Subscribe to The Post Most newsletter for the most important and interesting stories from The Washington Post.
Ever notice how nutrition advice keeps changing?
Yeah, don’t answer that.
I’ve been writing about how our diet affects our health for longer than I care to calculate. I’ve watched low-fat come and go. I’ve watched low-carb morph to paleo, and then keto, and then carnivore. I’ve watched people eat according to their blood type or food’s pH. And now we’re saddled with dietary guidelines that double down on meat, saturated fat and protein.
Nobody can get a straight answer to the simple question: What should I eat?
There’s a reason for that. Turns out, that simple question is very hard to study. Most evidence is lousy, and yields inconsistent answers. You can find some kind of support for almost every diet theory going.
There are, however, three issues where I think we have a robust body of reasonably reliable evidence. While there’s no such thing as absolute certainty when it comes to the complexities of nutrition and the human body, I am willing to say we’re pretty sure about these three things.
What the evidence for all three has in common is that it comes from controlled trials, and not from trying to study people in the wild. (If you come here often, you know I don’t believe observational data has the power to suss out most effects.)
Here goes.
1. Dietary fiber is good.
Let’s take the least controversial one first. The observational research that I’m not going to cite nearly always shows that people who eat a lot of fiber have better health outcomes, but fiber is a component of plants (animal foods have no fiber), and plants also have vitamins, minerals and other compounds that can affect health. So, is it the fiber or is it the phytochemicals?
You can answer that with trials, where you give some people fiber supplements and others placebos. And, in trial after trial – and the meta-analyses that aggregate those results – fiber shows benefits. It can help improve glycemic control, relieve constipation and alleviate symptoms of irritable bowel syndrome. It can help manage kidney disease, and even help a little with weight loss. It can reduce cholesterol, and one meta-analysis showed that adding a psyllium supplement to statins was the equivalent of doubling the statin dose. It doesn’t seem to help with depression and anxiety, but you can’t have everything.
There are nuances here. Some fiber is soluble, some is insoluble, and they have different benefits and effects. But the details matter less than the well-supported conclusion that we should all eat fiber-rich foods, and even consider taking a fiber supplement.
Bottom line: Fiber is good. Get more of it.
2. Salt raises blood pressure, which increases the risk of stroke and heart disease.
Nearly everyone agrees that a high level of salt intake – about 5 grams of sodium per day – increases blood pressure and disease risk. There is, however, disagreement about a low level. Should we aim for about 2.3 grams per day, the U.S. recommendation, or is the American average of about 3.6 grams per day actually optimal?
Michael Jacobson, former head of the nonprofit Center for Science in the Public Interest, tackled the question in his 2020 book “Salt Wars,” and I talked to him about it. The evidence is clear, he said: “We should reduce sodium to 2 to 2.3 grams.”
The contrarian position, he told me, maintains that the lowest risk corresponds to an intake of 3 to 5 grams per day, putting U.S. consumption squarely in the sweet spot. That position hangs its hat primarily on a large study called the Prospective Urban Rural Epidemiology study (PURE), done across 18 countries, that correlated urinary sodium with cardiovascular disease and stroke. That study found that both low- and high-sodium intakes were associated with increased risk.
Arguments about the study quickly devolve into the ins and outs of urine sampling. Is it spot urine (basically, a random sample)? Or first-thing-in-the-morning fasting urine? Or 24-hour total urine (the most accurate)? But the larger problem is that the PURE study is observational. Sure, the sodium measurements are inaccurate, but even if they weren’t, we would be left with insurmountable confounding issues.
When I told Jacobson that I had looked at the evidence, and the preponderance of it supported lowering sodium, he took issue. “‘Preponderance’ makes it sound like it’s 60/40,” he said. “There’s virtually nothing that contradicts the notion that we should reduce sodium.”
We have a robust body of trial data, and the vast majority points to the fact that cutting sodium lowers blood pressure, and lowering blood pressure decreases the risk for stroke and heart disease.
Some of the sodium disagreements are about levels below about 2 grams per day, and I don’t think there’s enough evidence to address that. Jacobson told me that lower levels are of academic interest, but they are not relevant to public health. “To get to 2.3 grams would be a major miracle,” he said, and he – and I – will happily take that. We can talk about going lower when we get to that point, by which time perhaps we’ll have better evidence.
Susceptibility to hypertension has a genetic component; some people have low blood pressure almost no matter what they eat. (I am lucky enough to be one of them.) I asked Jacobson whether people like me have to worry about salt. “When they’re old, no,” he said. (Okay, I’ll put that down as one of the very few compensations for aging.) “When they’re young, it’s important to keep that level low, as it tends to increase as we age.”
Bottom line: Less salt is better, particularly if you have high blood pressure or it runs in your family.
3. Most saturated fat raises heart disease risk.
Here we are again with saturated fat! I’ve tackled this before, and nothing has changed. Like the evidence about salt, the sat fat evidence I find persuasive is a one-two punch of results from clinical trials: Most saturated fat raises LDL cholesterol, and LDL is causal for heart disease.
The tallow-beef-butter crowd looks primarily at observational research and a handful of trials that were done a half-century ago. Because the observational research is, well, observational, and the trials are equivocal (and some wouldn’t pass muster in today’s scientific landscape), I find the evidence for the sat fat/LDL link to be much more compelling.
Bottom line: Limit saturated fat to the long-standing recommendation of 10 percent of your calories.
When it comes to the connection between diet and overall health, those three – fiber, salt and saturated fat – are virtually the only issues I think we have compelling enough evidence to have some certainty about. (There may be some other recommendations to treat specific diseases, such as a ketogenic diet for epilepsy.) Not beyond-a-shadow-of-a-doubt certainty, but enough that it makes sense to work those conclusions into your diet. Other than that, the best hedge against uncertainty is lots of different foods that still have their nutrients intact.
When I first started writing about nutrition, back in the Pleistocene Era, I used to say I could tell you everything important about nutrition in less than a minute. The more I learned about nutrition science and its problems, the shorter my spiel became.
Now it’s one sentence with three caveats: Eat a wide variety of whole-ish foods you enjoy, in quantities consistent with the weight you want to be. Get lots of fiber, and don’t overdo the salt or sat fat.
That’s the best advice nutrition science can give us, which means I’ve got my spiel down to about eight seconds.
Related Content
