BBC Article: Bike Now to Avoid Senility

http://news.bbc.co.uk/2/hi/health/4304822.stm

EFR
Ile de France

Elisa Francesca Roselli wrote:
http://news.bbc.co.uk/2/hi/health/4304822.stm


Yet another article filled with bad science. I don't doubt there is a
correllation between exercise and lowered risk of dementia. But they
failed to establish a true cause and effect relationship. This is a
typical ecological fallacy.

The true cause of the lowered risk may be eating habits. Perhaps it is
more frequent washing (exercisers get stinky and probably wash more
often). Maybe it is that exercisers have a lower exposure to chemical
propellants because they are less concerned with their state of their
hair - it's going to get messed up anyway, right? The point is that
there are a lot of differences between exercisers and non-exercisers
that are not accounted for in the article. If the researchers did
attempt to control for this, it certainly isn't presented in the bbc
article. If so, then this is just another case of the public press
butchering good science because they don't know how to report it
properly.

Here's a good article on ecological fallacy:
http://www.stanford.edu/class/ed260/freedman549.pdf

-Buck

Didn't look like bad science to me - anyway the science is in the
research paper whereas this is just a journalistic report. It would
seem that the conclusions are hedged with 'mays' and 'mights' and the
conclusion is that 'more research is needed'. You don't need to prove a
cause and effect to establish that something may be hazardous or
beneficial.
Anyway I'm going to keep on pedalling even if it is too late in my
case.

cheers

Jacob

wrote:
Didn't look like bad science to me - anyway the science is in the
research paper whereas this is just a journalistic report. It would
seem that the conclusions are hedged with 'mays' and 'mights' and the
conclusion is that 'more research is needed'. You don't need to prove a
cause and effect to establish that something may be hazardous or
beneficial.

Agreed.

I understand Buck's point, but there's no need to put the article down
so harshly. One might say "Hmmm. Perhaps this needs further study."
But if we immediately reject every study (or every report of a study)
that doesn't account for every confounding variable, science will slow
to a crawl.

I think the way these things generally go is this: Someone notices a
correlation like the one mentioned in the article, and publishes it.
Someone else notes what Buck did - that there may be other factors at
work - and questions it. A third team may refine the work done by the
first, controlling other variables in various ways. That process is
repeated, with new knowledge gained each time.

At some point, there's enough confidence in the knowledge to recommend
for or against the original idea, of course taking benefits and
detriments into account.

On this particular issue, I figure the benefits of exercise are many.
This _may_ be just one more. The detriments are few. I see no harm in
mentioning this possible benefit, at least until it's disproven.

- Frank Krygowski

Buck wrote:
Elisa Francesca Roselli wrote:
http://news.bbc.co.uk/2/hi/health/4304822.stm


Yet another article filled with bad science. I don't doubt there is a
correllation between exercise and lowered risk of dementia. But they
failed to establish a true cause and effect relationship. This is a
typical ecological fallacy.

Have you read the paper yet? It is interesting though my quick skim
suggests some worries about the stats. I'll have to go back and look
more closely.


The true cause of the lowered risk may be eating habits. Perhaps it is
more frequent washing (exercisers get stinky and probably wash more
often). Maybe it is that exercisers have a lower exposure to chemical
propellants because they are less concerned with their state of their
hair - it's going to get messed up anyway, right? The point is that
there are a lot of differences between exercisers and non-exercisers
that are not accounted for in the article.

Try the Lancet Neurology article. They appear to be controlling for a
few variables of interest.

From the Lancet Neurology pre-pub
"In brief, the baseline survey procedures included a self-administered
questionnaire on health behaviour, health status, and medical history.
Participants' blood pressure, height, and weight were measured, and
body-mass index (BMI) was calculated. A venous blood sample was taken
to determine serum cholesterol concentrations.13 In addition, the
presence of various locomotor disorders was determined. [...]



If the researchers did
attempt to control for this, it certainly isn't presented in the bbc
article. If so, then this is just another case of the public press
butchering good science because they don't know how to report it
properly.

Here's a good article on ecological fallacy:
http://www.stanford.edu/class/ed260/freedman549.pdf

-Buck

John Kane
Kingston ON

General Systems Theory Beginning with Wholes

General Systems Theory Beginning with Wholes

wrote:
wrote:
Didn't look like bad science to me - anyway the science is in the
research paper whereas this is just a journalistic report. It would
seem that the conclusions are hedged with 'mays' and 'mights' and the
conclusion is that 'more research is needed'. You don't need to prove a
cause and effect to establish that something may be hazardous or
beneficial.

Agreed.

I understand Buck's point, but there's no need to put the article down
so harshly. One might say "Hmmm. Perhaps this needs further study."
But if we immediately reject every study (or every report of a study)
that doesn't account for every confounding variable, science will slow
to a crawl.

I think the way these things generally go is this: Someone notices a
correlation like the one mentioned in the article, and publishes it.
Someone else notes what Buck did - that there may be other factors at
work - and questions it. A third team may refine the work done by the
first, controlling other variables in various ways. That process is
repeated, with new knowledge gained each time.

At some point, there's enough confidence in the knowledge to recommend
for or against the original idea, of course taking benefits and
detriments into account.

On this particular issue, I figure the benefits of exercise are many.
This _may_ be just one more. The detriments are few. I see no harm in
mentioning this possible benefit, at least until it's disproven.

- Frank Krygowski

Good points but the first thing to do is to read the actual journal
article rather than pre-judging on the basis of a quick media report. I
suspect that even most BBC science reporters are not Ph.Ds or MD in
the field

For those with access to the on-line version of The Lancet Neurology
the paper is available in the pre-pubs category

The abstract (taken from the on-line pre-pub.)reads
---------------------------------------------------------------------
Background
Physical activity may help maintain cognitive function and decrease
dementia risk, but epidemiological findings remain controversial. The
aim of our study was to investigate the association between
leisure-time physical activity at midlife and the subsequent
development of dementia and Alzheimer's disease (AD).
Methods

Participants were randomly selected from the survivors of a
population-based cohort previously surveyed in 1972, 1977, 1982, or
1987. 1449 persons (72·5%) age 65-79 years participated in the
re-examination in 1998 (mean follow-up, 21 years). 117 persons had
dementia and 76 had AD. Multiple logistic regression methods were used
to analyse the association between leisure-time physical activity and
dementia or AD.
Findings

Leisure-time physical activity at midlife at least twice a week was
associated with a reduced risk of dementia and AD (odds ratio [OR]
0·48 [95% CI 0·25-0·91] and 0·38 [0·17-0·85], respectively),
even after adjustments for age, sex, education, follow-up time,
locomotor disorders, APOE genotype, vascular disorders, smoking, and
alcohol drinking. The associations were more pronounced among the APOE
var epsilon4 carriers.
Interpretation

Leisure-time physical activity at midlife is associated with a
decreased risk of dementia and AD later in life. Regular physical
activity may reduce the risk or delay the onset of dementia and AD,
especially among genetically susceptible individuals.
----------------------------------------------------------------------------

They do seem to be doing a lot of chi-square and t-tests though .
I'll have to look at it more closely.
John Kane
Kingston ON Canada

Dunno why you didn't just go to the original Lancet article to see if
it's legit, but through the magic of the internet, here's a link for
those who are interested:

http://www.sciencedirect.com/science.../sdarticle.pdf




-JR

wrote:

Good points but the first thing to do is to read the actual journal
article rather than pre-judging on the basis of a quick media report. I
suspect that even most BBC science reporters are not Ph.Ds or MD in
the field

For those with access to the on-line version of The Lancet Neurology
the paper is available in the pre-pubs category

The abstract (taken from the on-line pre-pub.)reads
---------------------------------------------------------------------

Unfortunately the article is pay-for-view (while it is pre-pub?) As the
research was undertaken in Sweden I would be interested to know if the
form of regular exercise was analysed in any detail. Sweden has a higher
proportion of orienteers than many other countries. As physical exercise
goes it requires a fair degree of problem solving, thus using the brain
as much as the heart and lungs.

Colin