NYC Restaurant Inspection Database — A Complete User Guide

What Is Dataset 43nn-pn8j?

The dataset numbered 43nn-pn8j on NYC Open Data is the official public record of every food service establishment inspection conducted by the New York City Department of Health and Mental Hygiene (DOHMH). It includes inspections going back to 2010 and is updated on business days. As of 2026, the dataset covers more than 27,000 active food establishments across all five boroughs and contains millions of individual finding records.

Understanding how the data is structured — and where the common sources of confusion lie — makes the difference between a useful lookup and a misleading one.

How the Data Is Structured

The dataset is organized at the finding level, not the restaurant level. This means:

• One inspection visit can produce 0, 1, or many rows depending on what the inspector observed
• A restaurant with five findings on a single date will appear five times in the data for that date
• A perfect inspection (no findings) may appear as a single row with a blank violation description

The most important fields to understand are:

• CAMIS — a unique identifier for each establishment (useful for tracking an address that has changed names)
• DBA — the public-facing business name
• BORO — borough (Brooklyn, Manhattan, Queens, Bronx, Staten Island)
• INSPECTION_DATE — when the inspection took place
• INSPECTION_TYPE — Initial Inspection, Compliance Inspection, Pre-permit, etc.
• VIOLATION_CODE and VIOLATION_DESCRIPTION — the specific finding
• CRITICAL_FLAG — whether the finding is Critical, Not Critical, or Not Applicable
• SCORE — the total score for that inspection date (same value repeated on every row for that date)
• GRADE — A, B, C, Z (Grade Pending), or blank (for inspection types that do not result in a grade)

Step-by-Step: Filtering by Borough

On the dataset page at data.cityofnewyork.us, click the Filter button in the top toolbar. Select Add a New Filter Condition:

1. Column: BORO
2. Operator: is
3. Value: Brooklyn (note: capitalize exactly as shown; the dataset uses title case)

This narrows the view to Brooklyn establishments only — approximately 6,000 to 8,000 active food service locations depending on the year.

Filtering by Zip Code

To focus on a specific neighborhood, add a second filter condition:

1. Column: ZIPCODE
2. Operator: is
3. Value: enter the five-digit zip code (e.g., 11201 for Brooklyn Heights, 11238 for Prospect Heights, 11211 for Williamsburg)

You can add multiple zip code conditions using "OR" logic to cover several neighborhoods at once.

Filtering by Cuisine Type

The CUISINE_DESCRIPTION field contains a standardized list of cuisine categories assigned by DOHMH. Common values include American, Chinese, Japanese, Mexican, Pizza, Bakery, Café/Coffee/Tea, and many more. Filtering by cuisine lets you compare food safety records across similar establishments in your neighborhood.

Reading a Restaurant's Score History

To review a restaurant's inspection history:

1. Filter by DBA (the restaurant name) and optionally by BORO
2. Sort by INSPECTION_DATE descending to see the newest results first
3. Group rows by INSPECTION_DATE to understand which findings come from the same visit
4. Read the SCORE on any row for that date — it is the same for all rows sharing that inspection date
5. Check the INSPECTION_TYPE to distinguish Initial Inspections from Compliance Inspections

Initial vs. Compliance Inspections

DOHMH uses a two-step cycle when an establishment scores above 13 points. The first visit is the Initial Inspection. If the score is 14 or higher, DOHMH schedules a Compliance Inspection within 30 to 90 days. The grade posted in the restaurant window comes from the compliance visit score, not the initial visit. This means:

• A Grade A can appear even after an initial score in the 20s, if the establishment corrected the findings before the compliance visit
• Looking only at the most recent grade without checking inspection type can be misleading for recent visits

For a complete picture, look at both inspection types and note the scores on each.

Interpreting "Grade Pending"

A Grade Pending card (Grade "Z" in the dataset) means the establishment is in the period between inspections, during an adjudication process, or has recently received a score that is being reviewed. It does not necessarily mean the restaurant has failed — it means a final grade has not yet been assigned for the current cycle.

Downloading the Data for Deeper Analysis

The Open Data portal allows you to download filtered results as CSV or Excel for offline analysis. This is useful if you want to compare scores across multiple restaurants, build a neighborhood-level view, or track trends over time. The dataset is also accessible via the Socrata Open Data API (SODA) for programmatic access.

Frequently Asked Questions

Why does the same restaurant show multiple rows with the same inspection date?

Each row represents one finding. If an inspection resulted in four observations, there will be four rows with the same INSPECTION_DATE and SCORE. Add all findings together for a complete picture of that visit.

What does CAMIS stand for and why is it useful?

CAMIS is a unique establishment identifier assigned by DOHMH. It stays with the physical location even if the restaurant changes its name or ownership. Use it to track safety history at a specific address.

How often is the dataset updated?

The dataset is updated on New York City business days. Results from inspections typically appear within one to two business days of the visit.

Is there a simpler tool than the raw dataset?

Yes. The DOHMH consumer lookup at a816-restaurantinspection.nyc.gov provides a search interface with maps and summarized grades. The raw dataset offers more historical depth and custom filtering.