5. Not the Full Story

TV Kills was the headline that raced around the world as a result of this study.

As New Scientist pointed out:

The asked 8800 people about their health, lifestyle and television watching behaviour, and then followed them over the next six years, during which time 284 of them died. Among people who spent more than 4 hours a day in front of the TV, it found, the risk of their dying within the period of the study was 46 per cent higher than among those who watched less than 2 hours a day.

The error is confusing CORRELATION with CAUSATION. The risk might involve the sitting rather than TV. Or it could be that those who sit in front of TV longer are not well.

THIS IS A COMMON AND DANGEROUS ERROR INVOLVING HEALTH STATISTICS

Now that we have more screening tools we must be more cautious.

Angelia Jolie had her breasts removed because of testing positive for the BRCA 1 and BRCA 2 genes, which are linked to an increased risk of Breast Cancer.

This is her decision. Anyone facing breast cancer would seriously consider this option.

BUT, be wary if you face this decision.

Think of Asthma. If we started our research into asthma looking for asthma genes we would have found them.

Genes create a pre-disposition for Asthma (hence the correlation) but they do not cause asthma.  The dust mite, pollen, cat, dairy and other allergies cause asthma. If researchers had concentrated on genes alone our knowledge of asthma would be limited.

 Another study showed that 80% of prisoners in Australia smoke. Isn’t it obvious? Smoking causes criminal behaviour!!!!!

Lies, Damned Lies and Breast Surgery

Angelina Jolie has had both breasts removed because of the risk of developing breast cancer.

But did she understand the Maths?

She may be right. This could be the best way for her to avoid breast cancer. And it IS her decision.

But maths isn’t that clear cut. Referenced information about BRCA1 and BRCA2: Cancer Risk and Genetic Testing can be found at The National Cancer Institute here.

More information from the UK NHS here.

The estimates of lifetime risk are about

12.0 percent of women (120 out of 1,000)

in the general population will develop breast cancer sometime during their lives compared with about

60 percent of women (600 out of 1,000)

with gene mutation in BRCA1 or BRCA2. This is FIVE TIMES the risk over a lifetime.

BUT and this is a big but …..

It is important to note, however, according to The National Cancer Institute (USA), most

……………………………………………………

research related to BRCA1 and BRCA2 has been done on large families with many individuals affected by cancer and the cancer risk associated with BRCA1 and BRCA2 mutations have been calculated from studies of these families. Because family members

……………………………………….

share a proportion of their genes and, often, their environment, it is possible that the large number of cancer cases seen in these families may be due in part to other genetic or environmental factors.

Factors that could affect these risk factors include food, lifestyle, location, inherited temperament and even the air they breathe.

Statistics are a useful tool. But statistics are based on random selection. This is IMPORTANT. Once you bias the data, your results are corrupted.

Removing cancerous breast tissue makes sense. Removing perfectly healthy breast tissue based on the maths … well … you’d want to think about it.

To put this another way the BRCA genes may be correlated with breast cancer, but this event is not independent of all other events in one family. There may be many, yet to be discovered, links . This is a common error in statistics made not only by journalists, but by experts too. (See below)

……………………………………………………………………………………….

Conviction by Maths Error

On 9 November 1998 at Chester Crown Court Sally Clark, a Cheshire solicitor, was convicted, by 10-2 majority, of smothering her two baby boys.

……………………………………….

Clark’s first son died suddenly within a few weeks of his birth in 1996. In 1998, when her second son died in similar circumstances she was arrested and tried for the murder of both sons.

………………………………………………..

The prosecution used paedeatrician Prof Roy Meadows as a expert witness. He had discovered Munchausen Syndrome by Proxy (MSbP). Sally Clark was found guilty and spent 3 years in jail.

Prof Roy Meadows testified that the chance of two children from an affluent family suffering sudden infant death syndrome was 1 in 73 million. He arrived at this number by squaring 1 in 8500 for likelihood of a cot death in similar circumstances.

……………………………………………………….

He was wrong and was later struck off the medical registrar. He assumed that these were independent events. They were not. Maybe the heater was at fault. Or the cot. Or the house paint. Or, as suggested in The Observer article, faulty genes in the family.

Sally Clark died several years later of alcoholic poisoning. More information.