Commercially available business (CAB) datasets for food environments have been investigated for error in large urban contexts and some rural areas, but there is a relative dearth of literature that reports error across regions of variable rurality. The objective of the current study was to assess the validity of a CAB dataset using a government dataset at the provincial scale.

A ground-truthed dataset provided by the government of Newfoundland and Labrador (NL) was used to assess a popular commercial dataset. Concordance, sensitivity, positive-predictive value (PPV) and geocoding errors were calculated. Measures were stratified by store types and rurality to investigate any association between these variables and database accuracy.

NL, Canada.

The current analysis used store-level (ecological) data.

Of 1125 stores, there were 380 stores that existed in both datasets and were considered true-positive stores. The mean positional error between a ground-truthed and test point was 17·72 km. When compared with the provincial dataset of businesses, grocery stores had the greatest agreement, sensitivity = 0·64, PPV = 0·60 and concordance = 0·45. Gas stations had the least agreement, sensitivity = 0·26, PPV = 0·32 and concordance = 0·17. Only 4 % of commercial data points in rural areas matched every criterion examined.

The commercial dataset exhibits a low level of agreement with the ground-truthed provincial data. Particularly retailers in rural areas or belonging to the gas station category suffered from misclassification and/or geocoding errors. Taken together, the commercial dataset is differentially representative of the ground-truthed reality based on store-type and rurality/urbanity.

To assess the accuracy of government inspection records, relative to ground observation, for identifying businesses offering foods/drinks.

Agreement between city and state inspection records v. ground observations at two levels: businesses and street segments. Agreement could be ‘strict’ (by business name, e.g. ‘Rizzo’s’) or ‘lenient’ (by business type, e.g. ‘pizzeria’); using sensitivity and positive predictive value (PPV) for businesses and using sensitivity, PPV, specificity and negative predictive value (NPV) for street segments.

The Bronx and the Upper East Side (UES), New York City, USA.

All food/drink-offering businesses on sampled street segments (n 154 in the Bronx, n 51 in the UES).

By ‘strict’ criteria, sensitivity and PPV of government records for food/drink-offering businesses were 0·37 and 0·57 in the Bronx; 0·58 and 0·60 in the UES. ‘Lenient’ values were 0·40 and 0·62 in the Bronx; 0·60 and 0·62 in the UES. Sensitivity, PPV, specificity and NPV of government records for street segments having food/drink-offering businesses were 0·66, 0·73, 0·84 and 0·79 in the Bronx; 0·79, 0·92, 0·67, and 0·40 in the UES. In both areas, agreement varied by business category: restaurants; ‘food stores’; and government-recognized other storefront businesses (‘gov. OSB’, i.e. dollar stores, gas stations, pharmacies). Additional business categories – ‘other OSB’ (barbers, laundromats, newsstands, etc.) and street vendors – were absent from government records; together, they represented 28·4 % of all food/drink-offering businesses in the Bronx, 22·2 % in the UES (‘other OSB’ and street vendors were sources of both healthful and less-healthful foods/drinks in both areas).

Government records frequently miss or misrepresent businesses offering foods/drinks, suggesting caveats for food-environment assessments using such records.