Methods & attribution

Every figure on this site is computed by code from public data. No numbers are estimated, modelled or entered by hand. Each release states the date of every source and the time it was computed, and ships with a CSV of every metric.

Sources

1. New York City restaurant inspections

Source: NYC Department of Health and Mental Hygiene (DOHMH), “DOHMH New York City Restaurant Inspection Results,” NYC Open Data, https://data.cityofnewyork.us/d/43nn-pn8j. Version used: dataset record date 2026-10-19, pulled 2026-10-20.

Changes we make: violation rows are grouped into one record per establishment and inspection date. We keep only the business ID, borough, ZIP code, cuisine, inspection type, action, score, grade and violation codes. Phone numbers are dropped. We then compute monthly aggregates. Per DOHMH, the dataset covers only establishments that are active at the time of the pull, so closed businesses drop out of the history over time.

2. Chicago food inspections

Source: City of Chicago, Department of Public Health, “Food Inspections,” Chicago Data Portal, https://data.cityofchicago.org/d/4ijn-s7e5, pulled 2026-10-20. Changes we make: we keep the facility type, risk, inspection type, result and the number of violations cited. We drop the violation narrative text and compute monthly aggregates.

3. FDA food recalls

Source: U.S. Food and Drug Administration, food enforcement reports via openFDA, open.fda.gov/apis/food/enforcement. Data last updated 2026-10-07, pulled 2026-10-20 without an API key, within openFDA’s published rate limits.

Red Plate Index is an independent project. It is not affiliated with, sponsored by or endorsed by the U.S. Food and Drug Administration, the NYC Department of Health and Mental Hygiene, the City of New York, the Chicago Department of Public Health or the City of Chicago. This site uses no agency logos or seals.

Definitions

• Month: the calendar month of the NYC inspection date, the Chicago inspection date or the FDA report date.
• NYC initial-cycle inspection: an inspection whose type is “Cycle Inspection / Initial Inspection.” Critical: at least one violation flagged “Critical” by DOHMH. Score: the DOHMH point score (lower is better; 0–13 is the A range). Closures: inspections with the action “Establishment Closed by DOHMH.”
• Chicago fail rate: inspections with result “Fail” divided by completed inspections. Completed inspections exclude Out of Business, No Entry, Not Ready and Business Not Located.
• FDA product record: one row in the enforcement report, usually one product or lot. Event: a distinct FDA event ID that groups related products.
• Breakdowns with small samples are hidden (NYC cuisines need n≥40; Chicago inspection types need n≥10).

Recall cause & category rules

The FDA does not publish a structured cause or food category. We assign one using the first keyword rule that matches the reason for recall (cause) or the product description (category). Unmatched records go to “Other.” These labels are approximate.

Allergens named: milk, egg, peanut, tree nut, wheat, soy, fish, crustacean shellfish, sesame, sulfites.

Caveats

• Inspection counts depend on how many inspections each city schedules. A change in the fail rate can reflect where inspectors went, not only how restaurants performed.
• Complaint-driven inspections fail more often by design, so the mix of inspection types affects the overall rate.
• FDA enforcement reports appear weeks after a recall starts, and records can be reclassified later. Earlier months may change slightly in later pulls; each release keeps the numbers as computed.
• We publish business-level and product-level data only. We never publish data about individuals.

Licence

Red Plate Index’s derived statistics (CSV and JSON) are released under CC BY 4.0. The source datasets keep their own terms.