17 replies to this topic

[stocks]

Epidemiology -- Correlation Does Not Prove Causation

Wet streets don't cause rain.

Epidemiological studies look at associations between things. However, just because one thing is associated with another, this doesn’t mean this one thing is causing the other.
This is because of things called confounding factors – perhaps true causative factors that are associated with other factors that are ‘innocent bystanders’

Michael Crichton:

You open the newspaper to an article on some subject you know well. In my case, show business.
You read the article and see the journalist has absolutely no understanding of either the facts or the issues. Often, the article is so wrong it actually presents the story backward—
reversing cause and effect. I call these the "wet streets cause rain" stories. Paper's full of them.
In any case, you read with exasperation or amusement the multiple errors in a story, and then turn the page to national or international affairs, and read as if the
rest of the newspaper was somehow more accurate about Palestine than the baloney you just read


http://www.drbriffa....rove-causation/

Edited by stocks, 02 June 2012 - 09:34 AM.

[Rogerdodger]

"wet streets cause rain" and McDonald's causes fat people. (No other explanations are viable.)

[MaryAM]

"Often, the article is so wrong it actually presents the story backward— reversing cause and effect." Like whenever the weather turns cold, large numbers of turkeys die. Mary Anne

[mss]

"Often, the article is so wrong it actually presents the story backward—
reversing cause and effect."

Like whenever the weather turns cold, large numbers of turkeys die.
Mary Anne

So do, deer, rabbits, ----------

[stocks]

Relative Risk: Making a Mountain out of a Molehill

Both of these can be true:

Healthy people taking statins reduce their heart attach risk by 33%.
Healthy people taking statins have a 99% chance of receiving no benefit.

Reductions in relative risk don’t matter much when overall risk is low. Let’s say your risk of having a heart attack over 5 years is 3 per cent. Let’s say a statin reduces that risk to 2 per cent. Relative risk has fallen by a third, right?
But the real reduction in risk (known as the ‘absolute risk reduction’) is just 1 per cent (3 per cent minus 2 per cent).

Suddenly, the firepower of statins looks less like a bazooka and more like a peashooter.


http://www.drbriffa....marketing-hype/

[kamakazeman]

"Suddenly, the firepower of statins looks less like a bazooka and more like a peashooter."

[stocks]

Risk Factors—An Idea Whose Time Has Come—and Gone

Despite more than 50 years of trying, investigators have been unable to identify a single factor that shows a correlation with any disease that even remotely approaches the degree of correlation between cigarette smoking and lung cancer.

(MR FIT) This was the largest and most ambitious primary prevention trial ever undertaken. It was a cooperative study carried out in 28 regional medical centers around the country and spanned a ten year period.

They cut their cholesterol intake by more than 40%, saturated fats by almost 30%, and total calories by more than 20%. They increased their consumption of polyunsaturated fats by about 33%.

So what did we get with all this time, effort, and money? Nothing “statistically significant”.

The Women’s Health Initiative and enrolled over 48,000 healthy women who were followed for 8 years. The cost of the study was over $400 million. The investigators were convinced that their diet would significantly lower the incidence of breast cancer, colon cancer, heart attacks, and strokes.

Much like MR FIT, the women in the intervention group were able to maintain much lower levels of fat and saturated fat in the diet and most consumed the amounts of fruits, vegetables and grains prescribed at the outset.

The result: at the end of 8 years follow up there was no difference between the intervention and control groups in the incidence of breast or colon cancer, heart attacks, or strokes.



http://junkscience.c...one/#more-71773

[stocks]

Publication Bias

World Health Organization says withholding results from clinical trials is unethical.

Phantom clinical trials are a well-known phenomenon in biomedical research. Trials are conducted, but the results never see the light of day. To tackle this, the World Health Organization (WHO) released a statement calling for full disclosure of clinical-trial results

A 2013 study in the British Medical Journal showed that out of 585 large clinical trials, 171 studies involving nearly 300,000 patients were still unpublished some 5 years after completion.

Joseph Hayes, a psychiatrist at University College London, suggested in a tweet that such a policy would uncover many less-than-spectacular results.


http://www.nature.co...-trials-1.17331