"PK" <
[email protected]> wrote in message
news:[email protected]...
>
http://depts.washington.edu/hiprc/childinjury/topic/bicycles/index.htm
>
http://depts.washington.edu/hiprc/childinjury/topic/bicycles/helmeteffect.htm
> perhaps someone more familiar with all the data can set this in context
for
> me.
The context is pretty simple. This is a list of reports and meta-analyses, most of which can be
traced back to the original series of reports by Thompson, Rivara and Thompson and others. It
includes the original discredited 88% figure. It includes small case-control studies, and excludes
whole population and time series data. As usual with such meta-analyses it adds nothing to the body
of knowledge save the fact that the authors are not prepared to consider the reasons why the claimed
benefits are never exhibited at the population level.
And I would be disinclined to characterise this as "different" since I think most of these studies
have been analysed on <url:http://www.cyclehelmets.org>.
If you want to understand the weakness of this model, consider the following: analysis of casualty
data from emergency rooms indicates that Kevlar vests reduce the incidence of knife wounds.
Therefore Kevlar vests should be compulsory.
Note the flaws:
1. The solution is not proportional to the actual risk
2. The solution does not address the source of risk, only it's after-effects
3. The solution will not help those whose wounds are on parts of the body not covered by the vests
4. The solution ignores the possibility that people wearing Kevlar vests will confront knife-
wielding assailants rather than run, so increasing their level of risk
5. The solution ignores the possibility that the behaviour of knife-wielding assailants will change
for the worse
6. The solution enshrines in law the idea that the source of risk cannot be reduced
There may be other problems, these are just the ones which occur to me regarding the hospital-
centred case-control studies on which helmet promotion is predicated.
The whole helmet compulsion thing is close to being a my-dog-is-a-cat scenario, as per Yes Minister,
except for the crucial fact that in this case nothing need actually be done - or at least, not at
the victim level.
--
Guy
===
WARNING: may contain traces of irony. Contents may settle after posting.
http://www.chapmancentral.co.uk