A few weeks ago,
I wanted to know whether saturated fat (the kind that makes up most of
mammal and bird fat and tropical oils like coconut and palm oil) was
better or worse for you than unsaturated fat (the kind that makes up
most of fish oil and non-tropical oils of vegetable origin). At this
point the evidence strongly suggests that, at least if you live in the
rich world, saturated fat is pretty bad for you, and unsaturated fat
is much less bad for you (not counting trans fats, which almost don't occur naturally and are very bad for you indeed, but were abundant in margarine until they were outlawed).
The Cochrane Collaboration is kind of the gold standard for medical
research review studies. In my search I ran across this recent
open-access Cochrane review of only randomized clinical trials:
Hooper L, Martin N, Jimoh OF, Kirk C, Foster E, Abdelhamid AS.
Reduction in saturated fat intake for cardiovascular disease.
Cochrane Database of Systematic Reviews 2020, Issue 8. Art. No.:
CD011737. DOI: 10.1002/14651858.CD011737.pub3.
They conclude that over the four or so years in the 16 RCTs they
reviewed, the decreases in all-cause mortality and cardiovascular
mortality due to lower saturated fat intake, if any, were too small to
detect with the usual p-values, but their measure of combined
cardiovascular events did decrease significantly, with a RR CI from
0.70 to 0.98.
On the face of it, this sounds kind of optimistic for the thesis that
saturated fat is fine: are heart attacks really that bad if they don't
increase your risk of dying? Maybe there was some countervailing
health benefit, like you're more likely to die of a heart attack but
less likely to die of cancer or an accident. And they have some
quotes with a remarkably low degree of hedging: "Critical importance.
Reducing saturated fat intake probably makes little or no difference
to cardiovascular mortality."
But I don't think it's actually that optimistic, for three reasons.
The numbers I've looked at so far don't suggest an actual protective
effect; the average death rates and cardiovascular death rates in the
intervention groups (with lower saturated fat) were still lower, just
not significantly so. Presumably part of the reason for that is that,
almost necessarily, fewer experimental subjects had a heart attack and
died from it than merely had a heart attack. But that would only
account for wider confidence intervals.
I think the stronger argument is that the studies were only for (an
average of) four years or so, and usually it takes about 50 years for
cardiovascular disease to kill you on a standard unhealthy "Western"
diet (and this is an aspect of "Westernization" we didn't get from the
Song Dynasty). So it's really pretty alarming that lowering saturated
fat intake significantly reduces the "combined cardiovascular event"
rate over only four years. Presumably there is progressive damage (or
recovery) if you continue the intervention for ten or twenty years.
The third argument is that a stroke that doesn't kill you can still be
pretty bad.
I'm not any kind of expert on public health (that would be my ex-wife the epidemiologist) or on nutrition, so I could easily be overlooking something important. In fact, I overlook important things in fields where I am an expert all the time. But I don't think Jeff Nobbs is an expert either; he's a startup founder and restaurateur, and this post seems to be a sales pitch for his vegetable-oil-substitution business (according to https://www.jeffnobbs.com/about). Searching in Google Scholar finds literally zero publications by him in any area, especially including nutrition. And in https://news.ycombinator.com/item?id=34283915 it looks like Nifty3929 caught him saying that one of the studies said something very different from what it actually did say. So I think I'll trust the Cochrane Collaboration instead.
Heart disease in America is driven by hyperinsulinemia, which is driven by overconsumption of refined carbs and sugar. Hyperinsulinemia precedes diabetes by 10-20 years. If you want to prevent heart disease, stop eating refined carbs and sugar. And get a glucose insulin tolerance test. This is the test to get - as something like an a1c tells you NOTHING about your insulin resistance until you break past the thresholds.
i don't think sugar consumption is higher now in the usa than it was a century ago, when obesity and metabolic syndrome were rare (though statistics for america as a whole are hard to come by, since most countries in america didn't have anything like a public health system a century ago), and certainly very low compared to the hadza
by 'refined carbs' do you mean white flour
because white flour consumption was also pretty high in the usa a century ago
regardless, the topic at hand is the health effects of different kinds of fats, and it turns out that different kinds of fats do seem to have significant health effects (despite what effects other macronutrients like carbohydrate may or may not have) and that those effects run in the opposite direction to what nobbs is proposing
The Cochrane Collaboration is kind of the gold standard for medical research review studies. In my search I ran across this recent open-access Cochrane review of only randomized clinical trials:
Hooper L, Martin N, Jimoh OF, Kirk C, Foster E, Abdelhamid AS. Reduction in saturated fat intake for cardiovascular disease. Cochrane Database of Systematic Reviews 2020, Issue 8. Art. No.: CD011737. DOI: 10.1002/14651858.CD011737.pub3.
<https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD...>
They conclude that over the four or so years in the 16 RCTs they reviewed, the decreases in all-cause mortality and cardiovascular mortality due to lower saturated fat intake, if any, were too small to detect with the usual p-values, but their measure of combined cardiovascular events did decrease significantly, with a RR CI from 0.70 to 0.98.
On the face of it, this sounds kind of optimistic for the thesis that saturated fat is fine: are heart attacks really that bad if they don't increase your risk of dying? Maybe there was some countervailing health benefit, like you're more likely to die of a heart attack but less likely to die of cancer or an accident. And they have some quotes with a remarkably low degree of hedging: "Critical importance. Reducing saturated fat intake probably makes little or no difference to cardiovascular mortality."
But I don't think it's actually that optimistic, for three reasons. The numbers I've looked at so far don't suggest an actual protective effect; the average death rates and cardiovascular death rates in the intervention groups (with lower saturated fat) were still lower, just not significantly so. Presumably part of the reason for that is that, almost necessarily, fewer experimental subjects had a heart attack and died from it than merely had a heart attack. But that would only account for wider confidence intervals.
I think the stronger argument is that the studies were only for (an average of) four years or so, and usually it takes about 50 years for cardiovascular disease to kill you on a standard unhealthy "Western" diet (and this is an aspect of "Westernization" we didn't get from the Song Dynasty). So it's really pretty alarming that lowering saturated fat intake significantly reduces the "combined cardiovascular event" rate over only four years. Presumably there is progressive damage (or recovery) if you continue the intervention for ten or twenty years.
The third argument is that a stroke that doesn't kill you can still be pretty bad.
I'm not any kind of expert on public health (that would be my ex-wife the epidemiologist) or on nutrition, so I could easily be overlooking something important. In fact, I overlook important things in fields where I am an expert all the time. But I don't think Jeff Nobbs is an expert either; he's a startup founder and restaurateur, and this post seems to be a sales pitch for his vegetable-oil-substitution business (according to https://www.jeffnobbs.com/about). Searching in Google Scholar finds literally zero publications by him in any area, especially including nutrition. And in https://news.ycombinator.com/item?id=34283915 it looks like Nifty3929 caught him saying that one of the studies said something very different from what it actually did say. So I think I'll trust the Cochrane Collaboration instead.