Coeliac charity UK has good information about this http://www.coeliac.org.uk/healthcare-professionals/diet-information/oats-in-the-gluten-free-diet

It’s pretty much impossible  to be totally gluten free because contaminants happen in restaurants, in the production and up till recently there has been little enforcement or checking.

Glutten isn’t essential but wheat is an incredibly good source of vitamins and proteins and inadvertently has probably contributed more to good health over the last 2000 years than any other food source.

Between 1900 and 1940, at least 100000 individuals in the southern United States died of pellagra due to over reliance on a maize based diet rather than wheat. And similarly in parts of the world now which rely on deficient grain and a staple, cases of pellagra and beriberi unfortunately still occur.

It’s interesting that many of the vitamins that people think of as somehow virtuous and of nature were discovered by doctors who were trying to explain diseases like pellagra. Many of the important discoveries surrounding essential nutriens were by doctors who self experimented by eliminating certain foods and seeing what symptoms they got – some unfortunately died in the process. Its a fascinating story which I wish Seth would pay adequate lip service to.

you don’t just run random tests on people. the public have this idea that tests tell you ‘yes’ or ‘no’ . 
If I ran a HIV test on everyone in US then 1000s of people would get a false positive result and 100s of people with HIV would be falsely reassured.

People present with symptoms, and the doctor’s job is to put the history and examination findings together and order tests which help confirm the diagnosis. Coeliacs is tested for when people present with malabsorption, unexplained weight loss, unexplained diarrhoea, fatty stools, or in children as a failure to thrive. Characteristic rash of dermatitis hepetiformis.
If someone presents like this then it would be negligent not to check for coaeliac antibodies – but a rather uncomfortable and expensive camera down the throat and biopsy – which rarely but seriously can cause stomach perforation is required to make a diagnosis – the blood test on it’s own is far too inaccurate.

So you can see that for people with attributable symptoms, a positive result for coeliacs related antibodies and a convincing clinical picture means its worth going ahead and scoping.

If take someone who has headaches and feel a bit tied and you run a screen because you want to be a caring doctor – which shows a positive coealiacs antibody – do you tell them to quit gluten now? No you need to subject them to an invasive test which was not related to their original symptoms – if that comes back negative once, you probably should probably scope again after a week of high gluten diet.

most of the test for coeliacs antibodies are around 90% specific, which means 10 people in every 100 healthy people will be identified falsely as ‘positive’ and require an invasive test to confirm.

The problem with these test is that you have to be eating gluten in order to make a diagnosis- so if you were at all concerned about having coeliac - the worst thing to do would be to go onto a gluten free diet without discussing your doctor as he or she will have to convince you to come off your diet in order to make the diagnosis.

I know it’s long winded, but the devil is in the detail. 

I agree that people should feel empowered and encouraged to find foods that agree with them. People do this all the time, I can’t drink more than half a glass of orange juice as it repeats on me in a most uncomfortable way  but i found that a small amount is absolutely fine and so i don’t need to concoct some cargo cult science about oranges to explain that something disagrees with me – i think that’s what people take issue with.

rubbish. Wheat is consumed in massive quantities and was key to human population growth and civilisation.

It’s not the historical staple of East or South Asia, Africa or the New World. So what are you on about?

The interesting point then becomes the fact that if there was a new pill developed that would negate the effects of Coeliacs – does medicine prescribe that first line or suggest a gluten free diet as first line. I have friends who are self diagnosed ‘wheat intolerant’ who would probably stick with a diet as it was their choice to do that – where as my Coealiac friends would love to be able to eat their favourite foods – they had no choice but to go on the diet.

But, also, isn’t that where placebo controls come into the picture?  If one person says their fatigue improved but they didn’t know that they got the placebo rather than the actual treatment (drug/supplement/etc.), then we score their response as “wrong” (false positive, actually) and not anecdotal evidence of efficacy.

Unless there are a lot more people saying their fatigue improved in the treatment group than in the placebo group, then you can’t say the treatment worked.  Anecdotal information from self-experimentation has no placebo control (or any of the other ways to control for such comparisons).

So, is the person telling you how great their untested remedy works a false positive or a proof of treatment benefit?  The fact that we actually get false positives in carefully controlled clinical trials means you can’t tell based on how excited they are.

But, people who feel they benefited from an untested treatment are sharing what happened to them.  People should not be rude and say their personal experience is wrong (unless there was real harm).  A benefit to a person with no other options is a pretty nice benefit.

I am approaching tl;dr, so we can discuss sample size in another post if you like because new statistical techniques are hopefully going to get us down into the dozens not thousands for studies.

If you want to try a treatment option that has been proven ineffective in a clinical trial, you shouldn’t expect it to work for you (even if you feel that your set of symptoms is unique in the whole world), but there is always the long-odds chance it might.

If you have no clinically-proven treatment options (the case you are mentioning I think), then you don’t have any way of knowing ahead of time if an anecdotal treatment might work for you.  If it is not known to be harmful, why not roll the dice and try it?

If you feel better and your doctor says you didn’t (inadvertantly) harm yourself (things like kidney failure, not vague ruminations and finger shaking about what you might have possibly done) - woohoo, you are a new source of anecdotal  evidence for benefit.  Who cares about that – you feel better and didn’t hurt yourself!

So, using an untested remedy when there is a better option = bad.  Using a safe but untested remedy when there is no other option = seems OK to me.

Unless you’re using sequestered and monitored subjects, you also have to rely on anecdotes regarding diet, drugs/alcohol/smoking, activity level, whether or not they’re taking their meds, etc. Thus, you need rather large studies when you’re relying on anecdotal evidence. And you can still bet that they all said that they eat and drink less and exercise more than they really do.

I think I am  starting to see what I should have been saying.  You are right that data is made up of individual data points.  Some of the data points may come from the patient telling the doctor subjective information (anecdotal information in a sense).  The subjective information is cross-checked by objective tests and statistically compared to other patients (data).  Until you do that comparison, there is no way to know if the patient’s subjective observations (anecdotes) are typical or atypical (or maybe lying).

The collection of “anecdotes” in a clinical study does not equal Wisconsin Platt’s statement “the plural of anecdote is [not] data.”  If I don’t have data to compare to a person’s anecdotal experience, I have no idea if their experience is typical or atypical.  If 100 people claim a benefit based on their anecdotal experiences, I have no idea if they found a real cure or 100,000 people tried it and it didn’t work.  It is 100 anecdotes that don’t have any data for context – 100 anecdotes are not data.  It think “anecdotal data” is an oxymoron, but I am not sure. :)

Two of his relatives have the same reaction to not getting wheat, and as a result, I wonder if there are other people who have intractable symptoms because of something they’re not eating rather than something they are eating.

Also, if you supply your opcites, then you can be sure they’re the ones you had in mind.

For example, pain is very subjective to report, but mark where your pain is before and after on a visual analog scale and measure the distance between the two marks – you have a pretty consistent way to measure pain relief. Combine those in a clinical trial of a new pain reliever and you have data!

Yes, anecdotal data. That’s what I’ve been saying all along. How patients score pain has as much to do with whether their spouse was in a good mood this morning, whether they got good or bad news from their doctor, whether or not they like today’s nurse as it does with the pain med that they’re receiving. When one patient quantifies pain on one occasion, it’s an anecdote. When thousands of people quantify pain on thousands of occasions, it becomes data.

Please stop putting your links in parentheses.

For example, pain is very subjective to report, but mark where your pain is before and after on a visual analog scale and measure the distance between the two marks – you have a pretty consistent way to measure pain relief.  Combine those in a clinical trial of a new pain reliever and you have data!  Tell physicians how much pain relief is seen for which conditions, then they can do the guesswork as to whether the new drug will help a patient.

And, as you said before, alcohol consumption data relies on the honesty of the patient, but there are ways to cross-verify http://pubs.niaaa.nih.gov/publications/arh28-1/30-37.pdf

It is not as perfect a system as locking someone in a room (or London) under CCTV 24/7 to get accurate data, but it is a heck of a lot more predictive of efficacy than figuring out if the anecdote ”I feel better” applies to your own situation or not.

The Ionnidids reference is a theoretical calculation (not actual data) that talks about a “controversy” that the author created himself (refs 7 and 8 in the paper) based on false findings in microarray data (well known methodology problem).  The Atlantic article is about Ionnidids.  I haven’t had to retract half of my publications, so I am still looking for a good citation to back up the original statement.

Finding citations is quick and easy, finding the right ones is hard. :)

If your health was on the line, you would just do it. 

Instead of eating pasta every day, you would do what millions (billions?) of people around the world do. You would buy a rice cooker and make  a batch of rice every morning, or every second morning. (It keeps it hot all day while you work.) You would buy gluten-free soy sauce. 

You would buy a brand of stock that doesn’t have gluten. You would switch to rice crackers. And honestly, if you’re buying shredded cheese where there are more ingredients than, you know, cheese, there are larger issues. Half of Campbells soups are gluten free. 

BC is not a food shangri-la. I live on an island! When I go to grocery stores in the US there is approximately 5 times the amount of food. You can order millions of food products online that are unavailable to me. 

I have travelled in the US since switching to gluten-free. I’ve been to Fresno CA, Springfield MO, Binghamton NY, Woodinville WA and other small places for work. I have always managed to eat something. I mean, go to Red Lobster or the Outback or something and order meat, a baked potato and a salad with vinaigrette. Simple.

I just think if you start with the attitude that it will be impossible, then of course it will be impossible. 

We can’t all live in the gluten free shangra-la that is BC apparently.

Though I will one year grow my own wheat.

[1] Hippocrates (c. 400bce) On the Art of Medecine. Alexandria.

[2]  Oannidis JPA (2005) Why Most Published Research Findings Are False. PLoS Med 2(8): e124. doi:10.1371/journal.pmed.0020124 http://www.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal.pmed.0020124

[3] Freedmann (2010) Lies, Damned Lies and Medical Science. Atlantic Magazine http://www.theatlantic.com/magazine/archive/2010/11/lies-damned-lies-and-medical-science/8269/ (Accessed 26/06/2012)

(Has a personal hate-on for the kind of opcit people who are too lazy to google for themselves. This really wasn’t difficult).

My point is that there is a problem in western medicine; other countries do not have this problem. Is is the pharma lobby? Is it the wheat lobby? Is it invalid data or incorrect information in medical journals? Why do you think it’s more readily diagnosed in Europe than it is in the US?

Here’s the problem: American/western (yes, I know the story was in Vancouver) doctors are pushed by the drug industry to proscribe drugs. We don’t do a lot of healing anymore in this country. We just shovel drugs at people, and try to keep them hooked. Gluten intolerance/celiacs isn’t something you treat with drugs. You just don’t eat it. No money to be made for big pharma. You can call my cynical, but I had the same problem with doctors when I tried to get it solved. They shoveled pills at me and said it might all be in my head. I gave up eventually, and stumbled on the answer myself.

I don’t have a nut allergy but I have a nut intolerance, where some nuts make me feel very sick. I can eat peanuts, but walnuts, hazelnuts, macadamia nuts, pine nuts, and coconut milk seem to give me varying degrees of trouble. And it can be hard to avoid them, especially when it comes to arguing about ‘no, it’s not an allergy, it’s something else but they still make me sick’ and ‘what, you put coconut milk in that?!?’

I also have a nicotine allergy. I can’t eat eggplant and I have to avoid smokers. Sometimes that means dodging traffic to get to the opposite side of the street. Sometimes I’ll be sick for days because of exposure to someone’s cigarette smoke.

… society hasn’t yet adapted to that one. If I had been born a few decades earlier, I would have died of it.