Seth's Blog

• A good example of how misleading drug-company-sponsored analyses of drug trials can be. Independent reanalysis by Daniel Coyne, a professor of medicine at Washington University in St. Louis, reached opposite conclusions. Good work, Coyne.
• Coke contains a carcinogen.
• “I used sunflower seeds to lose weight.” Someone else used them to reduce addictions. The link between the Shangri-La Diet and reduction of non-food addictions (smoking, coffee) fascinates me. People start SLD to lose weight and say they become less addicted to smoking, coffee drinking, and so on. One possibility is that by reducing hunger, SLD reduces discomfort. Addictions gain strength from discomfort, often resemble self-medication.
• Steve McIntyre replies to Gavin Schmidt’s claim that McIntyre’s beliefs resemble “classic conspiracy theory”. I used to watch a lot of football — when the 49ers won most of their games. (I am a classic fairweather fan.) I get a similar pleasure reading Steve McIntyre’s posts as I did from watching 49er games.
• Congratulations, UCLA press office! A study that measured the effect of omega-3 by comparing two groups of rats — one gets omega-3, the other doesn’t — is called a study about the evils of fructose (both groups got a high-fructose diet). I am surprised the scientists involved didn’t object to this misrepresentation. The study supposedly shows — according to the press office — that fructose is bad because performance went down when the rats were switched from standard lab chow to a high-fructose diet. Let’s say you start with a diet (standard lab chow) that has a barely adequate amount of omega-3. You feed both groups lab chow for several months. Then you do an experiment in which both groups get 60% of their calories from the lab chow and 40% of their calories from a diet that contains no omega-3. Performance is likely to decline due to insufficient omega-3 no matter what the new diet contains.

Thanks to Tim Beneke.

Patrick Tucker, an editor at The Futurist, posted a request on the Quantified Self Forums for “astounding” predictions based on self-quantification. He is writing a book about using data to make predictions.

Here are examples from my self-measurement:

1. Drinking sugar water causes weight loss. The self-quantification was measuring my weight. It began when I found a new way to lose weight, which pushed me to try to explain why it worked. The explanation I came up with — a new theory of weight control — made two predictions that via self-experimentation I found to be true. That gave me faith in the theory. Then the theory suggested a really surprising conclusion, that loss of appetite during a trip to Paris was due to the sugar-sweetened soft drinks I had been drinking. If so, drinking sugar water should cause weight loss. (The nearly-universal belief is that sugar causes weight gain, of course.) I tested this prediction and it was true. More.

2. Seeing faces in the morning improves mood the next day (but not the same day). This is so surprising I’ll spell it out: Seeing faces Monday morning improves my mood on Tuesday but not Monday. For years I measured my sleep trying to reduce early awakening. Finally I figured out that not eating breakfast helped. There was no breakfast during the Stone Age; this led me to take seriously the idea that other non-Stone-Age aspects of my life were also hurting my sleep. That was one reason I decided to watch to watch a certain TV show one morning. It had no immediate effect. However, the next morning I woke up feeling great. Via self-measurement of mood, I determined it was the faces on TV that produced the effect, confirmed the effect many times, and learned what details of the situation (e.g., face size) controlled the effect. More.

3. One-legged standing improves sleep. Via self-measurement I determined that how much I stood during a day controlled how well I slept. If I stood a long time, I slept better. Ten years later I woke one day after having slept much better than usual. The previous day had been unusual in many ways. One of them was so tiny that at first I overlooked it: I had stood on one leg a few times. Just for a few minutes. Yet it turned out that it was the one-legged standing that had improved my sleep. Without the previous work on ordinary standing I would have ignored the one-legged standing — it seemed trivial.

4. Butter is healthy. I found that butter improved how fast I can do arithmetic problems. No doubt it improves brain function measured in other ways. Because the optimum nutrition for the brain will be close to the optimum nutrition for the rest of the body — at least, this is what I believe — I predict that butter will turn out to be healthy for my whole body, not just my brain.

5. Mainstream Vitamin D research is all messed up. Via self-measurement I confirmed Tara Grant’s conclusion that taking Vitamin D3 in the morning (rather than later) improved her sleep. It improved my sleep, too. When I had taken it at other times of day I had noticed nothing. Apparently the timing of Vitamin D — the time of day that you take it — matters enormously. Take it at the right time in the morning: obvious good effect. Take it late in the evening: obvious bad effect. Vitamin D researchers haven’t realized this. They have neither controlled when Vitamin D is taken (in experiments) nor measured when it is taken (in surveys). Because timing matters so much it is as if they have done their research failing to control or measure dose. If you fail to control/measure dose, whatever conclusion you reach (good/no effect/bad) depends entirely on what dose your subjects happened to take. And you have no idea what dose that is.

A few days ago I asked Mark Qualls, a 59-year-old truck driver who lives in Longmont, Colorado, about his success with the Shangri-La Diet, which he posted about.

How did you learn about it?

Freakonomics.  When I read about you in that book, it made sense to me. The whole idea of a setpoint. I used to be an accountant. I weighed 290 pounds. I’m 6′ 2″. I lost 25 pounds when I started driving a truck. I’ve been there for almost 12 years. Around 260. I get a lot of exercise delivering groceries. I can eat anything I want but the idea of going on a diet makes me hungry.  My doctor said lose a bit of weight but I just couldn’t do it. (more…)

In The Shangri-La Diet I noted that hospital patients given intravenous feeding often lose a lot of weight without hunger. I said this supported my theory that the body fat set point is raised by the smell of food. Without smell, the set point goes down. When your set point goes down you lose weight without becoming hungry.

You should be able achieve the same effect by nose-clipping all your food. A new diet, however, makes smell avoidance considerably more difficult and expensive.

The K-E diet, which boasts promises of shedding 20 pounds in 10 days, is an increasingly popular alternative to ordinary calorie-counting programs. The program has dieters inserting a feeding tube into their nose that runs to the stomach. They’re fed a constant slow drip of protein and fat, mixed with water, which contains zero carbohydrates and totals 800 calories a day. Body fat is burned off through a process called ketosis, which leaves muscle intact, Dr. Oliver Di Pietro of Bay Harbor Islands, Fla., said.

It is a hunger-free, effective way of dieting,” Di Pietro said. “Within a few hours and your hunger and appetite go away completely, so patients are actually not hungry at all for the whole 10 days. That’s what is so amazing about this diet.”

Di Pietro says patients are under a doctor’s supervision, although they’re not hospitalized during the dieting process. Instead, they carry the food solution with them, in a bag, like a purse, keeping the tube in their nose for 10 days straight. . . .

Schnaider said she was never hungry throughout the 10 days she was on the K-E diet, but admits that it still wasn’t easy. “It was emotionally difficult, the 10 days of not eating,” Schnaider said. . . . Although the K-E diet is new to the United States, it has been around for years in Europe. Dr. Di Pietro charges $1,500 for the 10-day plan, and says the before-and-after pictures sell themselves.

I sympathize with the “emotionally difficult.” When I lost 30 pounds in 3 months drinking sugar water, I ate maybe 50% of my usual calorie intake. I was never hungry and that too was bad. The world seemed drab without hunger.

Thanks to Tom George.

• Where are they now? J. S. Boggs, profiled by Lawrence Wechsler in The New Yorker. Boggs made small paintings closely resembling money (e.g., a $100 bill) that he offered in place of real money. He sold surrounding details (e.g., the receipt) to a collector who would try to get the bill Boggs had drawn from the merchant in order to “complete” the work of art.
• A SLDer (Shangri-La Dieter) loses 80 pounds in 18 months. That’s 1.0 pounds/week.
• More medicine does not equal better medicine.  I agree with every word of this critique by a Glasgow general practitioner named Des Spence. For example, “The prescribing of powerful antipsychotic and potentially addictive stimulant drugs to children is a societal norm. . . . A quarter of US women are taking mental health drugs.” As Spence says, these are signs of a healthcare system biased toward those who make money from it and against everyone else (including children). One way to sum up why this is a mistake: Your health is too important to be left to those who only make money if you are sick.
• Japan: from rice to wheat to rice.

Thanks to Bryan Castañeda.

• White rice associated with Type 2 diabetes.
• Whistleblower says drug company failed to report bad side effects. “She also claims that the company failed to report 28 of 100 cases of bladder cancers to the FDA, which she called a “serious discrepancy,” and that Takeda told her to change her notation that the cancers were “related” to “unrelated.” The suit says that she was fired after she complained to her supervisors about the company’s failure to report all serious adverse events.”
• Natto speed eating contest. Why do some foods (hot dogs, natto) get a speed-eating contest, but not others?
• Drinking olive oil with your nose held shut (short video).
• Is self-experimentation n=1? or is that misleading?

• The high value of free food. “About 1 in 4 doctors take cash directly from drug companies or medical device makers and 2 out of 3 take free food  . . . Those who [do so] are more willing to prescribe drugs in unapproved or potentially risky ways.”
• Beware the North Carolina Board of Dieticians.
• Tonsils make T-cells. One more reason cutting off part of your immune system (your tonsils) is a terrible idea.
• Long-term Shangri-La Diet success. “I’ve been doing the Shangri-La “Diet” for over 5 years now (see here, here, and here). My weight went down right away, and it’s stayed down.”

Thanks to Anne Weiss.

Here’s a new graph I’ve made of Alex Chernavsky’s data. In 2001, he started weighing himself and recording his weight with the hope that it would help him lose weight. His data shows several interesting things:

1. Long walks really helped. The walks lasted 1.5-2 hours. They weren’t sustainable but the weight loss they caused lasted a remarkably long time — years, apparently, in the sense that it took years to regain the lost weight.

2. A low-carb diet worked well, but only at first. Alex lost a lot of weight initially but then started to regain it. Just before he became vegetarian, he was regaining weight quickly. I don’t know if this is typical. The popularity of low-carb diets has not been matched by availability of data about long-term effects, where by “long-term” I mean four years. Even though low-carb diets are 150 years old (Banting wrote in 1863).

3. The Shangri-La Diet is working better than other alternatives. There’s a difference between (a) showing that a diet causes weight loss and (b) showing that it works better than other ways of losing weight. In this comparison, it appears more sustainable than long walks and the weight loss it causes appears more sustainable than the weight loss from a low-carb diet.

Alex originally used Shangri-La Diet principles by ingesting 4 tablespoons of flaxseed oil washed down with water. (Details here.) He lost weight but then started to slowly regain it. I suggested he increase his intake of flavorless calories so he started to eat 1 tablespoon of coconut oil (about 100 calories) each day with his nose clipped. He stopped slowly gaining weight.

I asked Alex why he has persisted weighing himself so long. He replied:

I had at best a vague idea of what I wanted to do with the data.  When I was in graduate school [in neuroscience], I enjoyed plotting the results of my experiments, so I thought it would be fun to have a dataset that consisted of my own weight measures.  After I started the SLD, I had a more-concrete reason why I needed to collect the data.  I explicitly set out to test the diet.

• In praise of Rush Limbaugh.
• Shangri-La Diet experience (“Bottom line: I lost three pounds in a week and a half”) of an artist named Elizabeth Periale.
• Long interview with Tucker Max. “His fridge . . . is in one way very different: where you’d expect the six-pack of cold ones waiting for the game, instead you’ll find rows and rows of kombucha, the fermented health beverage.”
• End of college campuses. Megan McArdle imagines a world in which college is replaced by distance learning. “95% of tenure-track jobs will be eliminated.” Jane Jacobs, in Systems of Survival, divided jobs into taking and trading. Teaching is trading if the student really wants to learn the subject. Teaching is taking if the student is forced to take (and pay for) the class. Scary thought: Every college student is asked about every class: would you take this class if you didn’t need to (and didn’t need to take other classes)?

The key point — as far as I’m concerned — in this article about the DIYization of beer brewing comes in the middle of a paragraph:

Home brewing is part of a broad spectrum of DIY activities including amateur astronomy, backyard biodiesel brewing, experimental architecture, open-source 3-D printing, even urban farming. . . . Many of these pastimes can lead to new ideas, processes, and apparatus that might not otherwise exist.

Likewise with the DIYization of science: It will produce new ideas, solutions, etc. The Shangri-La Diet is an example.

Thanks to David Archer.