Reduced-Item Food Audits Based on the Nutrition Environment Measures Surveys

Area of Science:

• Public Health and Nutrition Science
• Environmental Health
• Health Policy

Background:

• The community food environment significantly influences dietary choices and contributes to obesity.
• Assessing food environments through store and restaurant audits is crucial but often time-consuming and expensive.
• There is a need for brief, valid, and reliable measurement tools for food environment assessments.

Purpose of the Study:

• To develop and validate reduced-item food environment audit tools for both stores and restaurants.
• To create shorter versions of the Nutrition Environment Measures Surveys for Stores (NEMS-S) and Restaurants (NEMS-R).
• To ensure these brief tools maintain the quality of measurement compared to the full surveys.

Main Methods:

• Utilized data mining techniques, including correlation-based feature selection and linear regression, on NEMS-S and NEMS-R data from 820 stores and 1,795 restaurants.
• Identified key survey items highly correlated with total scores to create reduced-item audit tools.
• Validated the reduced-item tools against full NEMS surveys, using regression coefficients as weights for score comparability.

Main Results:

• Developed reduced-item tools comprising 5-10 items for stores and 13-19 items for restaurants, representing 10% (NEMS-S) and 25% (NEMS-R) of the full surveys.
• Found no significant differences in median scores between reduced-item and full NEMS surveys across various food outlet types.
• Achieved a 25%-50% reduction in median in-store audit time.

Conclusions:

• Reduced-item audit tools effectively decrease the burden and complexity of assessing the retail food environment.
• These brief tools offer a practical solution for large-scale or repeated food environment assessments.
• Measurement quality is maintained, ensuring reliable data for public health research and interventions.