Should you Trust the RDA?

Get our “getting started” mini-course:

About calories

It’s very convenient to talk about the “percent of recommended daily intake” instead of absolute amounts of nutrients. It provides one easily understandable scale for how high a food is in a certain nutrient, relative to how much you need. If we didn’t have the “% DV” category on nutrition labels, you’d have to remember how much you need of every single separate nutrient, and then get really good at doing percentage math – and that’s an incredible pain in the neck. Simplifying everything into a percentage gives you one simple number that everyone can understand and use.

This makes the %DV a very useful tool. But while there’s no question that it’s handy, there’s an increasing amount of evidence that it’s not much better than an educated guess.


To understand what’s really going on with the RDA and %DV, you have to know a little terminology. Unfortunately, the terminology surrounding nutrient recommendations is convoluted and often out of date – and full of abbreviations that all sound the same! Essentially, there are two completely different sets of numbers. Set #1 is produced by the Institute of Medicine. These numbers are reasonably precise because they’re broken down by age and sex, and they’re updated fairly often. Those numbers give us:

  • Recommended Dietary Allowance (RDA): an estimation of how much of a nutrient you need, based on your age, sex, and pregnancy status.
  • Adequate Intake (AI): how much of a nutrient you need to avoid an obvious deficiency disease.
  • Tolerable Upper Limit (UL): how much of a nutrient you can safely take without overdose.

All of these three numbers are released in reports called DRIs (for Dietary Reference Intakes). You can take a look at the latest DRI report here.

Set #2 is produced by the FDA. This set is much more simplified and hasn’t been updated since 1968!

  • Reference Daily Intake/Recommended Daily Intake (RDI): a simplified estimation of nutrient needs based on a 2,000 calorie diet that gives just one number for everyone regardless of age, sex, or pregnancy.
  • Percent Daily Value (DV): How much of a given nutrient is in a food, relative to the RDI. This is the number you see on nutrition labels, because those labels are regulated by the FDA, not the Institute of Medicine. In case your headache wasn’t bad enough yet, this used to be called the “US RDA” but they changed the name to make it less easily confused with the actual RDA.

From this, it should be pretty clear that the “%DV” column on nutrition labels leaves a lot to be desired.

Nutrient Needs Are Different

To start with the obvious, the %DV is extremely imprecise because it doesn’t account even for very broad categories like age and sex. It’s supposed to account for everyone ages 4 and up – as if a 5-year-old child and a 40-year-old man will have anything like the same nutrient requirements. If you take a look at the Institute of Medicine’s numbers (which do distinguish between men and women, and between adults and children), you’ll see for example that:

  • Men need significantly more zinc, magnesium, Vitamin A, and Vitamin C than women do.
  • Men need significantly less iron and calcium than women do.
  • Pregnant and breastfeeding women in particular need a lot more B vitamins than either men or non-pregnant women. And pregnant women need almost twice as much iron as anyone else.

Obviously, if you try to make just on number for all these different groups, it’s going to over- or under-estimate someone’s requirements most of the time. It’s clear to see why we like just one number (imagine how huge and complicated nutrition labels would be if they had all the categories of age and sex. You’d need a separate fold-out sheet for everything you bought!), but we’re sacrificing a lot of precision for that convenience.

Even the RDAs Aren’t Perfect

So the %DV isn’t perfect. But what if we took the Institute of Medicine’s numbers – the ones that get updated regularly and actually break the population down into groups? Are they any better?

Yes, but they’re not perfect. They’re much more precise than the %DV, and likely to be a little more accurate. But there are still some problems with the RDAs:

The groups are still huge

For example, take the category “Males, 19-30 years old.” Within that category, you could have:

  • An extremely fit, active, 22-year-old soldier on active duty.
  • His 23-year-old brother the couch potato, who weighs 350 pounds, is taking several different medications for his Type 2 Diabetes, and spends all day playing video games.
  • Their older brother, age 25, who is struggling with a chronic autoimmune disease and desperately trying to gain weight because he’s only 120 pounds at 5’8”.

These three men have vastly different nutrient requirements. One standard set of guidelines cannot possibly account for all of them.

Nutrient needs depend on other nutrients

Vitamin D helps you absorb calcium, so if you’re deficient in Vitamin D, your calcium needs will be greater. Vitamin C helps you absorb iron, so if you eat your iron with some Vitamin C, you’ll need to eat a smaller total amount. There are many other pairs of cofactors like this, and the RDA doesn’t account for any of them.

Nutrient needs depend on your environment.

The most obvious case is Vitamin D. In the winter, when you’re not get as much through sun exposure, you can make up for this with diet. But the RDA fails to account for this at all, giving just one number that you’re presumably supposed to use year-round.

Another example is Vitamin C. Smokers and people who get a lot of secondhand smoke need more Vitamin C than the general population. So for example, if you work as a host in a smoky bar, or a bartender in a smoky nightclub, you might need more Vitamin C than the average person even though you don’t smoke yourself. Again, the RDA fails to take this into account.

The numbers are for preventing deficiency

This might be an even more damning criticism. The numbers in the RDA report are designed to prevent obvious deficiency, not to achieve optimal health. So if you follow their recommendation for Vitamin D, you won’t get rickets; if you follow the recommendation for Vitamin B1, you won’t get beriberi. It’s definitely better not to have these diseases than to have them, but just the fact that you don’t have rickets or beriberi doesn’t make you healthy.

So how can you tell how much of a nutrient you really need? For a lot of them, we don’t actually know. There’s reasonably decent evidence that the RDA for Vitamin C, Vitamin D, and iodine might be too low – but it’s also possible that some of the RDAs (like calcium) are too high. Try to do a lot of research in this area, and you’ll keep bumping up against a wall of question marks.

Break Free from the Nutrient Mindset

The astonishing imprecision of all these supposedly “objective” numbers should be giving you some serious pause for thought about whether it’s at all useful to be counting nutrients in the first place. Hold onto that thought; it’ll do you a lot more good than driving yourself crazy trying to get “precise” nutrient counts and “accurate” numbers for what you really need.

The truth is that if your diet consists entirely – or even predominately – of nutrient-dense whole foods, you will meet almost all the RDAs, RDIs, %DVs, AIs, and all the other acronyms in the acronym soup without counting anything. The constant fixation on milligrams of this and micrograms of that makes it seem like you need to get precisely X amount of Y nutrient every single day, but that’s just not the case. You can get a little more some days, and a little less on others.

Realizing how imprecise the nutrient estimations are should also give you a healthy sense of doubt about taking a lot of supplements. If we don’t even have a rock-solid foundation for these recommendations in the first place, why would you spend a lot of money on vitamin supplements to meet a number that’s essentially imaginary? The vast majority of people will be fine if they simply eat real food, go out in the sun regularly, and get plenty of sleep.


The RDA (the numbers broken down by age and sex) can give us a reasonably decent estimation of nutritional needs for the average person (which you may or may not be) eating an average diet (which you may or may not eat). The %DV is even more imprecise, but even this number can provide a very rough idea of whether a food is high or low in a certain nutrient. That’s why you’ll keep seeing references to the “RDA” or “%DV” on this site: it’s not because you should be obsessively counting these things; it’s just to give you a very rough idea of how much nutrition is actually in a food. Don’t read 15% as 15%; read it as “reasonably good source.” Don’t read 80% as 80%; read it as “very high.”

This gives you all the information you really need on Paleo: a rough approximation of how nutritious a food is, and a general guideline for what foods to eat if you’re looking for a specific nutrient. It’s really just for your information, to help you choose the most overall nutritious staples, and to help the rare people who really do have a deficiency and need some advice about fixing it. If you aren’t suffering from a specific deficiency issue, it falls into the category of “nice to know,” not “crucial to memorize.” Human beings did perfectly fine long before we had food scientists around to tell us we needed so many milligrams of potassium or iron in our diet, and we can continue to do just fine without getting worked up over meeting some oversimplified standard of what we “should be getting.”

P.S. Have a look at Paleo Restart, our 30-day program. It has the tools to let you reset your body, lose weight and start feeling great.

+ Paleo Leap Tribe is now also available. It's a set of tools that we've built to make Paleo work for you. It has a meal plan generator, a weight loss tracker & tons of visual cheat sheets.

Get a PDF with our top 35 recipes