About the database

What it is

A supplement label is the most exact public statement of what is in the bottle: each ingredient, how much of it, in what form, and the share of a day's value it gives. The database keeps that statement word for word, one record per label, so it can be searched and compared across brands and over the years. A label stays in the database after the product leaves the shelves, marked off the market, so older products can still be looked up.

A label is listed only when we hold the label itself: the NIH's photo of the package for its originals, or for a label we read, its picture or the maker's own page. Records read only from a store listing or a web search wait until their label is found; the search can show them with one click.

Who uses it

What each label holds

• The Supplement Facts panel: every ingredient with its amount, unit and % Daily Value, in the order printed.
• Proprietary blends with their total weight and the ingredients inside them, in order.
• The form of each ingredient, such as magnesium as magnesium glycinate.
• Serving size, servings per container, net contents and other ingredients.
• The barcode, the supplement form (capsule, powder, gummy and so on) and the product type.
• Directions, warnings and storage, as printed.
• A photo of the label, where one was taken.
• Whether the product is on the market, and the date the label entered the database and where it came from.

History

The NIH's Office of Dietary Supplements opened the database to the public in June 2013, built with the National Library of Medicine and with input from the USDA, the CDC, the FDA and the Department of Defense. Labels were entered from photographs of the packages. In 2021 it was given a new design and a free public API.

The last label was entered on 25 September 2025. By then the database held 214,780 labels. On 19 October 2026 Reverse Age Lab began updating the database with labels from the date the NIH stopped.

The move to Reverse Age Lab

Every NIH label keeps its original NIH number, so existing links and research citations still find it here.

The search works the way the DSLD's did: one box for a product, brand, ingredient or barcode, then filters for the market status, form, type, brand, ingredient and date entered. The look and the name on the page are ours.

How we keep it current

• 1. New products, found by their barcode. Every day the GS1 barcode registry, where makers register each product's barcode (UPC), is searched for supplement barcodes we have not seen; each one becomes a label waiting to be read. Being set up
• 2. The label, read from the maker. A label is read from the maker's own website first, then from the stores that sell it; a barcode already in the NIH's records is matched to that original instead of being added twice. Live
• 3. Added every night. The labels from our own product research and from visitors' requests (when a search finds nothing, you can tell us the product) go into the database overnight, each with the date it was read and where. Live
• Corrections. A label that is wrong is corrected from the maker's own page or the bottle. Tell us.

One difference from the NIH: they photographed each package, and our new labels are read from the maker's pages and store listings. Each one says where it was read and when, and the bottle you buy is the final word, because makers change formulas.