Food coloring & hyperactivity?

[TransformX]

TX, some things are not clear to me and I hope you can enlighten me.
- Do you still believe that controlled testing methodologies for drugs on humans is evil?
- Do we agree on why Dr. Bandadi argues not to use controlled testing for his therapy (ethical consideration or scientific suitablility)?

-I believe that controlled testing in the form of drug vs. placebo, while being the best options in some cases, is the lesser evil in others and plain evil yet in other cases. I believe each kind of drug test should be customized for the drug and the situation in which it's being tested. Too bad too many people aren't open minded enough or too lazy to compare different kinds of data.

- Dr. Bandadi wants to test his therapy in his own way because 1. He's certain it works. 2. If more people are introduced to his group, why play with placebo if he can HELP them? and 3. He has a way of quantifying the drug effectiveness without needing a control group!

[James_D]

TX: I don't disagree with what you say, but you have to consider that there are a thing called the scientific method, which when you are doing research you have to follow. It contains certain principles that the process must be objective to reduce a biased interpretation of the results. Another basic expectation is to document, archive and share all data and methodology so it is available for careful scrutiny by other scientists, thereby allowing other researchers the opportunity to verify results by attempting to reproduce them. This also allows statistical measures of the reliability of these data to be established. And Dr Bandadi makes this difficult.

All data and methodology have to be reproducible.

JD

[Brian Ellis]

TX: I don't disagree with what you say, but you have to consider that there are a thing called the scientific method, which when you are doing research you have to follow. It contains certain principles that the process must be objective to reduce a biased interpretation of the results. Another basic expectation is to document, archive and share all data and methodology so it is available for careful scrutiny by other scientists, thereby allowing other researchers the opportunity to verify results by attempting to reproduce them. This also allows statistical measures of the reliability of these data to be established. And Dr Bandadi makes this difficult.

All data and methodology have to be reproducible.

JD

Absolutely!

Unfortunately, it appears that Tx is allergic to good science.

[Umfriend]

I'm out.

[Jammrock]

I'm out.

Me and Umf on the same side of an arguement...who would have thought. The two people who actually know people with food coloring related hyperactivity are dismissed offhand because a few people just don't believe it, and therefore the study was pseudo-science. I love it

[James_D]

I don't know if you where referring to some of my comments. I haven't been able to find the original article that the science daily was referring to. But the food coloring/additives debate started in 1975 and has been going on since then.
I assume that its Prof Stevenson at Southampton Uni, that they are referring to in the article. He made article in Proc Nutr Soc. 2006 Nov 65(4) page 361-5. Where he published some of their initial findings, but here also Fatty acids where seen as a contributing factors, and it was a report on their initial findings. So whether it is pseudoscience I will not say, untill the final report is published and reviewed.

JD

[Umfriend]

We disagree that often? I'd have thought you'd be on the right side of an argument more often!

In any case, I'm out on the debate whether controlled testing is evil or not. No use.

There are quite a number of decent tests, many double-blind with control groups. The results are mixed and, as far as I can tell, more recent studies tend to show (not neccesarily prove) a connection between various ingredients (a.o. often colorants) and adverse behavioural phenomena. The more recent studies appear more solid in that they more often subject participants to a diet for a longer period and then introduce more than just one substance.

With respect to having to be reproducible, Brian, you must realise that climatology is, to date, a science that relies on un-reproducible data. Sure, you can run the same model over and over, you can try to understand parts of the chemical and physical processes, but that does not mean you can test the outcomes of models and hence, validation is an extremely difficult issue. Furthermore, the tests I spreak about are all reproducible but still result in different conclusions. Reproducability can often fall short of what is required to be a decent research project. Shit, I need to take my son to swimming class, bbl.

[James_D]

There are quite a number of decent tests, many double-blind with control groups. The results are mixed and, as far as I can tell, more recent studies tend to show (not neccesarily prove) a connection between various ingredients (a.o. often colorants) and adverse behavioural phenomena. The more recent studies appear more solid in that they more often subject participants to a diet for a longer period and then introduce more than just one substance.

True, there seems to be a growing number of studies indicating that food additives can produce adverse effects in people.

JD.

[Brian Ellis]

With respect to having to be reproducible, Brian, you must realise that climatology is, to date, a science that relies on un-reproducible data. Sure, you can run the same model over and over, you can try to understand parts of the chemical and physical processes, but that does not mean you can test the outcomes of models and hence, validation is an extremely difficult issue. Furthermore, the tests I spreak about are all reproducible but still result in different conclusions. Reproducability can often fall short of what is required to be a decent research project. Shit, I need to take my son to swimming class, bbl.

I agree that the prediction of climate cannot be verified in advance, but retro-calculation allows the models to be compared with past climate and this is a key verification process. Then, as the future becomes the present, further verification becomes possible. Obviously, to take an extreme hypothesis, if all the major volcanoes throughout the world blew their top the day after tomorrow, this would provide a severe change in the climate but such an event would be totally unpredictable and therefore cannot be included in the models until after the event. Similarly, if someone decided to change Iran or N. Korea into glass. That is why several typical scenarios are calculated, but the atypical has to be ignored.

As an illustration, with weather, not climate, I run a sophisticated weather forecasting software. Early this week, I forecast heavy rain for the latter half of the week. My forecast was unprecedented for this time of year and all the local farmers, gurus etc. said that my modelling was faulty and it could never happen. Guess what? We got 30 mm (so far) today! Yet I admit the modelling was faulty, as it predicted about 10 mm max up until last night, when it upped the ante to a whopping 60 mm, so the modelling made quantitative errors, but not qualitative, but was sufficiently accurate to confound the gurus. OK, this is anecdotal but it does show that mathematics can be used for forecasting, when it is confirmed by past experience.

However, this is not the subject of this thread. I know nothing about sensitivity of individuals to food colouring. I know a little about scientific methodology and I know of no way of final testing new drugs except by double-blind testing with placebos, AFTER all the animal and computer tests prove positive, including on healthy persons and on those the drug is supposed to help. My knowledge of toxicology tells me that there can always be unforeseens because of different metabolic paths, sensitivities etc..

[TransformX]

http://en.wikipedia.org/wiki/Philosophy_of_science

http://www.allaboutscience.org/laws-of-thermodynamics-faq.htm

Sometimes faith in science can lead brilliant men to believe things that are not true, because they haven't seen everything that science - or the universe - has to show them. Lord William Thomson Kelvin, for whom the Kelvin temperature scale is named, stated in 1895, "Heavier-than-air flying machines are impossible." One of the greatest scientists of all time was proven wrong, along with many of his colleagues, just eight years later by the Wright Brothers. There are many popular and widely accepted theories that science cannot prove beyond doubt. Yet, the way they interpret the evidence is good enough for some people to put their faith in. There is great danger in ignoring evidence that contradicts a scientific theory, claiming that the theory is "Modern Science," and so it must be true. What would Lord Kelvin say to that? Like the Laws of Thermodynamics themselves, there have been some powerful, universal, undeniable truths that science has been wrong about in the past. What theories and ideas does "Modern Science" claim today that might not be true? Are there any views of the universe that haven't proven false?

No Bryan, I'm allergic to people who think they know everything, people who accept things because someone with a higher academic degree or a bunch of yes-men claim to be true. A healthy dose of scepticism is necessary in order to advance and not cling to religious, false scientific or simple fanatic ideas.
edit: We could go back to the thread about global warming if you'd like.. http://forums.murc.ws/showthread.php?t=61699

[Mehen]

Me and Umf on the same side of an arguement...who would have thought. The two people who actually know people with food coloring related hyperactivity are dismissed offhand because a few people just don't believe it, and therefore the study was pseudo-science. I love it

Your arguments are much more convincing than the original article. The article isn't nearly long enough and doesn't discuss their methods.

[Dr Mordrid]

Not that I'm anti randomized trials, but there are instances where non-randomized data is useful. An example is how many "native treatments" we still use, though we aren't aware of their origins. Long before trials people used them for the best of reasons; they worked. Sometimes experience and trial/error isn't all bad.

Examples: honey from certain flowers on infections (used on diabetic sores), various antibiotics, digitalis & its family, cloves for pain etc. etc.

The Egyptians of 3500 years ago had very effective treatments of these kinds.

That said here'a a peer review editorial of the Inoges, Bendandi et al publication in the Journal of the National Cancer Institute;

Inoges, Bendandi Synopsis link....

Peer review editorial....

Idiotype Vaccination in Follicular Lymphoma: Knocking on the Doorway to Cure

Dan L. Longo

Correspondence to: Dan L. Longo, MD, National Institute on Aging, National Institutes of Health, Intramural Research Program, Rm. 1E07, 5600 Nathan Shock Dr., Baltimore, MD 21224-6825 (e-mail: longod@grc.nia.nih.gov).
>
>
The Inoges et al. study (14) is not a randomized trial. But I find the data in the paper persuasive that idiotype vaccination is influencing the natural history of disease. If the observed remissions remain durable, the stage seems set for a head-to-head comparison between rituximab and idiotype vaccination as postremission therapy in patients who achieve an initial chemotherapy-induced complete remission. If that comparison reveals a clear winner based on a disease-free survival endpoint, it may be time to consider putting in the effort to conduct a major study assessing overall survival.
>

[Umfriend]

As I read it Doc, what Longo is saying is that the data is persuasive and that at some stage, a good comparison between rituximab and idiotype vaccination is in order. My guess would be that that would be a double blind test, no?

[Dr Mordrid]

Yup, but sometimes it takes this kind of non-randomized work to get the funding for a full d/b trial. Having Longo write an editorial like this is a good first step, but the real mover will be the progress of the patients. If they do well it'll be hard to deny funding.

This peer review seems on a good path, but there are big problems with the peer review system; reviewers may not be interested in advancing new or contrary data if it challenges their life's work or preconceived notions. This of course often causes funding, and post-grads, to chase the established preconceived notions.

Then there is the problem of peer reviewers who never read the work, they just draw and shoot based on irrelevant factors like the institution of origin, synopsis, authors etc. regardless of content.

Examples abound and many heavily cited papers, including some describing work that llater won a Nobel Prize, were rejected by peer review.

An interesting book on these problems is "Peer Review: A Critical Inquiry" by David Shatz of Yeshiva University.

[James_D]

Then there is the problem of peer reviewers who never read the work, they just draw and shoot based on irrelevant factors like the institution of origin, synopsis, authors etc. regardless of content.

Examples abound and many heavily cited papers, including some describing work that llater won a Nobel Prize, were rejected by peer review.

An interesting book on these problems is "Peer Review: A Critical Inquiry" by David Shatz of Yeshiva University.

Have you read the book? It is a good book but he is also favoring the peer reviews system, with some small modifications, like anonymous authors for the peer review. And also your claim that peer reviewers do a less than optimal job is somewhat flimsy. If you read Lancet, or BMJ or Nature, you will see that there is a very wide spread of authors and institutions that get published. Of course papers get rejected it but many get rejected, revised and readmitted and then published.

Kind regards

JD.