About this guide & important disclaimer

Who is behind this

Dr. John O. Ike, MD, MPH, is a Stanford-trained, board-certified nephrologist in Houston, Texas, with more than three decades in kidney medicine. He has spent his entire career working with his chronic kidney disease patients to protect their kidney function and keep them away from dialysis.

Why we built this

We built this guide because people with chronic kidney disease are often told to watch their phosphorus, yet food labels rarely show how much a product contains. Phosphate additives can also hide in processed foods, where they appear in the ingredient list only by name. This guide shows what ingredient lists reveal, so patients can ask better questions of their care team.

How we flag phosphate additives

• We search each product’s ingredient list for phosphate additive wording: phosphoric acid, and any “…phosphate” compound, including sodium, potassium, calcium, aluminum and ammonium phosphates, pyrophosphates, polyphosphates and metaphosphates.
• We ignore words that are not additives, such as phospholipids and phosphatidylcholine (found in lecithin), phosphatase and phospholipase enzymes, vitamin forms like riboflavin-5’-phosphate, and statements like “no phosphates.”
• Leavening phosphates (for example monocalcium phosphate or sodium acid pyrophosphate in baked goods and cereals) are included as additives, because they are still added phosphate.
• We do not measure how many milligrams of phosphorus a particular additive adds. Labels rarely give that number.

What the phosphorus numbers mean

• The phosphorus figures come from the “Phosphorus, P” value in the USDA database, which is copied from the Nutrition Facts panel when the manufacturer chose to print it. In the US, listing phosphorus is optional unless a claim is made or the product is fortified with it, so most products do not have a value. When there is none, we show “Not listed on label.” That is not the same as zero, and we never count it as zero. Where we can, we add a separate typical estimate (see “Typical estimates” below); it is never mixed with label values.
• Label values are rounded by the manufacturer, and a listed 0 mg means the label said 0, not that the food has no phosphorus. Manufacturers can calculate label values from ingredient tables, so a number may not fully capture the phosphorus that phosphate additives contribute. For that reason the phosphate additive flag stays our primary signal, and the milligrams are a rough label-based guide, not a measurement.
• Milligrams per serving use the serving size printed on the label, which differs between products. The per 100 g or mL figure is easier to compare. Sodium is shown only as extra information and is not part of how we flag or rank products.
• How “Best picks” are ordered: products with no phosphate additive found come first; among them, the lowest listed phosphorus comes first; products with no listed value follow, and each page says how many items lack one. Products that lack a label value are not ranked by phosphorus, and typical estimates never change the order. This is not advice on what to eat.

Typical estimates (not from the label)

• What they are. When a product’s label does not list phosphorus, we may show a typical range for that kind of food (for example “almond milk” or “frozen pasta meals”). It comes from USDA FoodData Central reference foods (SR Legacy 2018 and Foundation Foods 2026; phosphorus in mg per 100 g), not from the product itself.
• How the range is built. We match each product to reference foods using simple, published rules (the product’s name and type). The main range is the middle half (25th to 75th percentile) of the matched reference foods, and the “wider spread” is the 10th to 90th percentile. Both are scaled to the serving size on the label. We show how many reference foods were used (n), and we show no estimate when fewer than 3 match.
• Label first. The order is always: label value, then estimate, then “Not listed on label.” Estimates are never used to sort, filter or rank products, and they are never shown in the label columns.
• Phosphate additives. Reference foods are generic, so a product with a phosphate additive can have more phosphorus than the typical range. Where products of the same type that do list phosphorus show this, we say “likely higher than typical”; where we could not test it, we say “may be higher”. For carbonated sodas only, the range for products with phosphoric acid is based on colas and pepper-type sodas, and the range for others on non-cola sodas.
• Checks. We compared each type’s range with products of that type that do print phosphorus, and we do not show estimates for types where the range clearly did not fit. We currently show no typical estimate for coffee creamers or deli meats: almost no products of those types print phosphorus, so we cannot test the ranges, and the closest type we could test (sausages) came out lower than the label values. Those products show “Not listed on label.”
• Limits. An estimate can be wrong. Recipes differ, reference foods are generic and some are old, and a few types have only 3 to 5 reference foods. Estimates are not medical advice. If you follow a phosphorus limit, ask your renal dietitian how to use them. This prototype has not yet been reviewed by a physician.

Known limits of this data

• The USDA database stores many past versions of each product. We keep the most recent one, but a large share of records date from 2017–2023, so some recipes have changed.
• Ingredient lists can be typed with errors or cut off, so a few products may be missed or flagged by mistake.
• Phosphorus and sodium per serving are calculated from the amount per 100 g or 100 mL and the serving size on the label. Serving sizes vary. Phosphorus is missing on the label for most products, and a few obviously impossible values in the database were set aside and treated as not listed.
• Category membership follows the USDA category. For example, “Sodas” also includes some juice drinks and sparkling waters.
• Product names are shown as they appear in the database and are in English.

Data source

USDA, Agricultural Research Service. FoodData Central, Branded Foods, release 2026-04-30. Public domain. Source: fdc.nal.usda.gov.

Contact & corrections

[Placeholder: a contact email for corrections and brand questions goes here, once approved.]