Sigh. Correlation. Causation. No mechanism is offered, and all they studied was respondent's reported diet. Nothing to see here, other than the fact that there is evidence that eating lots of fresh produce correlates well with longevity. Really? Thanks, Mr. Scientist! Do both of those things also correlate well with upper middle class?
The problem with people like you who have this Pavlovian conditioning of belching out "correlation != causation!!!!!1" whenever an observational study is mentioned which doesn't affirm their own preconceptions is that they seem to be incapable of understanding that it is practically infeasible to run randomized controlled trials to study the interactions between complex lifestyle factors such as diets and complex diseases such as cancers and CVDs, as you would have to somehow ensure sufficiently high study compliance in a sufficiently large sample over a sufficiently long period of time, i.e., decades; not to mention ethical implications.
What "people like me" do you mean? Those who don't forgive you for doing bad science just because good science is hard? Yeah, those guys will totally fuck up a neighbourhood.
By "people like you" I mean people like you who go to social bookmarking sites, don't bother to properly read the article, state the (in this case: scientifically) obvious, mix in some ignorance about the topic at hand and then lean back thinking they had contributed anything insightful. In short, by "people like you" I mean the peanut gallery.
They do say that they controlled for class and physical activity: but the only thing open-access about the article is the abstract, so I can't tell how they controlled for those things.
The controlled for them in the standard way, by adding them into the (Cox) regression. The full list of controls is:
"Other variables included in the analysis were age (35–44 years, 45–54 years, 55–64 years, 65–74 years and 75+ years), sex, smoking status (current smoker, ex-regular smoker, never smoked regularly) and social class of the head of household (manual, non-manual, other). Education (degree or equivalent qualification, other, no qualification), measured body mass index (BMI) (<20 kg/m2, 20–24.9 kg/m2, 25–29.9 kg/m2, ≥30 kg/m2), physical activity (as maximum activity intensity level in the past 4 weeks: inactive, light activity, moderate activity, vigorous activity) and alcohol consumption on heaviest drinking day in the previous week (non-drinker, drank within limits, drank above limits, drank more than twice the recommended daily limit) were also added to the models as indicated."
The most obvious thing missing from this list to me is all other dietary related variables. For example, people who like the taste of plants are probably far less likely to enjoy eating red meat, salty foods, and many other things that have been associated with increased mortality.
>For example, people who like the taste of plants are probably far less likely to enjoy eating red meat, salty foods, and many other things that have been associated with increased mortality.
Which, incidentally, have really only been associated with increased mortality through flawed epidemiological studies such as this one.
Flawed is a strong word. There is little doubt that meat and sodium intake are associated with a number of poor health outcomes. (Run almost any regression you can think of.) That is a much weaker statement than saying the former causes the latter, which is why responsible researchers are reluctant to do so unless there is very convincing evidence. Even if you are not convinced of the causal link, you would be well served to moderate your consumption of these foods on the chance that one exists.
>social class of the head of household (manual, non-manual, other). Education (degree or equivalent qualification, other, no qualification)
Social class is manual/non-manual? What do retail and telemarketing count as?
edit: I'm being a bit pithy here, but what I want to know is if someone who went to DeVry and works in telemarketing ends up higher class than me, a six-figure college dropout with six-figure college (Northwestern/Northern Illinois) graduate parents.
Social class is almost fully determined by the wealth of people's parents, why don't we use it?
The problem is that people have been saying that fruits and vegetables are good for you for a long time, so conscientious who exercise and go to the doctors and take the medicines they're prescribed will also disproportionately eat more fruits and vegetables. With health, the conventional wisdom is the hardest thing to test and is where you really need controlled trials, or at least natural experiments where people eat more or less vegetables for extrinsic reasons.
Mod to high SES is correlated with eating healthier food, we all get that. However, by convincing low SES to eat healthier, you can actually increase their SES. For example, if you eat better, you are healthier and therefore take less sick days at work. If you don't get paid for sick days, this means more money in pocket. Showing up for work more may increase your chance at a raise or promotion and therefore increase your SES.
So I think there is value in these studies. As well, food diaries are a standard measure of diet. Not ideal, but accepted standard.
I think SES is just a very poor proxy for IQ. Higher IQ people are a lot healthier than lower IQ people regardless of SES. Probably quite a bit of that is just genetic (mutational load lowers both IQ and general health) and diet details are secondary.
These observational studies are mostly worthless because they are not controlling for the right things.
> and all they studied was respondent's reported diet
If you read the actual journal article referenced [1], a number of other factors were included in the analysis, and the stated effects were after controlling for those factors (which included smoking, age, and social class, among other factors.)
A valid criticism. I was basing my response on the originally linked article from Science Daily, not on the cited study. Notably, the original article also doesn't make the egregious causation claim that irritated me so in the linked report.
