and it works. Adding to the unpleasantness, if you have recurrent infection it works much better to use donor poop than your own, which obviously has not been up to the task. (You may only be able to read the first paragraph.)

I note this not to gross you out, particularly, but because it points as dramatically as possible to our evolving understanding of ourselves not simply as clonal colonies of eukaryotic cells derived from a single zygote, but as ecosystems including not just those, but also a much larger number of prokaryotic organisms. Our eukaryotic genome has co-evolved with the rest of the ecosystem such that it supports beneficial  symbionts that not only are essential to digesting our food but also protect us from unfriendly organisms and quite possibly do other stuff we don't yet understand.

That doesn't mean that the stuff you buy at the GNC that purports to be probiotic will actually do you any good. We're still just beginning to figure this out and  come up with evidence based interventions. But it will become a bigger deal in coming years, I hereby predict.

Thursday, August 25, 2016

Have I ever mentioned that drug companies are evil?

They aren't really of course -- corporations aren't people nor are they moral agents. But human beings do act on their behalf. With price gouging by pharmaceutical manufacturers all of a sudden in the news (despite it has been going on forever), some of our friends from Hahvahd, aka World's Greatest University, explain some of the ins and outs of drug pricing wonkery (and wankery).

Some of the main points are:

  • The U.S. grants long periods of monopoly over new drugs, and let's companies use various loopholes and fancy tricks to extend their monopolies.
  • Even when they lose legal monopolies, they often have practical monopolies for various structural reasons in the industry. The simplest is that if you are in a competitive market for generic drugs, there isn't much profit there in many cases, so you don't get a lot of competition.
  • Medicare is not allowed to negotiate drug prices, and drugs can't be imported, even though they're cheaper everywhere else in the world.
The companies claim they need to haul in the big bucks to pay for R&D, but that's a total crock. They conclude:

High drug prices are the result of the approach the United States has taken to granting government-protected monopolies to drug manufacturers, combined with coverage requirements imposed on government-funded drug benefits. The most realistic short-term strategies to address high prices include enforcing more stringent requirements for the award and extension of exclusivity rights; enhancing competition by ensuring timely generic drug availability; providing greater opportunities for meaningful price negotiation by governmental payers; generating more evidence about comparative cost-effectiveness of therapeutic alternatives; and more effectively educating patients, prescribers, payers, and policy makers about these choices.
The problem is, of course, that the companies own congress, so none of this is going to happen.

Monday, August 22, 2016

The Party of Ignorance

A United States senator says that there is no need for college professors, students can just watch documentaries. He specifically proposes watching the Ken Burns documentary instead of paying some egghead to teach Civil War history.

By the way, this clown has a college degree.

Burns is a Republican, He is not from the South, but I presume he must know that many of his colleagues and their constituents actually wouldn't like that proposal because they think the Civil War was actually the War of Northern Aggression and it was all about state's rights and the preservation of heritage and tradition. How do universities deal with these differences in point of view?

Unless they are Liberty or Bob Jones university, we expose people to basic facts, often in the form of a textbook; and to various interpretations and viewpoints on those facts through diverse reading assignments. We then encourage them to discuss and debate the issues, along the way teaching critical thinking skills and the technique and etiquette of adversarial dialogue. Then we ask them to express their own ideas and conclusions through written essays, which we evaluate and critique.

In other words, we don't just stuff them full of single viewpoints on complex issues. We develop the capacity to learn and think for oneself. A capacity sadly lacking in Ron Johnson.

Thursday, August 18, 2016

The Chrono-Synclastic Infundibulum

Vonnegut fans will recognize the title of this post as a singularity in the space-time continuum where all possible opinions are true. That is perhaps not quite the same as saying that there are no facts. Peter Pomerantsev, in Granta, finds that we are living in a post-factual society, a development for which he blames a combination of mendacious politicians, Your Intertubes, and post-modern philosophers.

I don't think our situation is as yet quite that bad. In Putin's Russia, it probably is, but that's because he controls all of the mass media and can maintain a fact-free environment. Pomerantsev invokes the counterfactual, emotion-driven Brexit campaign, but as we know the Brits woke up from their dream the following morning to find themselves doused in the ice water of reality. Karl Rove mocked the Reality Based Community but ultimately, he discovered that he could not create the new realities of his imagination: the Weapons of Mass Destruction™ weren't there, and Iraq did not become a bourgeois republic allied with the West. Donald Trump's fans are impervious to facts but they will discover on November 9 that they are in the minority and their hero is a loser.

I do share Pomerantsev's contempt for post-modernism and it's bizarre claim that truth is the servant of power, that scientific claims are arbitrary, and that denial of intersubjective reality is liberating. And I am as frustrated as anyone by the power of confirmation bias and motivated reasoning, which has indeed been amplified by the instant availability of whatever congenial bullshit you are looking for at the click of a mouse.

Nevertheless, sitting here in a university, I know that we still have the tools to recognize truth and a large corps of people who live by them. We may go down the infundibulum soon, but not today. 

Wednesday, August 17, 2016

Fat Al told you so

With California burning  and Louisiana drowning, NASA informs us that we've just lived through the hottest month in recorded history. If you go to the last link, to Joe Romm's blog, you'll see a graphic that ought to terrify you.

What appears to be happening is a long-anticipated step up in global temperature -- with 2016 landing close to .2 degrees centigrade hotter than last year, which was already the all-time hottest year by a lot. The Middle East is literally becoming uninhabitable -- Baghdad has had temperatures above 109 Fahrenheit every day since June 19, and other places in the region are even hotter. In case you didn't know, the collapse of agriculture in Syria due to climate change was the main reason why Syrian society disintegrated, leading to the greatest humanitarian catastrophe since WWII.

One of the presidential candidates says this is not actually happening, it's a hoax perpetrated by the Chinese. The other one never mentions it. It is the single most important thing that is happening on earth.

Thursday, August 11, 2016

Yep, they're still evil

That would be the drug companies. Boston University School of Public Health prof. Alan Sager (with whom I am acquainted) runs down their profiteering at society's expense. I would say, do read the whole thing because it takes him 11 pages to properly debunk all of the prevalent falsehoods about this. I'll just offer you this excerpt which reiterates stuff I keep saying here, which is that the mythological and non-existent "free market" bears absolutely no resemblance to the reality of pharmaceutical development, manufacture and selling:

Unfortunately, functioning free markets are simply unattainable in health care, and the drug sector is no exception. That’s because not one of the six requirements for a competitive free market is met, or can be met, in the realm of pharmaceuticals. A market of small buyers and sellers doesn’t make the price; drug makers with patents or market power are dominant. Price competition among generics, biosimilars, and me-too brand name drugs does little to cut U.S. drug spending. Sovereign consumers don’t make decisions; they rely on physicians who are often swayed by drug makers’ marketing or detailers. Entry of new competitors can be hard; drug makers merge with competitors or acquire them to reduce competition; big vertically-integrated drug makers often finance smaller ones or buy up their discoveries. Information is asymmetric; patients lack it and doctors and drug makers have lots of it. Subtly, but importantly, the price of drugs doesn’t remotely track the cost of production. And the injunction that buyers should beware and mistrustful carries little weight in the absence of good information about which drugs are needed or valuable.
There are actually additional reasons why the fictitious "free market" doesn't come near this reality, but those are probably enough for now. As Sager explains, it is the immense political power of the drug companies, along with our generally brainwashed population, that let's them keep ripping us off.

Sager is an advocate of single payer national health care, but given that isn't happening any time soon, he offers some interim measures. Not on Hillary's docket, however, at least not so far.

Tuesday, August 09, 2016

Down the Rabbit Hole

John Ioannidis is what I would call a science critic. I don't mean that he's anti-science. On the contrary. He's a science critic in the same sense that a movie critic or a literary critic likes movies or books, but has something to say about which ones are better than others, and how they are composed and achieve or fail to achieve their artistic objectives.

He joins colleagues in critiquing the Big Science initiatives of NIH, which has been directing more than half of its funding to genomic and stem cell research. Contrary to early expectations, it turns out that the relationship between the genome and specific diseases is far too complex to provide targets for prevention or treatment. Even when a specific genetic cause of a disease is known, as in sickle cell anemia, the knowledge doesn't necessarily translate into a therapeutic target.

Similarly (and relevant to the cancer "moon shot,") the extremely diverse mutations associated with cancer, and the adaptive nature of tumors (they evolve in response to selection pressure from treatments), mean that understanding the specific mutations associated with cancers is of limited value for treatment. Stem cells have also so far failed to yield any approved treatments.

The writers call for a redirection of NIH funding away from these "big ideas," and to more strictly evaluate proposals in these areas. The challenge is that right now, these fields are academically prestigious, and yield high impact publications, promotions and tenure, and awards. Furthermore, scientists get pigeonholed in their narrow programs of research, and proposal review committees generally don't like it when they try to step out in a new direction. So there are vested interests which are difficult to overcome.

I actually think that there is more to be gained by what they call translational science -- getting doctors to do what we already know is best for patients -- and investment in social determinants of health, than by the "blue sky" biomedical research Ioannidis and friends call for. Sure, we should do some of that, but we can save lives and improve the health of billions of people just by putting what we already know to work. But that isn't very glamorous.

Thursday, August 04, 2016

Assault and battery for money

NYT's Gina Kolata tells you what all of us health services researchers already know far too well. Surgical procedures don't have to be approved by the FDA or anybody else, and even when they are proven to be useless, surgeons keep doing them. She leads with the most notorious example, spinal fusion.

In fact this goes way back. In the 1990s, what was then called the Agency for Health Care Policy and Research concluded that there was no evidence that spinal fusion was beneficial to patients. So, the spinal surgeons society persuaded the Republican congress to eliminate the agency entirely. Quoth:

AHCPR was also confronted in 1995 with an advocacy organization’s active efforts to get it defunded. The source was an association of back surgeons who disagreed with conclusions reached by the [agency] on low-back pain and with practice guidelines based on that work. . . .

The surgeons found sympathetic ears among House Republicans who, for reasons already discussed, were prepared to believe the worst about the agency. The events of 1995 followed many years of controversy over the merits of surgical procedures for low-back disorders. AHCPR entered this fray when its PORT on low-back pain reviewed the research and concluded that there was no evidence to support spinal fusion surgery and that such surgery commonly had complications.30 The North American Spine Society (NASS) created an ad hoc committee, which attacked the literature review and the subsequent AHCPR practice guideline on acute care of low-back pain. In a letter published in 1994 in the journal Spine, the committee not only criticized the methods used in the literature review and expressed concern that the conclusions might be used by payers or regulators to limit the number and types of spinal fusion procedures, but it also charged that AHCPR had wasted taxpayer dollars on the study.31 An entity known as the Center for Patient Advocacy was formed by Neil Kahanovitz, a back surgeon from Arlington, Virginia, to lobby on the issue. It organized a letter-writing campaign to gain congressional support for its attack on AHCPR. Kahanovitz used personal contacts to gain the support of Representatives Bonilla, who had a staffer who was Kahanovitz’s patient; Johnson; Gerald Solomon (R-NY); and Joe Barton (R-TX). Solomon, Bonilla, and Johnson led the effort in the House to end the agency’s funding, energetically promulgating the NASS/Kahanovitz argument that it was supporting unsound research and wasting the taxpayers’ money.
The agency managed to survive, but it endured a 21% budget cut. (It's successor is now known as the Agency for Healthcare Research and Quality.)

Here we are, 20 years later, and surgeons are still doing this procedure, even though, as Kolata reports, randomized controlled trials since then have consistently found no benefit. It's finally stopping now that insurers have decided not to pay for it.

Similar phenomena happen all the time. The hue and cry when the Preventive Services Task Force recommended against routine mammograms for women in their 40s was orchestrated by radiologists, oncologists and surgeons who feared losing business. (Debbie Wasserman Shultz, bless her, was their willing dupe.)  We are finally seeing a movement in medicine to eliminate useless or harmful procedures, as spearheaded by the Choosing Wisely campaign lead by the American Board of Internal Medicine and joined by many other medical specialty societies. Still, physicians' financial incentives work against the goal, and we really need more than voluntarism.