Obey

For your daily dose of surreality, be sure to check out the thread Is Barack Obama the Antichrist? thread over at apostropher. This thread now has a logo. To understand the title of this post and the logo, check out this entry at Wikipedia. And once you’ve done all that, congratulate yourself on a good waste of 5 minutes or so. Apparently there are some other threads over at the Mineshaft. And apparently this is the leading source of all Barack Obama antichrist news. You could kill a good hour, if you wanted.

Technorati Tags: ,

A short-lived pleasant surprise while reading Boortz

Along with my fast on news and pseudoskepticism I’ve left “Neal Boortz”:http://www.boortz.com/ alone for a while. I mean, once you get past the “Most overpaid and underrated talk show host” and “Somebody’s gotta say it” crap, the guy’s a real jerk, driven insane by 9/11 and the War in Iraq. So I wandered over today[1] because, I just guess I felt like I wanted to get angry. I wasn’t disappointed, but before I got the blood temperature up I did get a little pleasant surprise.

His first entry was on “Dr. Eric Keroack”:http://boortz.com/nuze/200611/11272006.html#births, the space cadet that Bush has just appointed to deputy assistant secretary of population affairs of HHS(Department of Health and Human Services). This guy is facing a situation where a significant number of babies are born to single mothers, and he makes the following assertions:

# Condoms offer no real protection from herpes or HIV.
# Fifteen out of 100 people who have sex (outside of marriage, I presume) will get AIDS.
# The use of birth control is demeaning to women and degrading of human sexuality.
# A woman who has an abortion will be eight times more likely to get breast cancer before she’s 45.
# If you have premarital sex you won’t be able to bond with your wife or husband and your marriage will most likely be a bust.

And then Boortz made the astute observation that Bush felt his loss was do to the fact that he alienated conservatives (never said he isn’t intelligent, he’s just a jerk).

But then we get to the “I hate liberals” diatribes: complaining about minimum wages, Social security, paranoid pronouncements that Democrats are going to destroy talk radio, basic rehashings of his old paranoid rantings.

And then we get to the Iraq war. This from his “whining about Chuck Hagel’s column”:http://boortz.com/nuze/200611/11272006.html#iraq:

The promotion of any and all bad news. Something bad happens in Baghdad? It’s front page news. A single car bomb goes off? There’s nothing more important. All news from Iraq must be bad to push the anti-war, anti-Bush Administration template.

The majority of Americans finally recognize that Iraq isn’t just “one or two car bombs going off in Baghdad”:http://www.apostropher.com/blog/archives/003534.html. In fact, I don’t think there’s much controversy in calling it a civil war over there, except for those who have PR reasons for choosing other words.

He finishes up by complaining about “Yoko Ono”:http://boortz.com/nuze/200611/11272006.html#yoko (I don’t know about you, but a day of healing sounds pretty nice — we have plenty of days about war, dontcha think?), a woman who “wrote her dissertation”:http://boortz.com/nuze/200611/11272006.html#fat on overweight lesbians, and a finale: “If your daughter smokes, she’s having sex.”

Well, that’s about all the hot air I can take for a couple of months from him. It’s back to reading the clips from media matters.

fn1. Reading Boortz.com is not right for every one. If you have hypertension, heart problems, are overweight, underweight, or even the correct weight, if you have liver problems or a normal liver, then ask your doctor before reading. If you experience fever, seizures, uncontrollable temper, uncontrollable shaking of your head, clenching your fist until it bleeds, intermittent screaming of expletives, or just primal screaming, then stop reading Boortz, reach for the nearest bottle of beer, drink it down, and then call your doctor. Serious side effects may include running down the street screaming, bruised vocal cords from the screaming, bleeding and bruising in the chest, and people looking at you real funny.

Technorati Tags:

Pseudoskepticism

It’s been a while since I’ve tangled with the skeptics, and for good reason. I’ve pretty much stopped reading anything at “science”blogs.com or anyone that advertises the skeptics circle.

However, I did come across “this article”:http://en.wikipedia.org/wiki/Pseudoskepticism on pseudoskepticism, and pretty much points out everything that bothers me about these blogs I’ve kicked from my periodic reading list. Not listed in the bullet points is that trying to carry on a reasonable, logical argument against a pseudoskeptic is like trying to convince a brick wall to back away so it doesn’t hurt so bad when you bang your head.

Maybe one day I’ll write about my good ol’ nemesis Orac again. After all, every once in a while it is fun to give a thrashing to a pseudoskeptic who pretends to believe that science disproves things he doesn’t believe in.

In the meantime, I will confess an interest in a lot of fields that may be better described as “protoscience”:http://en.wikipedia.org/wiki/Protoscience, or fields that strive to remain coherent within the scientific framework, but which have not quite yet achieved experimental verifiability according to the principles of repeatability and replicability. I think we have a lot to learn from such fields, and they tend to stretch our understanding about reality and how we learn about reality.

Technorati Tags:

It’s beginning to look a lot like Christmas

In years past, I’ve pretty much said “Bah Humbug!” to Christmas until well into December. I always avoided the crowds on Black Friday, and wouldn’t get a tree until the second or third week of December.

Right now, the tree is up and lighted, and the outside decorations are up. A wreath made from real pine decorates our door. We’ve even had Bing Crosby’s _White Christmas_ going at one point. Christmas carols fill our bedtime routine. There are still some ornaments to put on the trees, and the garland remains to be put up.

This, folks, is what having kids does. Beware!

Law of the iterated logarithm, and a reminder on the limitations of statistics

We trust confidence intervals, and they are reported with just about every study in existence. However, they are often wrong, not just because they are wrong 5% of the time but also because they are often incorrectly computed.

In doing applied statistics day after day, it’s difficult to remember all of the theoretical statistics I learned in graduate school. (I got my degree at UNC-Chapel Hill, which is known for its theoretical statistics program.) So I brought home my copy of Van der Vaart’s “??Asymptotic Statistics??”:http://www.amazon.com/Asymptotic-Statistics-Statistical-Probabilistic-Mathematics/dp/0521784506/sr=1-1/qid=1164382703/ref=pd_bbs_sr_1/002-3849726-6409614?ie=UTF8&s=books. It’s a very nice combination of applied and theoretical statistics, with lots of examples and reminders about why the statistical methods we use in the applied world work the way they do. With this study came a stark reminder about some of the dangers in overinterpreting the results of statistical methods.

There’s this very odd looking result called the law of the iterated logarithm: lim supn→∞ (Y1 + … + Yn)/√(n log log n) = √2 for a sequence of random variables Y1, … with mean 0 and variance 1.

So what does this mean. lim sup means maximum “in the limit,” i.e. what is the maximum value of that strange-looking expression if you ignore the first 10, 100, or even 1,000,000 sums. Now, what about that that strange looking expression? It’s the average of Y1, …, Yn, multiplied by √(n/log log n). If you’ve happened to have gone through a few statistics classes and an advanced calculus class or two, you might recognize that the average goes to 0 (because that’s the mean of all the Yn) so “in the limit” that strange-looking expression is an indeterminate 0×∞. It converges to 0, but infinitely often it’s close to √2.

This is a result that applied statistics don’t use very often, but it still has important implications. I won’t go into the full argument, but an example illustrates how this means that confidence intervals are wrong.

Let’s say that we are measuring the average height in a population (and let’s say there are an infinite number of people). We measure 100 people and construct a 95% confidence interval AVE-2/√n – AVE+2/√n (I use 2 to make this simple, though statisticians might rather see 1.965.) And then say we do a sequential testing scheme: we add another person into the study and redo the average and confidence interval. Then add another person and redo the average and confidence interval again.

The confidence interval covers the true average height only if (AVE-true mean)*√(n/log log n) < 2/√log log n. But because of the law of the iterated logarithm, this fails infinitely often in our sequential testing scheme above.

Two take home lessons:
# 95% confidence intervals are wrong 5% of the time, and refining them to make them smaller doesn’t change that fact
# approaching sequential testing in a naive way leads, infinitely often, to confidence intervals that are wrong

Technorati Tags:

Zinc – “Clinically proven”

Whether stated in an ad for conventional or alternative medicine, I typically take claims of “clinically proven” with a grain of salt. That’s because the statistical methodology used to “prove” these claims basically says “if we assume these claims aren’t true, then the results we have seen in studies would have been too bizarre.” This is even though advertisement language is regulated by the DSHEA(Dietary Supplement Health Education Act?), Food, Drug and Cosmetic Act, the Code of Federal Regulations Title 21 Part 101, and so forth. So, when I saw the words “clinically proven to cut your cold nearly in half” by the Cold-Eeze® product manufactured by the “Quigley Corporation”:http://www.quigleyco.com, I naturally got a bit curious.

Cold-Eeze is the homeopathic remedy Zincum Gluconicum along with inactive ingredients. The form I saw was a lozenge, and the dilution was 2X (i.e. a factor of 10^2^=100). The back of the box makes the following claim:

Two clinical studies have shown: Cold-Eeze proprietary formula reduces the duration and severity of cols by 42% or 3 to 4 days.

The independent double blind studies were conducted at the Cleveland clinic and Dartmouth College and published in peer-reviewed journals.

The two articles are as follows:
* Mossad, _et al._ ??Annals of Internal Medicine??. *126*:2, July 15, 1996. (“PubMed”:http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=8678384&query_hl=1&itool=pubmed_docsum | “Full Text – free”:http://www.annals.org/cgi/content/full/125/2/81)
* Godfrey, _et al._ ??Journal of International Medicine Research??. *20*:3, June 1992. (“PubMed”:http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=1397668&query_hl=3&itool=pubmed_docsum)

I don’t know that much about ??JIMR(Journal of International Medicine Research)??, but the ??AIM(Annals of Internal Medicine)?? is certainly one of the top-tier publications. (This means little in my book, but hey, we’re talking about one of the topics that James Randi has staked his $1 million prize on from my understanding.)

So, the Mossad article in ??AIM(Annals of Internal Medicine)?? does show a well-designed and well-controlled study of zincum gluconicum in which the severity of symptoms was reduced from a median of 7.6 days in the placebo group to a median of 4.4 days in the zincum gluconicum group. The full text is available, and this study makes it look like this formulation does reduce duration of symptoms. Do read the discussion section of the article to see the limitations of the study. No mention was made of the method of preparation of the active ingredient (i.e. dilution), and the discussion of mechanism was phrased in terms of clinical pharmacology rather than homeopathy. This may reflect a bias in the journal or the authors, or Quigley may have produced non-homeopathic zinc lozenges for the study. It’s also worth noting that on the box the company states that the active ingredient is a homeopathic cold remedy, but I didn’t see a mention of the Homeopathic Pharmacopoeia of the US.

The Godfrey article is not in full text online, but from the abstract it looks like zinc took 1.2 days or 4.9 days off the duration of symptoms, depending on when therapy was started. The formulation and treatment schedules were not discussed.

So, it does look like the Cold-Eeze product performed pretty well in those two studies, if in fact is was a similar formulation. However, what about other studies? Some seem to be more negative on zinc:
* Eby, GA and Halcomb, WW. ??Altern Ther Health Med.?? 2006 Jan-Feb;12(1):34-8. “We found no reason to recommend intranasal zinc gluconate or zinc orotate lozenges in treating common colds.” The measured number of patients free of symptoms after 7 days, and 10/16 (63%) in the zinc group compared with 9/17 (53%) in the placebo group. I actually find their conclusions bizarre in light of their sample size and measure. If they had wanted to detect a 20% difference between placebo and zinc, the study would have had 20% power. It’s pretty awful to have an underpowered study and then claim no difference when you can’t reject the null hypothesis. Heck, if we were allowed to do that, I could make it look like penicillin was ineffective. So I’d take the numbers from that study as a bit of information, but ignore the conclusions.
* Wintergerst, ES, _et al._ ??Ann Nutr Metab.?? 2006;50(2):85-94. Epub 2005 Dec 21. This is a review of done on zinc and vitamin C, basically as nutrients. The article does note that adequate amounts of zinc and vitamin C seem to shorten duration of symptoms.
* Arroll, B. ??Respir Med. 2005 Dec;99(12):1477-84.?? This Cochrane database review notes that zinc does seem to have some efficacy, and might be useful.

There are many others. A “PubMed search”:http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?CMD=search&DB=pubmed of “zinc treatment cold” (no quotes) returns 192 articles. Sorting through these studies and accounting for differences in formulation, dosing schedule, actual dose, delivery, and other factors that greatly affect drug efficacy is dizzying and daunting at best. It’s also important to remember that zinc lozenges do carry the potential for adverse events, such as bad taste and nausea.

Also not discussed in these articles are the issues of homeopathic dilution of zinc compounds. This is going to be hard to find because of the bias of journals, and because I still haven’t seen evidence that zinc is in the HPUS.

Finally, the long-term consequences of the suppression of common cold symptoms has not been discussed. Zinc has no effect on virus-shedding, so presumably it doesn’t help the body dump the cause of cold any faster. Whether reducing the severity of symptoms is, in the long run, useful, has not been answered.

For me, the jury is out on zinc as a cold remedy. It seems to do something. Exactly what, I want to understand a little bit better.

Technorati Tags: , ,

Wired for statistics

Aleks over at the “Statistical Modeling, Causal Inference, and Social Science”:http://www.stat.columbia.edu/~cook/movabletype/archives/2006/11/is_the_brain_do.html
(put out by a bunch of social science-based Bayesian statisticians from Columbia University) wonders whether our brain is wired to do a Markov Chain Monte Carlo, which is a sophisticated simulation-based method of doing statistics. Apparently, our brain puts out activity that looks like noise before going into peak-state, and the idea is that we might be simulating various outcomes from prior information (i.e. MCMC) to make the best decision (technically “Bayes-optimal”).

Fascinating theory to entertain. I think that maybe we are doing some sort of MCMC or simulation unconsciously when we make decisions, but that may not be the whole story.

Technorati Tags: ,