I had a small head-injury while biking a few months ago, and have been suffering from persistent headaches every day since. The headaches aren't usually too bad, but they get much worse when I try to do things that normal people do all the time (like traveling by car). I'm told that this is sometimes called Post Traumatic Headache (although I did not receive a formal diagnosis), and it usually goes away on it's own with time.
Well, a few months was long enough so i started talking with a friend of mine who's a life-long migraine sufferer. I had noticed in my reading that different kinds of headaches are often treated the same way, and I thought he might have some good advice. Turns out he had some great advice: a diet described in the book "Heal Your Headache" by Johns Hopkins doctor David Buchholtz.
I was skeptical at first but he told me he's been on the diet for two months, and he's gone off his usual medications and he feels great. When I got the book, and flipped through it, two things struck me: 1. in the introduction, the author explains that his idea is, in scientific terms, a hypothesis (which is an untested idea based on observations), and 2, many of the listed dietary triggers are either widely known to be bad for you, or else often found in food that's bad for you. It was pretty clear reading the book that although little scientific evidence stood in favor of the diet, very little harm could come of it -- aside from avoiding citrus, most of the food Buchholtz recommends avoiding are in the category of "I've heard that some people think that's bad for you".
It's a tricky diet because the categories of food don't make too much sense, and many things you are supposed to avoid, such as MSG, are often not listed in the ingredients in food (MSG may be listed by aliases, such as "Natural Flavorings", or it may be hiding in other ingredients, such as "Malted Barley"). But it's manageable and hasn't meant that much change for me, aside from eliminating a few things that I usually cook and eat and being more careful about restaurants.
I've been on the diet now for four days and I have already noticed that although my morning headache hasn't improved, I feel much better in the evenings. (My headache used to get progressively worse over the course of the day) Well, it may not be science, but his technique seems to work, and that's good enough for me. (A placebo cure is better than no cure, that's for sure)
Sadly, I think Buchholtz discredits himself in three ways: 1. he criticizes doctors a lot, which may appeal to headache sufferers who's doctors have been no help to them, but it hurts the chances of other doctors taking his book seriously; 2. the book looks and reads much like any other self help book, most of which I am naturally skeptical of; and 3. he poo-poos the prospect of scientifically testing his ideas: "Scientists may with to test these hypotheses formally, using randomized controlled trials. Memo to colleges who undertake this challenge: Good Luck and I hope you do it right. If that's possible." He gives several reasons why it might be hard to do so, including his observation that some people are sensitive to some dietary "triggers" and not others, and that the effect of all the "triggers" (dietary or not) is cumulative, so eating chocolate or other dietary trigger might cause a headache one day in one subject, but not on another day or with another subject.
I agree that it would be nearly impossible to test each trigger individually, and that it would be a challenge to convince participants in an experimental trial to really follow the diet (many of his patients say they spend the first month or two just learning and understanding the diet). But there is no reason the hypothesis could not be tested. Rather than studying each trigger, the entire diet should be evaluated. Yes it will be hard to get people on the diet, but scientists do diet studies all the time, many of which are far more restrictive than his diet, so it can be done.
Perhaps one of Buchholz's most controversial claims is that virtually all headaches are migraines. Put another way that might be easier to swallow: there is no fundamental difference between migraines and other headaches. I don't have the medical knowledge to assess this, other than to say he makes a good argument, but I would say that from what I've read online there is generally no difference between the treatment of the various kinds of headaches, even if the diagnosis is different, so it makes sense to me that the underlying cause may be the same.