Here are some especially notable results (most notable first).

Other People’s

1. Bedtime honey greatly improved sleep. Stuart King found this after many other things had failed to help him. He got the idea from Dave Asprey, who got it from The Honey Revolution (2009) by Ron Kessenden and Mike McInnes, but Stuart made by far the best case that the effect was important and determined some boundary conditions (e.g., don’t eat a lot of sugar during the day). The improvement is so big and easy (honey tastes good) that it’s quite possible this is why evolution shaped us to enjoy sweets after dinner — to improve sleep. In the future, I believe, it will be understood that sugars (at the right times in the right amounts) are a necessary nutrient — exactly the opposite of what all nutrition experts, including paleo ones and Weston Price, say. When this stunning reversal will happen I don’t know — but no one will have foretold it more than Stuart.

2. Avoidance of glutamate cured autism. “Cure” is not too strong. Katherine Reid, who has a Ph.D. in protein chemistry, realized that many many foods, including ultrapasteurized milk, contain glutamate. When all glutamate was removed from her daughter’s diet, her daughter, who had been autistic, became completely normal. The generality of this solution is unclear but to cure even one case of autism is more than anyone else has done.

3. Xylitol eliminated lichen planus. Evelyn M. found that if she swished xylitol around in her mouth several times/day, her lichen planus and overall gum health greatly improved. This effect is well-known in Scandinavia but barely known elsewhere. American dentists don’t know it, for example. I didn’t know it.

4. Caffeine reduced reaction time. Alex Chernavsky wrote his own version of my brain tracking reaction-time test and found, in a well-designed experiment, that caffeine made him faster, at least for a few hours. What’s important here is the method, not the result: Alex’s success with a test that costs nothing and takes only a few minutes/day. His success brings closer the day that many people can do these tests. When non-experts realize they can study their own health in good experiments, the world will change.

Mine

1. Bedtime honey greatly increased my strength. (More here .) This convinced me that bedtime honey was something special, more than another way to improve sleep. (I also improved my sleep by darkening my bedroom — much less important.) Bedtime honey also seemed to improve mood and motivation. I have been trying to improve my sleep my whole adult life — this, I now see, is the secret. My discovery in the1990s that lots of standing improves my sleep led to the daily one-legged standing that made the strength increase so clear.

2. A banana a few hours before bedtime (in addition to bedtime honey) improved my sleep even m ore. The Honey Prescription says it’s been known for thousands of years that bedtime honey improves sleep, but no one seems to have noticed this, in spite of the importance of sleep. As Robb Wolf has said, “If someone sleeps well, you can’t kill them. If they sleep badly, you can’t keep them alive.”

3. Tofu made me stupid. One 20 g piece of fermented tofu slowed me down for two days on a reaction-time test. Other evidence has suggested that tofu is bad for the brain, but I find this the most persuasive. It addition to what it says about tofu, it suggests the power of brain tracking to reveal important things few people know. (An earlier example involved butter.) Billions of people eat tofu often.

4. Percentile feedback helped me work. The version of percentile feedback I use now (an R program that tracks my work and compares how well I am doing today to the last 100 days) is an improvement, I like to think, over an earlier version. The biggest difference is the addition of weighting: doing a hard task for one hour counts more than doing an easy task for an hour (e.g., a factor of 1.5, so that 60 minutes counts as 90 minutes). I kept track of how much I worked for a whole year, the first time I’ve managed to do that. Here’s what happened.

I think the lesson is that with the right push, I can slowly improve.

In my first post about the use of bedtime honey to improve sleep, I included a graph that showed my legs suddenly got much stronger when I started the honey. I measured how long I could stand on one leg (bent). I had been doing this four times per day (left leg twice, right leg twice) for a long time to sleep better.

A reader of this blog named Nile McAdams found bedtime honey caused him to get stronger, too. He measured arm strength.

Soon after my legs got much stronger I reduced my one-leg standing from four/day to two/day to save time. The improvement stopped, but the gains persisted:

Note the logarithmic y axis. Each point is a different day, an average of the first left leg stand and the first right leg stand of that day.

The graph shows I am much stronger with half as much effort. As I said, the strength improvement has also been easy to notice in everyday activities, including walking, stair climbing (I live on the top floor of a six-floor walkup) and bike riding.

I have been unable to find research that shows a similar effect. Judging by a 2010 textbook, exercise physiologists don’t know about it.

People who believe in “evidence-based medicine” say that double-blind clinical trials are the best form of evidence. Generally this is said by people who know very little about double-blind clinical trials. One reason they are not always the best form of evidence is that data may be missing. Nowadays more data is missing than in the past:

By [missing data] he [Thomas Marciniak] means participants who withdrew their consent to continue participating in the trial or went “missing” from the dataset and were not followed up to see what happened to them. Marciniak says that this has been getting worse in his 13 years as an FDA drug reviewer and is something that he has repeatedly clashed with his bosses about.

“They [his bosses] appear to believe that they can ignore missing and bad data, not mention them in the labels, and interpret the results just as if there was no missing or bad data,” he says, adding: “I have repeatedly asked them how much missing or bad data would lead them to distrust the results and they have consistently refused to answer that question.”

In one FDA presentation, he charted an increase in missing data in trials set up to measure cardiovascular outcomes.

“I actually plotted out what the missing data rates were in the various trials from 2001 on,” he adds. “It’s virtually an exponential curve.”

Another sort of missing data involves what is measured. In one study of whether a certain drug (losartan) increased cancer, lung cancer wasn’t counted as cancer. In another case, involving Avandia, a diabetes drug, “serious heart problems . . . were not counted in the study’s tally of adverse events.”

Here is a presentation by Marciniak. At one point, he asks the audience, Why should you believe me rather than the drug company (GSK)? His answer: “Neither my job nor (for me) $100,000,000’s are riding on the results.” It’s horrible, but true: Our health care system is almost entirely run by people who make more money (or make the same amount of money for less work) if they exaggerate its value — if they ignore missing data and bad side effects, for example. Why the rest of us put up with this in the face of overwhelming evidence of exaggeration (for example, tonsillectomies) is an interesting question.

Thanks to Alex Chernavsky.

• Jared Diamond wrong about Easter Island
• Why are the Kuna Indians (in Panama) so healthy? Answer: low-processed cocoa. See this study for details.
• Asthma associated with depression
• Sufficient conversion of ALA to DHA in rats. “ This work indicates that DHA [long-chain omega-3] synthesis from ALA [short-chain omega-3] may be sufficient to supply the brain.” Many people have complained that I use flaxseed or flaxseed oil to get better brain function. Don’t you know conversion from short-chain omega-3 to long-chain omega-3 is poor? I tell them it works — but somehow I must be wrong. This study shows the issue is more complicated than they realize and it is quite possible that ALA is a good source of DHA. Good news for anyone who can’t eat seafood.
• National Sleep Foundation gives terrible advice. “Finish eating at least 2-3 hours before bedtime,” they say. Whereas I believe the single best thing you can do for your sleep is eat a source of glucose and fructose, such as honey or fruit, at bedtime.
• Sleep apnea associated with cancer. It is unlikely that cancer causes sleep apnea, but quite plausible that bad sleep — including sleep apnea — increases cancer.

Thanks to Eugenia Loli.

Yesterday I posted Kristen Marcum’s list of general rules she’d learned from this blog. (For example, “be skeptical of experts.”) Behind her list, I think there is one idea, slightly hidden from view:

non-experts can discover important things about health

By non-experts I mean people who are not health professionals. People who do not make a living from health research. By discover I mean learn from data for the first time, actually discover — in contrast to learn from an expert. By important I mean stuff that matters to many people. (It’s obvious that studying yourself you can help yourself.) I haven’t heard anyone else say this, although it isn’t far from the Quantified Self movement.

The first example of this rule was the work of Richard Bernstein, an engineer with diabetes. In the 1960s, he pioneered home blood glucose testing, now enormously important. Another example, I hope, is my work. I used self-tracking and self-experimentation to find important new cause-effect relationships in several areas — new ways to sleep better or lose weight, for example. I believe my conclusions will turn out be true for many people, not just me, because they fit well with research done with other people and animals. I’m a professional scientist, which obviously helped, but not a health researcher.

Why is this rule true? I found it hard to explain even my discoveries. The first time I tried, I didn’t do well. I did better the second time.

I came to see that discoveries require several things — more precisely, rate of discovery is a product of several factors. Everyone sees that experts have several advantages over non-experts:

1. more training (graduate school)
2. more resources (equipment, grants, space, assistants, journals, specialized libraries, and so on)
3. more hours per day (all working day)
4. more experience (their entire working life)

What I’ve never seen pointed out is that non-experts have several advantages over experts:

1. more time overall (with no pressure to publish, a non-expert can easily spend many years on one question)
2. more freedom (no pressure to follow a party line, obey your boss or be popular)
3. more motivation (non-experts want to improve their own health, or a loved one’s health; for health researchers it’s a job)
4. ability to self-track and self-experiment (this is too humble for experts)

Because of my job, I was only slightly behind health experts in terms of their four advantages, and I was far ahead of them in terms of the four non-expert advantages. This explains why I discovered many things that experts had not managed to discover.

Once I had a reasonable understanding of what had happened, I saw there was no reason I should be the only one. As a professor, I had excellent library access, much better than most, but now, with the Internet, many people have comparable access to research. Supporting my prediction that I am not the only non-expert who can do this, I have found other non-experts who have made important discoveries. Tara Grant figured out that the time of day she takes Vitamin D made a big difference. Professional researchers have not yet figured this out. Katie Reid found that her autistic daughter completely recovered — autism symptoms completely gone — when all glutamate was removed from her diet. Professional autism researchers haven’t managed to cure even one case of autism, as far as I know. Stuart King figured out that a tablespoon of honey at bedtime greatly improved his sleep. No sleep or nutrition researcher has realized this.

In the 1940s, my mother’s mother sent my mother newspaper clippings about computers, of which, at the time, there were only a few. This is going to be important, she said. Let me continue the tradition by telling you this is going to be important.

When even a tiny fraction of non-experts become “health knowledge generators” (for lack of a better term) the world will change. Some of the reasons:

1. The IQ advantage. There are far more non-experts than health researchers. A factor of 100,000? Let’s say health researchers have an average IQ of 130. If you take 100,000 non-experts (mean IQ = 100, standard deviation = 15) the top IQ among them will be about 165.

2. Trust and generalizability. When a health expert advises you to do Treatment X (e.g., surgery), what should you think? Can you (a) trust him and (b) is what he says true for you? Any reader of this blog knows that health experts routinely overstate benefits and understate costs of the treatment they are promoting. Even if they are telling the whole truth, their advice is based, at best, on several studies. Is what those studies found — studying animals, or at best, other humans — true for you? When you study yourself – for example, carefully measuring the effect of a drug — error due to distortions and differences from the study population go to zero.

3. Good and bad chemicals. We encounter thousands of chemicals, many of recent origin. How we react to them — do they help or hurt? — is determined by our genome (billions of possibilities) and environmental history (countless possibilities). Any sort of expert testing will be woefully incomplete. The danger in assuming that a chemical safe for most people is safe for you is illustrated by a woman who found her almost constant migraines were caused by household cleaning products and Katie Reid’s discovery that her daughter was very sensitive to glutamate (found in a large fraction of supermarket foods). That things claimed to be safe may be dangerous is illustrated by the Berkeley graduate student who became poisoned by mercury when he started grinding his teeth and my discovery that fermented tofu made me stupid for a few days. My experience with tofu and butter shows how a non-expert can do better than simply accept what experts say. My brain tracking, which anyone could do, detected previously-unknown or -unclear costs and benefits.

Our economy runs on specialization. We buy cars from car specialists, for example. Long ago, professionals had a monopoly on knowledge. To learn math, you had to pay a math teacher. Then came cheap printing and mass literacy and mathematics books and libraries. The same knowledge became not just free but widely available. Then came the Internet and Wikipedia and Google — even more available. But all this time, the source of the knowledge, its creation, was restricted to professionals. Only professionals (or rich amateurs, like Darwin) could “make” (or “mine”) the knowledge. Now that will change.

Kirsten Marcum told me she had “put a number of [my] findings to use in [her] own life.” I asked how. She replied:

I’ve put a few of your specific recommendations to work (SLD, standing on one leg each day, omega-3s, more animal fat/pork fat, butter tea, fermented foods)…but in thinking about this, I realized I’ve gotten even more use out of general principles I’ve drawn from your blog over the years:

1. Trust your results. I’m thinking of your own (and other people’s) experiments with what caused acne (dairy, etc.) or migraines (soy) and helped restless leg syndrome (b6) or sleep, etc. You and others tried solutions, and the solutions either worked or they didn’t, and you/they looked to correct root causes, and those corrections either worked or they didn’t. If you still have acne, the acne treatment didn’t work–even if it should have, and even if an expert recommended it…and you probably shouldn’t just keep doing the treatment just because it’s supposed to work. And if you cleared your acne, then the intervention worked–regardless of whether it “should” have. If your brain performance improves in a measurable way after eating butter, trust that result–and be skeptical about the people who tell you butter is damaging your health. If treating a leaky gut improves your thyroid labs, trust that–not the doctor who says your leaky gut diet is too restrictive and is going to damage your health.

2. Many unpleasant circumstances are reversible. It’s often possible to find positive interventions that ameliorate, reduce, or remove things that are bothering you–many of which are cheap or free and not hard to implement. It’s possible to fix health issues, productivity problems, make learning more fun or effective, etc. It’s amazing to me how many people don’t think this is true–that you’re stuck with extra weight, or acne, or creaky joints, or back pain, or migraines, or the inability to motivate yourself to work–and there’s nothing you can do.

3. Most things that are beneficial are also pleasurable, and vice versa. When something is good for us, we usually find it pleasurable (animal fat, salt, sugar at night, umami flavors, learning while moving, faces in the morning). Otherwise, how would we have evolved to do it?

I extend this to say: If something’s not pleasurable but supposed to be beneficial, it’s either bunk or you’re doing it wrong (i.e., in a way that removes the pleasure and possibly the benefit.) So look for the pleasure–because then you’re more likely to do it and to see a benefit.

And: If something is pleasurable but not beneficial, your body probably wants something very similar and it’s confused–junk food umami vs. fermented foods. In which case, find a pleasurable AND beneficial substitute, and it will be easy to stop doing the thing that is not beneficial.

4. Be skeptical of experts. Most experts have incentives that have nothing to do with the reasons you’re consulting them. Understand this, look for their motivations/weaknesses/blind spots, and always check what they’re telling you against what your body (or other results) are telling you. Also look for non-experts with interesting ideas.

5. What’s good for the brain is good for the body. I would add to this: Many things that we think of as personality or mental/emotional issues actually have a physiological basis–and optimizing your health will likely have a positive effect.

They sound elementary, but they’re counter to the way nearly everyone I know thinks.

Over the years, thinking like this has improved my life in multiple ways:

– I 100% cured my back pain (after 3 rough years) with Esther Gokhale’s 8 Steps to a Pain Free back–which cost $17 and provided more practical and helpful advice than the two doctors, physical therapist, and personal trainer I consulted.

– After a bunch of looking, I found a doctor who, when presented with new ideas, says: “That’s interesting. Where did you find out about it, and what have you tried?” More than once, she has looked into the subject, found the new recommendation convincing, and has started to share the information with similar patients–which means I also benefit from what she learns from others.

– I fixed my complexion, have nearly fixed my weight issues, and reversed dozens of nagging health issues (peeling fingernails, bad breath, rosacea flushing, random bad stomachaches, food sensitivities, mood swings, lack of motivation and focus etc.) through a whole-body thyroid/adrenal/gut repair approach that’s considered bunk by most mainstream endocrinologists. Parts of this approach happen to incorporate things you’ve found beneficial–increased amounts of animal fat, fermented foods, focus on optimal sleep, focus on aligning with circadian rhythms. (Interestingly, one of leading proponents of this approach started out focusing on thyroid health and has found even better results by broadening his focus to brain health….very much like your “what’s good for the brain is good for the body.”)

FWIW, I tend to use topics you’ve investigated as a check on other people’s recommendations–you’re not an “expert” in the field of thyroid repair, or gut health, etc., so you have no incentives to adhere or not adhere to a particular line of thinking. So if both you and a thyroid doctor find a particular approach beneficial, that seems very convincing to me.

Those are good lessons. Tomorrow I will explain why I think they all come from one somewhat hidden underlying lesson.

During the last week I have looked into the possibility that my sleep can be further improved — in addition to the bedtime honey improvement – by eating a similar amount of sugar (fructose and glucose) a few hours before bedtime. After I accidentally slept better than usual (or even better than usual), I tried to determine why. Several things had been unusual the day before. Two tests (here and here) pointed to the sugar (honey or banana) a few hours before bedtime.

Last night (Christmas Eve) I tried again. I ate a banana (132 g, peeled) about 3 hours (7 pm) before I fell asleep (10 pm). I fell asleep within a minute and woke up, after an apparently dreamless night, feeling perfectly rested. On my 0-100 percentage scale (100% = completely rested, no detectable tiredness), which I have been using for about 8 years, it was the first ever 100%. I had slept about 6 hours, a good amount of time.

To celebrate, I had a cup of black tea. I didn’t need it to wake up but I like the taste. I reflected that countless people had drunk tea or coffee to wake up. I had found a better way.

Discovery that an hours-before-bedtime sweet improves sleep (in addition to bedtime honey — that’s what’s interesting) is significant not just for the obvious practical reason (better sleep) but also because it is the confirmation of a prediction. After I slept unusually well, I thought of six possible reasons. The notion that sugar improves sleep pointed to one of them. The results of every test I’ve done (three nights) have agreed with that prediction. I believe the only real test of a theory (such as an explanation) is whether it makes correct predictions — especially, whether it leads to the discovery of new cause-effect relationships. Many things people say haven’t passed that test. An example is weight control. That low-carb diets cause weight loss has been known since the 1800s. Many explanations have been proposed; not one has made correct predictions, as far as I know. In contrast, my theory of weight control led me to three new ways to lose weight (sushi, low-glycemic foods, and fructose water).

I doubt it’s a placebo effect because the sleep improvement has happened whether I expect it or not. A commenter named Paolo Paiva, after reading my posts about this, realized something similar had happened to him:

Today I told my wife how deep I had slept and connected it to the 1 tbspoon of honey and 1 tbspoon of apple cider vinegar mixed with half a cup of water before bed (it tastes really good). Then I saw this post and remembered that yesterday I had had banana flour pancakes topped with honey 3 hours before bedtime!

Thanks, Paolo. May you continue to sleep well. May the rest of you sleep equally well.

Merry Christmas!

A reader writes:

My eight-year-old son has autism, on the serious end of the spectrum, with equally serious sleep troubles. A month ago I tried giving honey at bedtime after reading about it on your blog. It has made a tremendous difference.

My son has had trouble sleeping since he was a toddler. Since age four he has seen a pediatric neurologist who is also a sleep disorder specialist. With the doctor’s guidance we have tried different over-the-counter and prescription-drug approaches, but none has given him regular sleep, and the side effects from the drugs make increasing dosages risky.

His basic problem is not insomnia (a general inability to sleep) but rather delayed sleep phase disorder. He could sleep fine from the wee hours of the morning to noon. But whenever we managed to advance sleep onset [= get him to fall asleep earlier], he would wake up early. It was hard when we struggled to get him to sleep by midnight and then struggled to wake him at 7:00 a.m. for school. But it was even harder when he would fall asleep at 9:00 p.m., then wake up at 4:30 a.m., often falling back to sleep at 6:30 a.m, just in time for us to wake him at 7:00 a.m.. It’s not safe for him to be awake on his own for hours, and he was desperate to go back to sleep until about 9:00 a.m.

Since we started giving him a teaspoon of honey at bedtime he has slept through the night. We still have to give him meds to go to sleep between 9:00 p.m. and 10:00 p.m., but once he’s asleep he stays asleep all night until we wake him at 7:00 a.m. The one night, early on, when we forgot to give him honey, he woke up just as he used to.

One factor that may make honey particularly effective for my son is that we’ve been eating a low-carbohydrate paleo diet for the past two years (because it helps his autism symptoms). Before that he was an extremely picky eater who preferred baked goods, milk, and cheese. With high-glycemic foods all day long a teaspoon of honey at bedtime might not have had much effect.

This is the best my son has slept in his life, and it has been a huge improvement in his and our family’s lives. Before the past month, the last time my son had gotten three good nights of sleep in a row was because he’d had a concussion.

To recap: Three days ago I slept extremely well, better than usual. I wondered why. What had made the difference? That day (the day before the night I slept so well) had been different from previous days in at least five ways (e.g., chocolate, new brand of honey). I repeated four of them, and did not sleep better than usual. That suggested the remaining difference — I had eaten yogurt, blueberries (125 g) and honey (8 g?) a few hours before bedtime — was responsible. (Every night I had 1 tablespoon — 20 g — honey at bedtime. It wasn’t that.) Then I repeated all five elements, including yogurt, blueberries (125 g) and honey (14 g) two hours before bedtime. I woke up wired (jittery). Very rested, but wired, which wasn’t pleasant. Too much sugar, perhaps.

The next night I had a banana roughly two hours before bedtime. (In addition, I repeat, to 1 tablespoon honey at bedtime.) A banana has about 6 g glucose, 6 g fructose, and 3 g sucrose, similar to 1 tablespoon honey. I had a strong craving for something sweet at that time, which was new to me — I almost never eat dessert. In the evening I had more brain power than usual. Yet at bedtime I fell asleep quickly, in about a minute.

The next morning I woke up and felt great. Almost perfectly rested, neither tired nor wired. Even though I’d only slept 4.7 hours, a bit low for me. It really was the yogurt, blueberries and honey — almost surely their sugar, which is almost all they have in common with a banana — that had made me sleep so well.

Conclusion: For the best sleep, have sugar after dinner and sugar at bedtime. By sugar I mean a glucose/fructose mixture but for all I know sucrose would work, too.

A few days ago I noted that I had slept unusually well. I wondered why. The previous day had been unusual in five ways (yogurt blueberries and honey 2 hours before bedtime, only one blackout curtain, chocolate, unusual timing of morning faces, new brand of honey). Was one of them responsible? Or was it random variation?

The next day I repeated four of the five unusual things: only one blackout curtain, chocolate, unusual timing of morning faces, new brand of honey. Result: I did not sleep unusually well. This pointed to either yogurt blueberries and honey or random variation as the explanation. Those happened to be the explanations I had considered most plausible (yogurt blueberries and honey) and least plausible (random variation).

Yesterday I repeated all five of the unusual things. Yesterday evening it took longer than usual to fall asleep. Usually I fall asleep within a minute, but this time it took about 4 minutes. As I was lying there, I attributed the long latency to lack of exercise that day. In the morning, for the first time in my whole life, I woke up what a friend called “full of jitters”. Very rested, yes, but also too much energy. As if I’d had too much caffeine. Usually I have a cup of black tea first thing in the morning. I started to make one and realized it would make things worse, not better. As I said, this has never happened before.

So it was the yogurt blueberries and honey. The first time I had it I had added the honey just for sweetener and hadn’t measured it. The second time I did measure the honey — 14 g, about 2 teaspoons. Probably more than I had used the first time, which may explain the different results (1st time: very rested, 2nd time: very rested and jittery). Both times I ate 125 g blueberries, which has 12 g sugar (half glucose, half fructose). I doubt the yogurt (homemade, no sugar added) matters.

Perhaps the best dose (for me) will turn out to be one teaspoon honey 2 hours before bedtime and one tablespoon of honey at bedtime. I had tried taking more than one tablespoon of honey at bedtime; it seemed to produce the same effect as one tablespoon.

I want to test with and without blueberries, of course, and different amounts of honey 2 hours before bedtime. If honey alone is powerful, then I will test different forms of sugar. I don’t want to be at the mercy of differences between brands of honey, although honey is very convenient.

Let’s say it turns out to be the sugar in honey that produces these big improvements in sleep. Then it will turn out that the nutritionists of the world have been even more wrong about sugar than they were about saturated fat. All the data is not yet in, let me repeat. But the data so far unquestionably point in a surprising direction. It is entirely possible that sugar — in the right amounts at the right times of day — will turn out to be the greatest health food of all.

Which would explain what has never been well-explained: why we like it so much.