Register | Sign In


Understanding through Discussion


EvC Forum active members: 65 (9164 total)
3 online now:
Newest Member: ChatGPT
Post Volume: Total: 916,910 Year: 4,167/9,624 Month: 1,038/974 Week: 365/286 Day: 8/13 Hour: 1/0


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Sequel Thread To Holistic Doctors, and medicine
Percy
Member
Posts: 22505
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


Message 263 of 307 (427384)
10-11-2007 10:23 AM
Reply to: Message 262 by purpledawn
10-11-2007 7:46 AM


Re: Statistical Bias
The problem is not anecdotal data. When anecdotal data is all you have, then that's what you go with. Or when testing says that Midol works for about x% of women, then if you try it and find it doesn't work for you, then that's what you go with. Tylenol doesn't work for me, and there didn't need to be a double-blind placebo-based study for me to figure that out. But I wouldn't even consider Tylenol if it were a folk remedy instead of a carefully studied drug with regard to efficacy and safety.
But how to assess the available evidence? Ah, there's the rub! Though most people don't realize this, people in general are weak at informally assessing evidence except of the simplest kind, like whether Midol works for you. When the dataset is just yourself in the present, you can usually make a pretty good determination. But one can't measure longterm effects on oneself, or reach valid conclusions from a varied dataset, say of reports from friends. How many people have had a restaurant or a movie highly recommended by friends, only to discover they didn't like it at all?
And so the real problem is not anecdotal data, which is known to be unreliable but in some circumstances it is all you have, but of rejecting available scientific data in favor of anecdotal data.
Anecdotal reports should always be considered with a cake of salt. At a minimum you want reliable reports by trained observers (say, a doctor's experience in his practice), and ideally you want carefully conducted scientific studies. And you definitely should be skeptical of advice from someone who doesn't understand this, which would be most people.
Here's an example of the problem. You have two medications to consider, each recommended by a friend. One friend raves about his medication, saying how quickly it worked and how wonderful it made him feel. The other friend says his medication worked for him, but only after a week and only for the specific symptoms, and that there are possible side-effects, and that there were a number of other medications that it shouldn't be combined with, and that seeing a doctor before making a decision might be a good idea.
Which medication would you try? The first one, right? But the advice you received might say much more about your friends than it does about the medications. You'd have to take their reliability into account, too, and you can see how quickly it becomes very subjective, especially given placebo effects, common wisdom, cultural background information about known folk remedies, etc.
Bottom line: if there's scientifically valid information, go with that.
But even with scientific information there's a problem. Both you and LindaLou have been very successful at finding scientific views contrary to the mainstream, both in technical papers and, in LindaLou's case, also much older work, such as Breggin et. el., a book from 1912, testimony before Congress in 1962, etc. Those who want to believe something generally have no problem finding support for it, and experience tells us that talking someone out of a false belief is usually just not possible.
A different spin on the same problem is to put greater weight on evidence that agrees with what you already believe. Christopher Hitchens put it well when he recently said in an interview on the Skeptics Guide to the Universe that (paraphrasing), "Objectivity means that in a confrontation with the evidence I'm willing to change my mind." Naturally when he refers to evidence, he means all the evidence.
--Percy

This message is a reply to:
 Message 262 by purpledawn, posted 10-11-2007 7:46 AM purpledawn has replied

Replies to this message:
 Message 264 by Kitsune, posted 10-11-2007 10:49 AM Percy has replied
 Message 269 by purpledawn, posted 10-11-2007 1:27 PM Percy has replied

Percy
Member
Posts: 22505
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


Message 265 of 307 (427411)
10-11-2007 11:37 AM
Reply to: Message 264 by Kitsune
10-11-2007 10:49 AM


Re: Statistical Bias
LindaLou writes:
experience tells us that talking someone out of a false belief is usually just not possible.
No chance of persuading you of anything then Percy
But don't miss the main point. While one take on what I said is simply that people can be difficult to persuade, the fuller context points to significant differences in the types of evidence cited by the mainstream versus the fringe. For you, rejection by the mainstream is a plus, e.g., Breggin. Data from any era is significant if it agrees with you, such as your citations from 1962 and 1912. Data inconsistent with your view is minimized, such as the historical information I provided about epidemics that are now a thing of the past. The motives of those in the mainstream are questioned.
A scientific approach would stress the best and latest data as measured by how well well-established scientific protocols were followed, and would leave out subjective issues such as motive or whether the findings are consistent with what is already believed.
This is the approach taken to arrive at the current scientific consensus, which exists because scientists judged which were the best studies and accepted their findings. Fringe views are fringe because few scientists judge the studies supporting them to be of sufficient quality.
--Percy

This message is a reply to:
 Message 264 by Kitsune, posted 10-11-2007 10:49 AM Kitsune has not replied

Percy
Member
Posts: 22505
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


Message 270 of 307 (427462)
10-11-2007 2:57 PM
Reply to: Message 269 by purpledawn
10-11-2007 1:27 PM


Re: Statistical Bias
Since you're making a similar misinterpretation as LindaLou, I can see I didn't word things clearly enough.
It isn't an issue of natural versus non-natural. And it's not an issue of how do you persuade people.
The issue is the selection and interpretation of evidence. Those determined to believe something will always be able to select and interpret evidence in supportive ways. Those infected by the "true believer" disease can rarely be convinced that their beliefs are coloring their selection and interpretation of evidence.
So what are the clues that one might have this disease? Well, citing data from 1912 is a good clue. Concluding that being part of the consensus is bad is another one. Calling people's motives into question is another good one. Avoiding mainstream peer-reviewed journals is another one.
Basically, if you're not seeking the most recent and highest quality data based upon how well established scientific protocols have been followed, then you're not really exercising good scientific judgement.
In the entire history of quackery, there is not a single instance of any of it ever moving into the scientific mainstream. ESP is a great example. The few early ESP studies of quality indicated that perhaps it was possible that there was an actual phenomenon lurking on the very edge of detectability, but after at least 50 years of study and after huge improvements in experimental and statistical approaches, ESP studies can still only indicate that perhaps there is a phenomenon lurking on the very edge of detectability. And of course, only true believers actually believe this. Most scientists conclude that all ESP research has succeeded in doing is demonstrating that the phenomenon does not exist. The Princeton Engineering Anomalies Research laboratory (PEAR) closed its doors this year after 28 years of futility (A Princeton Lab on ESP Plans to Close Its Doors).
And so it is with alternative medicine. Believers will continue to believe that if science would just study the claims of alternative medicine properly that it would be found valid, despite that the quality studies at best indicate that there may perhaps be an effect lurking around the edges of detectability.
And the same is true of the anti-vaccination crowd. No matter how many studies demonstrate that vaccinations don't cause autism or whatever the claim might be, they'll continue to believe that when the science is done right they'll find the effect.
And the true believers might be right! After all, science is tentative and will always change its views in light of new evidence or improved insight. But history is against them. It just doesn't happen. A hundred years from now there will still be believers in UFOs and ESP and Bigfoot and alternative medicine, just as there were a hundred years ago.
--Percy
Edited by Percy, : Missing "not".

This message is a reply to:
 Message 269 by purpledawn, posted 10-11-2007 1:27 PM purpledawn has replied

Replies to this message:
 Message 271 by purpledawn, posted 10-11-2007 3:19 PM Percy has replied
 Message 272 by Kitsune, posted 10-11-2007 3:48 PM Percy has replied

Percy
Member
Posts: 22505
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


Message 274 of 307 (427526)
10-11-2007 8:51 PM
Reply to: Message 253 by Percy
10-09-2007 3:50 PM


Re: Staying alive to 100: Living longer, living stronger
Today I went to Doctor Jane Sullivan-Durand's presentation titled, "Staying alive to 100: Living longer, living stronger". Because the announcement about the presentation mentioned "Nutritional, Behavioral, and/or Alternative & Complementary therapies" I assumed this would be a presentation from the perspective of alternative medicine. It was anything but.
I entered the presentation prepared to exert much energy on self-control, but I needn't have been concerned. This doctor didn't say anything from woo-woo land. The conclusions she presented were that decreased caloric intake combined with a diet high in vegetables and soy products and with no meat (but fish is fine, nobody asked about chicken or turkey) appeared to contribute to extended life.
The were a couple minor things I thought were off. Someone asked about the percentages listed on vitamin supplements, like 400% RDA. She said that RDA (Recommended Daily Allowance) was just the minimum to prevent ill health effects, such as rickets for insufficient vitamin D. I thought the minimum was actually MDR (Minimum Daily Requirement). Maybe someone here knows what the story is here.
The other minor thing that seemed off was her answer to my question. I had decided to ask a question about vaccines, but since the talk was about longevity I didn't think a question about child vaccinations would be appropriate, so I asked, "Every winter another flu season arrives, and people are encouraged to take flu vaccines. As one gets older, is taking these vaccines a good idea?"
Her answer was excellent, I'll mention the one thing that seemed off last. She said that healthy people are perfectly capable of fighting off flu viruses, so there's no short term necessity for taking flu vaccines. People who should take these vaccines are those already in ill health, or who live with someone in ill health who shouldn't be exposed to potentially dangerous viruses, or who work in the health industry and wish to prevent infection because of the certainty that they will be exposed.
But even for very healthy people, suffering through a bout of the flu is not without danger. The problem with viruses is that they invade and take over the cell's reproductive machinery so that they can turn out copies of themselves. But not all virus invasions of cells are completely successful, and another common result is that the virus insinuates itself into the cell's DNA but fails to take over. In other words, infected individuals who suffer through a bout with the flu will have their DNA modified in some cells of their body. This is a type of mutation, and mutations are one contributor to shortened longevity.
Viral illnesses can be prevented by taking vaccines, but the vaccines themselves are not without risk. One of the dangers she mentioned is that injected vaccines are normally carried in a compound that contains mercury, and that mercury is suspected of being associated with increased autism in this country. She recommended using a nasal spray version of vaccines.
The mention of a possible link between vaccines and autism is the other thing that I thought might be off. I thought studies had shown no link. This has been discussed in this thread, and I thought the science side was pretty certain there was nothing to it. But this doctor seemed to have a lot on the ball. Was she wrong?
--Percy

This message is a reply to:
 Message 253 by Percy, posted 10-09-2007 3:50 PM Percy has not replied

Replies to this message:
 Message 277 by molbiogirl, posted 10-11-2007 10:13 PM Percy has not replied

Percy
Member
Posts: 22505
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


Message 280 of 307 (427610)
10-12-2007 8:29 AM
Reply to: Message 271 by purpledawn
10-11-2007 3:19 PM


Re: Statistical Bias
purpledawn writes:
The average person doesn't usually have access or doesn't know where to go. We can only interpret what we have access to.
Where specifically would you have us go. (Be nice. )
What is the highest quality data available to the average joe and his wife?
I can't even count the number of times when I decided that I'd take a spare half hour and investigate the scientific support for some claim, for instance the health benefits of megadoses of Vitamin C, and just thrown up my hands. I find tons of articles, and since I'm not familiar with the field I have to first investigate each and every journal to see if it's a legitimate peer-reviewed journal or just a refuge for fringe elements to publish low-quality papers. Going to each journal's website and reading their purpose and goals take time, and before I've even composed a list of articles to scan I'm out of time.
Someone like Molbiogirl who is much more familiar with the field would have a big head start on me. And I have a big head start on you. In other words, you haven't got a prayer, and I often don't have much of one, either.
So if you distrust the consensus, which has an absolutely remarkable record of success in the medical field over the past century, then all you've done is cast yourself adrift in a sea of charlatans. Laypeople researching on their own with the goal of forming their own conclusions are often on a fool's errand. It's one thing to want to know about a field, it's quite another to believe that having learned a little that you're qualified to draw conclusions.
As the presentation on longevity I attended yesterday indicates, there are legitimate medical doctors out there who integrate naturalistic approaches into their practice while not going overboard to the point of rejecting the findings of traditional medicine or advocating quackery.
--Percy

This message is a reply to:
 Message 271 by purpledawn, posted 10-11-2007 3:19 PM purpledawn has replied

Replies to this message:
 Message 293 by purpledawn, posted 10-12-2007 10:07 AM Percy has not replied

Percy
Member
Posts: 22505
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


Message 283 of 307 (427618)
10-12-2007 8:59 AM
Reply to: Message 272 by Kitsune
10-11-2007 3:48 PM


Re: Statistical Bias
LindaLou writes:
I thank you for helping to pin me to the ground a little more. But I'm never going to look at things 100% in a scientific way. There are times for that, and other times when it is perhaps more of a hinderance than a help.
The only time non-scientific approaches are preferred is when scientific data is unavailable and can't be made available within existing time constraints. In that case you go with what you got.
Scientific methods are far and away the best and most reliable way of ferreting out information about the real world. Oftentimes other methods will give you the same answers. For example, you hear a funny noise in your car, and to you it sounds like a valve click. You tell the mechanic you think you've got a valve click. He shrugs his shoulders, puts your car on a scientific analysis machine, then says, "Yep, you've got a valve click."
But when other methods give answers that don't agree with science, alarm bells should start going off in your head. For instance, say you took your car in, they put it on the analysis machine, then told you you had a valve click and that it will cost $1500 to rebuild the cylinder head. This is too much money, so you take it to another mechanic who's much cheaper but who has no analysis machine, but he listens to the car and tells you all you need is a new exhaust system. Time to back away slowly. I'd be backing away slowly because I trust the analysis machine, but it would also be valid to back away slowly simply to give yourself time to figure things out or to seek another opinion.
In other words, when there's little to no congruence in the conclusions reached by different methods, the best answer is not to choose the one you prefer. The best answer is to figure out why there's a difference. And in the case of anecdotal evidence versus the scientific consensus underpinned by the best studies available, the reason for the difference is obvious: anecdotal approaches are an extremely poor way of figuring out complex relationships.
--Percy

This message is a reply to:
 Message 272 by Kitsune, posted 10-11-2007 3:48 PM Kitsune has replied

Replies to this message:
 Message 284 by Kitsune, posted 10-12-2007 9:08 AM Percy has replied

Percy
Member
Posts: 22505
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


Message 285 of 307 (427622)
10-12-2007 9:13 AM
Reply to: Message 281 by Kitsune
10-12-2007 8:42 AM


Re: Statistical Bias
You know what scientists think about Bigfoot?
They think it would be incredibly neat if Bigfoot actually existed. Even better, it would be an enormous boon to both anthropology and the search for human origins because of what it would tell us to not only discover a previously unknown hominid, but a living one at that!!!!
You know what else scientists think about Bigfoot?
That there's no reliable scientific evidence supporting its existence.
What scientists object to about Bigfoot is not the idea that he might exist. What they object to is the bogus evidence and interpretations of evidence used to reach the conclusion that Bigfoot exists.
Bigfoot as a possibility? Fine. Unlikely, but fine.
Reliable evidence exists for Bigfoot? No way.
That's how scientists see it.
It's the same for alternative medicine.
Alternative medicine as a possible source of viable medical approaches? Fine. Unlikely, but fine.
Reliable evidence exists for the efficacy and safety of alternative medical practices? No way.
That's how scientists see it.
--Percy

This message is a reply to:
 Message 281 by Kitsune, posted 10-12-2007 8:42 AM Kitsune has not replied

Percy
Member
Posts: 22505
From: New Hampshire
Joined: 12-23-2000
Member Rating: 4.9


Message 286 of 307 (427623)
10-12-2007 9:28 AM
Reply to: Message 284 by Kitsune
10-12-2007 9:08 AM


Re: Statistical Bias
LindaLou writes:
Not going to happen, so matter how many lessons in skepticism people give me here. The scientific consensus underpinned by the best studies available can still be wrong, and I believe it is in many different ways.
No one is claiming that science can't be wrong. But the answer is not to return to methods proven to be even more wrong.
As has been said here many times, you're not wrong to use your own experience to decide your course of action for entire mainstream medical treatment. Where you're wrong is to generalize your experience into a condemnation and rejection of the medical establishment from drug companies to research facilities to practitioners. In the case of depression you're accepting only the evidence of those with negative experiences that correspond to your own, and rejecting the evidence of those that traditional medicine has helped because you would be unlikely to hear from them.
Those like my wife who have tried anti-depressants and found them to work (and work so incredibly well it's like magic) do not go to websites and complain. How many times do I have to tell you that valid conclusions do not derive from the experiences of a self-selected group who, even worse, are self-analyzed. If there are a thousand women at such websites, but 15 million women in the US take anti-depressants, how can you believe that that tiny self-selected subset can tell you anything reliable about the experience of the other 15 million women?
Let's assume your experience and the experience of all the other women at anti-depression websites are real, and that there are many more who don't go to websites, and that the total is 15,000 (I'm trying to be generous). That's only .1% of the 15 million total. If the ill effects are real but only one in a thousand, that doesn't justify rejection of the entire traditional medical establishment.
What you're doing is just incredibly wrongheaded.
--Percy

This message is a reply to:
 Message 284 by Kitsune, posted 10-12-2007 9:08 AM Kitsune has not replied

Newer Topic | Older Topic
Jump to:


Copyright 2001-2023 by EvC Forum, All Rights Reserved

™ Version 4.2
Innovative software from Qwixotic © 2024