A man picks up a multivitamin and turns the bottle over. Iron, 18 milligrams, 100% of the Daily Value. One hundred percent sounds like exactly the right amount.
According to the NIH Office of Dietary Supplements, the recommended intake for an adult man is 8 milligrams a day, at every adult age. He is looking at more than twice his recommended intake, and the label is completely correct.
Nothing on that panel is wrong. It was built to a federal regulation, 21 CFR 101.36, and a fully compliant panel can leave out more than most shoppers expect. The useful skill is knowing which questions the panel was never designed to answer, and there are more of them than you would guess.
The daily value is not a recommended amount
The Daily Value is a labeling number. Its legal source is the Reference Daily Intake table in 21 CFR 101.9, which has four population columns: adults and children four years and older, infants one through twelve months, children one through three years, and pregnant and lactating women. Most products on a general shelf use the first column. One number per nutrient, for everybody over four.
That single number has to stand in for a whole table of recommended intakes that vary by age and sex, and in the cases you can check, it lands at the top of the adult range.
Magnesium is the clean example. FDA’s Daily Value for magnesium is 420 mg. The Office of Dietary Supplements lists recommended intakes of 400 mg for men aged 19 to 30, 420 mg for men 31 and older, 310 mg for women 19 to 30 and 320 mg for women 31 and older. The Daily Value matches the highest adult figure on the table, which is a man over 30. A woman in her forties reading 100% of the Daily Value is looking at about 130 percent of her own recommended intake.
Iron runs the other way, which is why the opening example works. The Daily Value is 18 mg, matching the recommended intake for women aged 19 to 50. For men at any age and for women 51 and older, ODS lists 8 mg. Same label, same percentage, two completely different relationships to the published figure.
None of this is a defect. A label cannot ask who is reading it. What it means in practice is that the percentage on the panel is a comparison to a fixed reference, not a measurement of your needs, and that the direction of the error depends on who you are.
The numbers themselves are newer than most people realize. They come from an FDA final rule titled “Food Labeling: Revision of the Nutrition and Supplement Facts Labels,” published 27 May 2016 and effective 26 July 2016. The title names the supplement panel, and the rule revised the reference values that drive the percentages on both panels, though the coverage at the time treated it almost entirely as a food-label story.
Date this carefully if you ever read the rule yourself. Its own text sets compliance dates of July 2018 and July 2019. FDA later extended them: to 1 January 2020 for manufacturers with $10 million or more in annual food sales, and to 1 January 2021 for smaller ones. Anyone reading the 2016 rule alone gets the wrong years.
One more thing about FDA’s consumer page on Daily Values, which is dated as current to 5 March 2024. It describes Daily Values as recommended amounts to consume or not to exceed each day, and it never mentions the Recommended Dietary Allowance or the Dietary Reference Intakes at all. The distinction this section is built on does not appear on the page most readers would land on. Article 322 covers the recommended-intake system, and article 323 covers the ceilings, which are a third set of numbers again.
Serving size comes from the manufacturer, not from a survey
Here is where the two panels diverge most sharply, and the mechanism is more interesting than the result.
Section 101.36(b)(1)(i) says the serving size shall be determined in accordance with sections 101.9(b) and 101.12(b), Table 2. Follow that cross-reference to Table 2 and look up the row for dietary supplements. The reference amount given there is “The maximum amount recommended, as appropriate, on the label” for consumption per eating occasion, and where the label carries no recommendation, one unit, such as a tablet, capsule, packet or teaspoonful.
Now read the paragraph just above the table, 101.12(a)(1), which explains where the numbers for conventional foods came from. FDA calculated those reference amounts for people four years and older to reflect the amount of food customarily consumed per eating occasion by people in that group, based on data in national food consumption surveys.
The same regulation sends one category to survey data about what Americans eat, and sends the other category back to the label it is being printed on. FDA’s own labeling guide states the supplement rule the same way: one serving equals the maximum amount recommended on the label.
The practical consequence shows up every time anyone compares two bottles. If one product calls a serving two capsules and another calls it one, the amounts in the panel are not comparable and neither are the percentages, even if the capsules are identical. A “servings per container” count can also be set against a serving the manufacturer chose. Checking the serving line before reading anything else below it is not fussiness. It is the only way the rest of the panel means anything.
What the panel must show, and what it may leave out
Section 101.36(b)(2) covers the familiar nutrients that have a reference value: calories, fats, cholesterol, sodium, carbohydrate, fiber, sugars, added sugars, protein, vitamin D, calcium, iron and potassium. Those must be declared when they are present in amounts above the level that could be declared as zero on a food label. Then comes the part that surprises people: ingredients that are not present, or are present in amounts that can be declared as zero, shall not be declared. Leaving them off is not a manufacturer’s choice. It is required, and the regulation offers amounts corresponding to less than 2 percent of the Reference Daily Intake for vitamins and minerals as its example.
That is the opposite of how the food panel works, and FDA says so directly in its labeling guide: on a Supplement Facts panel you are not permitted to list zero amounts. On a Nutrition Facts panel, zeros must be declared. The guide also says you must list dietary ingredients without reference values in a Supplement Facts panel, and that you are not permitted to list those same ingredients in a Nutrition Facts panel for foods.
Hold those two inversions together and you get the single most useful thing in this article. A Nutrition Facts panel tells you what is absent. A Supplement Facts panel does not. A nutrient missing from a supplement panel might be genuinely absent, or present below the declarable threshold, or present as something that belongs in a different list further down the label. Absence from the panel carries almost no information.
“Daily Value not established” is about the science, not the product
Section 101.36 requires an asterisk, or a similar symbol, in the percentage column for ingredients that have no reference value, pointing to a statement at the bottom of the panel reading that the Daily Value is not established. The same mechanism applies to a proprietary blend‘s total weight.
It means FDA has set no Reference Daily Intake or Daily Reference Value for that substance, so no percentage can be calculated. It says nothing about whether the ingredient works, whether the amount is reasonable, or whether it is safe.
Notice which lines carry the asterisk. On most products the asterisked entries are the botanicals and novel compounds the product is being sold on, while the lines with an actual percentage are the ordinary vitamins and minerals. The most-read string on the panel attaches itself precisely to the ingredients a buyer most wants context for.
“Other ingredients” lives in a different regulation
This one trips up people who go looking in the right spirit. There is no “other ingredients” paragraph in 101.36. The requirement sits in 21 CFR 101.4(g).
That paragraph puts the ingredient list immediately below the nutrition label, or contiguous and to the right of it if there is not enough room below, preceded by the word “Ingredients,” with one exception: if some ingredients, meaning sources, were already identified inside the nutrition label under 101.36(d), then the list outside the panel is headed “Other ingredients” instead.
The word “other” is doing real regulatory work. It means other than the ones already named upstairs. The same paragraph says that excipients, fillers, artificial colors, artificial sweeteners, flavors and binders belong in that list, and section 101.4(a)(1) requires ingredients to be listed by common or usual name in descending order of predominance by weight.
How a proprietary blend stays legal while hiding the proportions
Section 101.36(c) is short, and it describes a mechanism worth understanding exactly rather than condemning vaguely.
A proprietary blend goes into the list of dietary ingredients identified by the term “Proprietary Blend” or another appropriately descriptive term or fanciful name, and it may be highlighted in bold. Any ingredient in the blend that has a reference value still has to be declared individually, with its amount, under the ordinary rule. The rest, the ones with no Daily Value, are listed in descending order of predominance by weight, in a column or in a line, indented under the blend name.
Then paragraph (c)(3) does the work: the quantitative amount by weight given for the blend is the total weight of all those other dietary ingredients, placed on the same line as the blend name, with the asterisk pointing to the statement that the Daily Value is not established.
So you learn the order and you learn one total. You do not learn the proportions. The first ingredient listed could be most of the blend or a small fraction of it, and the panel is fully compliant either way. This is permitted design, not a violation and not a loophole, and it is why two products with identical blend names and identical total weights can be entirely different products.
The disclaimer, and what puts it there
Most people read the sentence at the bottom of a supplement label as FDA distancing itself from the product. It is narrower than that, and the exact wording is set by regulation.
Under 21 CFR 101.93(c)(1), where a label carries one such claim, the required disclaimer reads: “This statement has not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.”
Where there is more than one claim, paragraph (c)(2) gives a choice: each statement can carry the singular disclaimer, or one plural version can cover them all, changing “This statement” to “These statements” and leaving the second sentence alone. That substitution is the detail people reproduce wrongly from memory.
Placement is also specified. The disclaimer must sit next to the statement with nothing in between, or be linked to it with a matching symbol such as an asterisk, and where it is not adjacent to the statement it must be set off in a box.
What triggers it is a structure or function claim: a statement describing the role of a nutrient or dietary ingredient in affecting the structure or function of the human body, which the regulation permits provided it is not a disease claim. The disclaimer travels with the claim, not with the product. A bottle making no claims at all does not need it. Article 810 covers the line between structure and function claims and disease claims, which is where most of the interesting argument happens.
The background to all of this is on FDA’s own question-and-answer page, current as of 27 August 2026: under DSHEA, FDA does not have the authority to approve dietary supplements before they are marketed.
The number on the label and the number in the capsule
There is one more gap, and it is measurable. Federal scientists have been chemically analyzing supplements and comparing results against label claims for years through the Dietary Supplement Ingredient Database, a USDA and NIH project.
A 2016 analysis in the American Journal of Clinical Nutrition by Andrews and colleagues bought adult multivitamins on a national sampling plan and assayed them. The mean measured content of every ingredient except thiamin came out above the labeled amount. Predicted average differences ran 1.5 to 13 percent above label for copper, manganese, magnesium, niacin, phosphorus, potassium, folic acid, riboflavin and vitamins B12, C and E, and about 25 percent above label for selenium and iodine.
A 2018 paper in the Journal of Nutrition by the same group put it in population terms. Overages above 20 percent of label levels were predicted for several nutrients in 50 to 99 percent of the multivitamin and mineral products reported in national survey data, including iodine and selenium in adult products, and overages above 10 percent for folic acid in the vast majority of adult and children’s products.
The reason is mundane. Manufacturers add extra during production to cover expected losses over shelf life, and the amount of extra is not standardized. The practice is legal and routine, and it means the panel is a claim about what was intended to go in, not a measurement of what is in the bottle you are holding. Whether anyone independently checked the contents is a separate question with its own answer, and article 326 covers the third-party verification marks that attempt it.
One guide from 2005, two bills in Congress
Two timing facts round out the picture, and both get reported wrongly.
The first: FDA’s Dietary Supplement Labeling Guide is the main plain-language federal explanation of how a Supplement Facts panel works. Its issue date is April 2005, its content is listed as current as of 1 April 2005, it is marked as containing nonbinding recommendations, and FDA states in it that its guidance documents do not establish legally enforceable responsibilities. It predates the 2016 rule that revised every Daily Value it explains. The regulation and the guidance are twenty-one years apart, and only one of them binds anyone.
The second: there is no FDA rule requiring companies to tell FDA which supplement products they sell, and FDA cannot create one on its own. The proposal exists as legislation. Two bills in the 119th Congress carry the same name, the Dietary Supplement Listing Act of 2026. S. 3677 was introduced by Senator Richard J. Durbin of Illinois on 15 January 2026 and referred to the Committee on Health, Education, Labor, and Pensions. H.R. 8370 was introduced by Representative Maxine Dexter of Oregon on 20 April 2026 and referred to the Committee on Energy and Commerce. As of late October 2026, the Government Publishing Office carries only the introduced version of each.
A bill sitting in committee changes nothing on any label. If you read that FDA has proposed mandatory product listing, the word is wrong, and the difference matters because it tells you who holds the power to change this.
Five lines to read, in this order
Everything above collapses into a short reading order, which is worth more than memorizing regulation numbers.
- The serving size line. Every amount and percentage below it is per that serving, and the manufacturer set it.
- Servings per container, against the serving size. This tells you how long the bottle lasts at the recommended amount, which is often not thirty days.
- Any line carrying an asterisk. The amount is disclosed; the reference value does not exist. Treat the percentage column as empty there, not as reassuring.
- The blend name, if there is one. You have the order and one total weight. You do not have the proportions.
- The other-ingredients list below the panel. Binders, fillers, colors, sweeteners and flavors live there, in descending order by weight.
The panel answers two questions well: what the manufacturer says is in the product, and how much they suggest taking at a time. It does not tell you whether that amount matches what was studied, which chemical form you are getting, whether anyone verified the contents, or where the amount sits against a published upper limit. Those are four separate questions: article 326 covers whether the dose matches what was studied and who verified the contents, article 325 covers chemical form, and article 323 covers the upper limit.
One question the panel will never answer is whether anything in it interacts with a medicine you already take. That is not a label problem and it is not a general worry about supplement safety. It is a specific, answerable request: bring the bottle and your prescription list to a pharmacist and ask them to read the two side by side. Article 677 covers what those interactions look like.
Sources
- Legal Information Institute, Cornell Law School. “21 CFR § 101.36 – Nutrition labeling of dietary supplements.” law.cornell.edu
- Electronic Code of Federal Regulations (eCFR). “21 CFR § 101.4 Food; designation of ingredients.” ecfr.gov
- Electronic Code of Federal Regulations (eCFR). “21 CFR § 101.12 Reference amounts customarily consumed per eating occasion.” ecfr.gov
- Electronic Code of Federal Regulations (eCFR). “21 CFR § 101.93 Certain types of statements for dietary supplements.” ecfr.gov
- Electronic Code of Federal Regulations (eCFR). “21 CFR § 101.9 Nutrition labeling of food, including the Reference Daily Intake table at (c)(8)(iv).” ecfr.gov
- U.S. Food and Drug Administration. “Daily Value on the Nutrition and Supplement Facts Labels.” fda.gov
- U.S. Food and Drug Administration. “Dietary Supplement Labeling Guide (April 2005).” fda.gov
- U.S. Food and Drug Administration. “Dietary Supplement Labeling Guide: Chapter IV, Nutrition Labeling.” fda.gov
- U.S. Food and Drug Administration. “FDA Extends Nutrition Facts Label Compliance Dates.” fda.gov
- U.S. Food and Drug Administration. “Questions and Answers on Dietary Supplements.” fda.gov
- Office of the Federal Register. “Food Labeling: Revision of the Nutrition and Supplement Facts Labels, final rule, document 2016-11867, published 27 May 2016.” federalregister.gov
- U.S. Government Publishing Office (govinfo). “S. 3677, Dietary Supplement Listing Act of 2026, 119th Congress, introduced version.” govinfo.gov
- U.S. Government Publishing Office (govinfo). “H.R. 8370, Dietary Supplement Listing Act of 2026, 119th Congress, introduced version.” govinfo.gov
- NIH Office of Dietary Supplements. “Magnesium: Fact Sheet for Health Professionals.” ods.od.nih.gov
- NIH Office of Dietary Supplements. “Iron: Fact Sheet for Health Professionals.” ods.od.nih.gov
- PubMed. “Andrews KW, Roseland JM, Gusev PA, et al. "Analytical ingredient content and variability of adult multivitamin/mineral products: national estimates for the Dietary Supplement Ingredient Database." The American Journal of Clinical Nutrition, 2016. DOI: 10.3945/ajcn.116.134544.” doi.org
- PubMed. “Andrews KW, Gusev PA, McNeal M, et al. "Dietary Supplement Ingredient Database (DSID) and the Application of Analytically Based Estimates of Ingredient Amount to Intake Calculations." The Journal of Nutrition, 2018. DOI: 10.1093/jn/nxy092.” doi.org