"A major new report, "Helping Doctors and Patients Make Sense of Health Statistics," in Psychological Science in the Public Interest, a journal of the Association for Psychological Science, shows that statistical illiteracy is a significant problem having widespread negative impact on healthcare and society. The authors of the report are an international and interdisciplinary team of psychologists — Gerd Gigerenzer and his colleagues Wolfgang Gaissmaier and Elke Kurz-Milcke at the Max Planck Institute for Human Development and Harding Center for Risk Literacy, Berlin, Germany — and physicians — Lisa M. Schwartz and Steven Woloshin at Dartmouth Medical School.

The problem of statistical illiteracy in health has two sides to it, according to Gigerenzer and his team. It is a combined problem caused by the misleading or confusing ways statistics are ordinarily presented in health communication, coupled with a lack of statistical thinking skills among consumers of health information.

The combination can be explosive. A mid-1990s report in Britain showed that new oral contraceptive pills posed a twofold (or 100%) increased risk of blood clots led to a mass panic among women, who opted to stop using this form of birth control as a result. There were an estimated 13,000 more abortions in England and Wales in the year following the report, due to the pill scare. Had the public been informed that the absolute risk with the new pills only increased from 1 in 7,000 women having a blood clot to 2 in 7,000, this panic would not have occurred, and the UK National Health Service would have been saved an estimated $70 million. "
http://www.physorg.com/news142876231.html

Are you confident that the methodology and data support the conclusion you posted? What countries were studied?

I totally agree!

Lets be careful though that we don’t paint the problem with too wide of a brush. We might overlook the two 'hypothesis' out of the hundred that are correct!

Thanks for the link above and yes, I did read all 9 pages!

Bo

Sorry, only the abstract information is available unless you’re a $subscribed$ member.

Bo

When they talk about controlling for variables such as these, it means that they were entered as covariates in the statistical analysis to try to isolate their effect on the outcome. Usually these factors are called confounders, since they confound the relationship between the variable of interest and the outcome. This can be a useful statistical tool, but it is definitely not realistic in some cases: for instance, I once saw a dataset where they had controlled for the effect of a child's age on height. Now, I don't know about you, but most children do not grow taller while their age remains the same--physics probably wouldn't allow it. I wouldn't call this voodoo, but it can sometimes leave something to be desired. I would agree that we need better statistical methods that would allow us to get at the relationship of interest without having to worry so much about confounding.

One other problem is that so-called 'residual confounding' may remain. It's possible that your measurement of the confounding factor is not precise enough to allow its effect to be removed from the analysis. Therefore, a residual effect may remain that would still be confounding the relationship of interest.

There is no use of Excel in this world--Excel does not have sophisticated statistical capabilities. SAS is commonly used, as are Stata, R, S-Plus, and SPSS. As for playing around with the numbers, any researcher worth their salt will have laid out an analysis plan for their study before they begin the analytical process. This prevents any urge to play around with the numbers until they give you the answer you were hoping for. I am afraid that a large number of the low-fat hypothesis researchers that Taubes describes were probably guilty of this after-the-fact manipulation but were never called out on it by anyone with the authority to enforce statistical rigor.



Yes, this is a big issue in statistics. This is know as the unmeasured confounders. This is something people worry about quite a bit, as there's no way to truly account for every possible confounding factor. That is part of the reason that ecological studies are so unreliable. Those are the studies where they compare the rates in different countries--like Ancel Keys's countries study. Since you are looking at population level factors, you cannot conclude anything at the level of the individual--you can only make generalizations at the population level, which are not terribly useful in a field like this. Ecological studies are really only good for hypothesis generation.




As an epidemiologist, I can say with confidence that observational studies can never prove causation. Randomized trials are a subset of epidemiologic studies, but the randomization, blinding, etc, if done properly, can begin to answer the causation question with more certainty. Now clinical trials can also be fraught with statistical issues if the studies were not designed or conducted properly, as we saw with the ACCORD study. People act as if all clinical trials are perfect, but this is far from the truth. There have been clinical trials whose conclusions were later proven wrong by cleaner trials that did not have as many issues with their methods.

However, observational studies are far less expensive and allow us to accumulate evidence for a particular theory that can then be tested rigorously in a randomized trial. Therefore, observational epidemiologic studies do have value. What often frustrates me is when the media, the public, heck even sometimes the researchers themselves, make conclusions about an observational study that are just not supported by the data. One statistical association does not a causation make.

Ok, getting off my soapbox now. Thanks for reading this far.

--Melissa

I actually think that there is a third side to statistical illiteracy. Many people who conduct health research, such as physicians and psychologists, have very little, if any, statistical training. Therefore, they often are guilty of making conclusions that their data does not support. Moreover, the editors of the journals in which they publish are also often statistically illiterate, and don't notice the leaps of logic in the discussion sections of manuscripts. Then the media misconstrue the meaning of the paper, which then confuses the consumer even further.

There is a huge huge massive difference between a relative risk and an absolute risk. Your relative risk of getting breast cancer in your 20s might be quadrupled if you live in an area with a lot of air pollution, but if your original risk was miniscule, then a quadrupling of that risk is still going to be small enough that it shouldn't concern you. These types of numbers can be very confusing to anyone who does not have statistical training, which is why I feel strongly that all researchers in the health field as well as all journalists who cover the field should receive statistical training. This would help prevent the type of mass panic that you discussed.

--Melissa

Infectious diseases and poverty related malnutrition affect the statistics on India, they therefore do not indicate anything regarding the long term health impact of a vegetarian diet.

http://www.country-data.com/cgi-bin/query/r-6000.html

Also, if the way a life expectancy is calculated is by adding up all the ages of death in a group, and then averaging the total, a high rate of infant mortality will bring the final figure down.

Here are some interesting new views of the globe reflecting different aspects of life/death
It covers such issues as Human poverty, War deaths in 2002, International emigrants,
It's interesting to see statistics represented as images.

I should also have said how much I enjoyed reading the Taubes article on epidemiology.

However given all the caveats and warnings from Taubes there will be people reading this thread who are concerned about their risk of Colorectal cancer and they may wish to read

Calcium, Dairy Foods, Vitamin D, and Colorectal Cancer Risk: The Fukuoka Colorectal Cancer Study A decreased risk of colorectal cancer associated with high calcium intake was observed among those who had higher levels of vitamin D intake or among those who had a greater chance of daily sunlight exposure, but not among those with medium or lower intake of vitamin D or among those with potentially decreased sunlight exposure. These results add to support for a joint action of calcium and vitamin D in the prevention of colorectal carcinogenesis.

The study was based in Fukuoka latitude 33 where walking to work will generate more incidental vitamin d through the year than walking to work in London. (If only they had tested 25(OH)D levels initially and through through the trial we would have some really useful and more reliable, information)
You should perhaps also remember that Calcium absorption is optimised when 25(OH)D status is above 80nmol/l (32ng) and some people still improve bone mineral density above 100nmol/l (40ng).

Thanks for the clarifications and explanations. I didn't mean to disparage the field, only to point out that its findings don't mean what some journalists or researchers THINK they mean. And I didn't mean to harp on the Excel business, which I got wrong. The point was only that the work is done on computer screens and not with patients or in labs.

As Taubes's article points out, epidemiology serves many critical functions, and has had some spectacular successes. It's just not good at telling us what to eat.

Oh, no offense taken. Epidemiology is not well known or understood by the general public, so I felt that the situation warranted some more info. But I definitely agree with you--epidemiology is important, but I would never make health behavior recommendations based on one epidemiologic study, no matter how compelling it may be. Especially recommendations that could possibly have unintended consequences for vast segments of the population.And that includes randomized trials. There are too many potential problems 'when clinical trials go bad' to base recommendations on one study that may be flawed. I think the CONSORT criteria are a step in the right direction, but all journals that publish these studies should require one of the reviewers to be a biostatistician--to root out issues before they're published, rather than leaving it to the judgment of the media and the public.

As for the computer screens vs. patients question, all epidemiologic studies must collect data at the patient level...but eventually that data needs to be analyzed using statistical software. A study is only as good as the accuracy and precision of the measurement tools used to collect patient data--if the tools are useless, then the results of the study will be largely useless, regardless of how much the study investigators wish or state otherwise. Garbage in, garbage out. Epidemiology has proved invaluable in many areas where the measurement tools are excellent--infectious disease is a good example.

The tools for measuring dietary intake, however, are extremely rudimentary at best. This is in large part why these studies tend to come up with hypotheses that are later disproven. Most studies that look at diet utilize questionnaires or 72-hour recall, which are notorious for recall bias since they rely on the accuracy of the participant's memory. I know how flawed my responses would be--I can barely remember what I ate yesterday, let alone what I ate for dinner a week ago.

The other related issue here is that most people are very imprecise when it comes to estimating the portion size of the food that they ate. Unless you break out a food scale every time you eat or have gained the ability to accurately eyeball your food from using one, this is an exceedingly difficult task, especially for someone who has not been keeping a food diary on Fitday for months like some of us. If you pulled a random person off the street and asked them how many grams were in the chicken breast in front of them, they would probably be off by at least 20% or more.

This makes it virtually impossible to run these studies with any degree of reliability. There is no pharmaceutical incentive to run these types of studies since the companies do not stand to benefit from an effective dietary regimen. Doing these studies correctly would be extremely expensive, therefore what we get instead is a patchwork of subpar studies with small numbers of participants. I would love if the studies that Taubes suggested were carried out with a commitment to quality--I just don't see it happening any time soon.

--Melissa

Melissa:

You're really good with this stuff. We need an epidemiology expert on here to explain -- with authority -- all the drivel that comes from the news, and university PR departments.

I nominate you for the job.

Set us straight.

Thanks Walt. I will try my best--we are just inundated with it. Even my poor mom got me a subscription to Health magazine once, which is just chock full of recommendations based on shoddy or questionable science at best. One study doth not a recommendation case make.

As of next week, I will be working at a medical center again--so I will have full text access to virtually every journal out there. That will help me get to the bottom of this.

--Melissa

I think, in short, too many people do not understand the difference between information that is bound to sell as a newspaper or magazine headline, or even sometimes a book, and information that can really make a positive difference to human health. Most of the people who buy the magazines, are so concerned about human health that it does not occur to them that all the publisher was doing was trying to get that magazine off the shelves. The trouble is, exciting hogswash is so much more likely to sell than boring proven truth.

Diet of worms protects against bowel cancer

REGULAR doses of worms really do rid people of inflammatory bowel disease. The first trials of the treatment have been a success, and a drinkable concoction containing thousands of pig whipworm eggs could soon be launched in Europe.

The product will be called TSO, short for Trichuris suis ova, and will be made by a new German company called BioCure, whose sister company BioMonde sells leeches and maggots for treating wounds. Chief executive Detlev Goj says he expects sales of TSO in Europe will start in May, after approval by the European Agency for the Evaluation of Medicinal Products. The agency would not comment. The pig whipworm was chosen as it does not survive very long in people. Patients would have to take TSO around twice a month. The human whipworm, which infects half a billion people, can occasionally cause problems such as anemia.

The latest trials, carried out in the US, involved 100 people with ulcerative colitis and 100 with Crohn's disease, both incurable and potentially serious diseases collectively known as inflammatory bowel disease. In many of the volunteers the symptoms of IBD- such as abdominal pain, bleeding and diarrhea- disappeared. The remission rate was 50 per cent for ulcerative colitis and 70 per cent for Crohn's, says gastroenterologist Joel Weinstock of the University of Iowa, who devised the treatment.

"A lot of researchers couldn't believe this treatment was effective, but people are always skeptical when confronted with new ideas," Weinstock says. He will announce the results in May at a conference in New Orleans, and full details will soon be published. "With our new impressive results, we can come out of the closet," he says.

The trials follow the success of a pilot study, revealed by New Scientist. Weinstock came up with the idea of using worms to treat IBD after noticing that the sharp rise in the disease over the past 50 years in western countries coincided with a fall in infections by parasites such as roundworms and human whipworms. IBD is still rare in developing countries where parasitic infections remain common.

Weinstock's theory is that our immune systems have evolved to cope with the presence of such parasites, and can become overactive without them.

Author: Frank van Kolfshooten
New Scientist

Finally, a resolution of our low carb and red meat bowel cancer concerns.

You’ve just gotta love science!

Bo

You may want to post this in the IBS forum in case they haven't seen it.

--Melissa