[Alex]

Originally Posted by tracilicious

I'm ok with most any alternative treatment. I don't believe in a lot of them, but if others want to then that's ok. At worst they do nothing in most cases.

And, for many things (such as cancer), nothing is a pretty horrible result. Which is why courts have long interceded (to address the original topic) when treatments that have been shown to be at least minimally effective are disregarded in favor of nothing or things that have no scientific support.

At least when it is a minor. The question is a very tricky one in deciding what parents are not allowed to decide for their children, but this recent decision is hardly precedent setting.

I believe in chi/meridians, but I have no proof. I'm going to assume that there is some factual/logical basis in there somewhere. Someday I'll do the actual research to confirm it.

I'm honestly ok with people (adults) deciding to follow faith-based approaches to life and health. Just call it that. It is when claims of scientific support are made for faith-based decisions that my back gets up a bit.

If you believe in forms of energy that can not be detected or tested then who am I to argue otherwise, but this is essentially the same view as Christian Scientists praying for God's intervention. Now, I am of the view that if those faith-based approaches actually do any good then the results will show up pretty clearly in standard evidence-based analysis. If people would prefer such analysis not be done, or not believe the results when they are done, then what can be done.

As far as anecdotes go, what isn't an anecdote? Don't many studies consist of giving a medicine and asking patients how they feel? Thus, a great deal of science is based in anecdote. I don't feel that anecdotes automatically disqualify anything.

Of course anecdote isn't automatically disqualifying and frequently it is indicative of something real. But it is also frequently indicative of misperception, false positives, selection bias, etc.

There are procedures for creating studies that remove those biases. In a properly double-blinded study, neither giver nor receiver of care knows what type of care was given. Therefore, biases of perception can't easily influence results. Particularly in the area of pain and discomfort it is important to remove these because perception of pain is so subjective. What is intolerable to one person is hardly noted by another and there is no way to measure it other than self-evaluation.

There are also statistical models for evaluating results that are truly significant. And the one things missing from evidence labelled "anecdotal" are control groups. An important element of non-anecdotal evidence is looking for a differential between two courses of action. Let's say I feed each of my 11-year-old triplets a pesto of oak leaves every morning for a year. After that year they are all about 2.5 inches taller.

Anecdotally, oak pesto promotes growth. (and a tendency to hide acorns in bedding). Obviously, that probably isn't the case. The kids were very likely to get taller in that year regardless of what you did with them. Did they get taller than they otherwise would have? Maybe they actually grew less than they already would have and the actual impact of oak pesto is the opposite of the observation.

So, if you were trying to do a non-anecdotal study of the impact of oak pesto on growth in pre-adolescent children, ideally you would take something approaching these steps:

1) You would only give oak pesto to some of the children. You'd have a control group that receives no oak pesto so you have something to compare to.

2) You'd try to otherwise make the pesto-receiving group and the non-pesto receiving group as identical as possible. If you, for example, split the group based on gender this would likely introduce confounding factors since girls that age are more likely to see a large growth spurt than boys.

3) You'd blind the children as to whether they were eating oak pesto or not. So, all the children would consume something that looked like oak pesto and tasted like oak pesto but only half of it would actually be oak pesto. This prevents the children from behaving differently in ways that might affect growth (perhaps, kniwng that they are receiving what may be a growth proponent and being focused on the idea of growing taller they subconsciously start drinking more milk).

4) You'd blind the researcher so that they wouldn't know which children were receiving oak pesto and which were receiving the faux-pesto. So a different person would prepare the dishes than gives it to the kids. The real pesto always goes in Bowl A, but the person who puts it in Bowl A doesn't know which kid eats Bowl A. The person who gives Bowl A to the kid doesn't know whether it has real pesto in it or not. This prevents the researchers from treating differently the kids receiving what they hope is a growth proponent (researchers always want their theory to be right, or they probably wouldn't persue it; so perhaps they'd subconsciously give the pesto-kids larger portions of their other food items).

5) You'd pre-determine what is a significant result. This isn't actually done anew with each experiment but there are standard statistical models. If, after a year of this study the pesto group, on average, saw an additional 0.73 inches of growth? Is this significant? You can't say based on just that number (but many pseudoscientists do; they don't care a whit for confidence intervals). Sample size and measurement method is important here. If there were only 10 kids in the study that difference could be generated by a single child experiencing a very large growth spurt that had nothing to do with pesto. If there were 500 kids in the sample than random distribution of natural growth has less impact. Also, what is the method used to measure height and how accurate and precise is it? With growth it is easy to be very precide, but on a different topic, such as perception of pain, measurement is accurate only for the moment (the exact same pain might feel like an 8 today but tomorrow when you learned you just became an aunt it may only feel like a 5) and precision is impossible (nobody is going to say, today the pain is a 7.352). These factors introduce the margin of error to the result. So, a 0.73 inch differential with only ten participants probably isn't significant and a 0.73 inch differential when the margin of error is plus or minus 0.75 inches also isn't significant. But it is important to predetermine what measure of significance is going to be used because the natural humna instinct is to grasp any indication of significance to support the year (or frequently in large scale medical studies many years) of work you put into the examination.

To summarize more bluntly: No, proper scientific research is not essentially anecdotal in nature.

Yes, there is a lot of anecdote in our scientific understanding because proper double-blinding and other techniques for removing the anecdotal element are frequently impossible or unethical. This particularly can make it difficult to create properly homogenous cohort groups.

However, when done properly, these confounding factors are reported up front, everybody does their best to minimize them, and the results are understood to be fuzzy. This is why sometimes it feels like you get contradictory health information every week. Study A find a minor heart health benefit to Vitamin Whatever and the media is all over it, not properly passing along the hedges that are likely in the paper about to be published. And then Study B finds that Vitamin Whatever increases the chances of cancer, and the media is all over it, not properly passing along the hedges that are likely in the paper about to be published. And certainly they won't attempt to do any kind of analysis of relative risk. Is a 2% reduction in heart disease risk outweighed by a 4.5% increase in the risk of melanoma?

I think some things work and I can't explain why. We know little about thee body, even less about energy of any sort, and almost nothing about the mind. If you compare what we don't know to what we do, we're practically in the dark ages.

This is what I don't get. We know so incredibly much more than we used to, but because we don't know everything all things are essentially equally believable.

To think our current knowledge is essentially the same as in the dark ages is to not show a proper understanding of just how little we knew about the body back then (when people weren't exactly sure of what role the heart played, it was believed that each sperm contained a full miniature person just waiting to grow, and the shape of your bowel movement indicated your future).

I am just baffled that one can look at the last 100 years of medical advancement and doubt that it has been vastly more effective than all the ancient alternative methods combined.

Yes, we've gone too far in depersonalizing medicine and that turns a lot of people off. But I'll without hesitation take the quality of life offered by modern Western medicine over the quality of life that has been historically provided by alternative medicines.