Linda,
I just don't think you get it. A scientific theory was founded on a bogus piece of research by Ancel Keys. Richard Smith summarised what happened, but you can find corroborative accounts all over the internet. Senator McGovern also leant his weight behind the saturated fat hypothesis, and the whole thing became political.
By the time Jacob Yerushalmy and Herman Hilleboe showed that the graph of heart disease against saturated fat intake was deliberately cherry picked, the bandwagon was already rolling. Richard Smith catalogued the way that the opportunities to stop the bandwagon were repeatedly missed.
When I said "start at the bottom", I thought you would understand that I meant start with the original piece of faulty research - that is where this whole theory came from. Have you even viewed Ancel Key's graph and the way he cherry picked the data. For example, he left out France because it had (and I think still has) low levels of CVD and high saturated fat consumption.
What we are talking about - and what I think Richard Smith is talking about - is a form of corruption of science. Once an idea has caught hold - even if it turns out to be based on fraud - a strong confirmation bias then operates, driven by the fact that large institutions are using the theory. It is obviously going to be far easier to bend the rules slightly to confirm what you 'know' to be true, than to face the enormous brouhaha which would follow a volte-face on an issue like saturated fat.
By quoting the latest research without examining the roots of the mistake, you are tainted with that same confirmation bias. There are innumerable ways in which diet/health studies can be distorted - Richard Smith commented on exactly that. My favourite - which you didn't comment on - is that people who live an otherwise healthy lifestyle will tend to eat whatever is perceived as the most healthy diet. That bias will always favour status-quo science.
Corporations and governments can handle awkward news from scientists, just so long as it stays the same, but they don't like scientists who change their mind.
You of all people should be able to see the myriad of ways that this can happen.
Yes! And that is the point of foregoing crappy evidence (an ecological study, like Keys', is crappy evidence no matter how meticulously bias-free its performance) in favour of depending upon good-quality evidence, like RCTs. "Quality" means that the stuff you mentioned - confirmation bias, confounders, personal preferences and even political interference - are specifically avoided. That is, the way the research is designed, implemented and analyzed obviates the ability of those factors to alter the results.
The current recommendations depend on an evidence-based review, which means that the evidence is weighed by the extent to which confirmation bias, confounders, personal preferences, political interference, and even fraud could alter the results. This is what you seem to be asking for, but then you turn around and reject the process.
So I ask you again, given that all these factors tend to come into play for any idea someone comes up with, how would you suggest we evaluate whether or not an idea is legitimate, if we do as you say and abandon medicine's evidence-based approach (where ideas are evaluated on the basis of whether or not these factors are in play)?
Anyway, I am getting bored with this, if it has made you the slightest bit worried that medical science might have made a mistake, why not contact Richard Smith. If he tells you his article was a joke, well the laugh is on me, but if he fills you in with far more information to back up his case..............
David
I couldn't care less what Richard Smith has to say. You are the one who hung their hat on the article. If this is about you trying to save face (which you don't need to do for my sake), then you contact him. I'm interested in the story from the perspective of the process of sorting ideas on the basis of reliability and validity. But it's a story about what happens when that process isn't in place, not when it is.
There seems to be this perception that science operates in some sort of pristine environment, shut off from the ordinary influences of humankind, so that any time there is an example of someone behaving badly, or even hastily, it spells the end of science, as though research depends upon that pristine environment. In actuality, the messy, biased, adversely motivated environment in which it is performed is well-known by those working within it. The only people who seem to think that science is pure is everyone but the scientists. The methods used don't depend upon everyone behaving well. The methods used have been developed so they work regardless of whether someone behaves badly.
Linda
Last edited: