Creating 'obesogenic realities'; do our methodological choices make a difference when measuring the food environment?

Area of Science:

• Environmental Health
• Geographic Information Science
• Public Health Nutrition

Background:

• Geographical Information Systems (GIS) are increasingly used to measure obesogenic food environments.
• Methodological heterogeneity in GIS studies poses challenges for comparing results.
• Understanding the covariance between different food environment metrics is crucial.

Purpose of the Study:

• To compare density and proximity metrics of food environments.
• To assess the comparability of different GIS-based food environment measures.
• To identify potential areas for standardization in food environment research.

Main Methods:

• Calculated food outlet densities (per 1000 population, per km²) and proximity measures (street network, Euclidean distances) in North East England.
• Used administrative areas and population-weighted centroids with buffers.
• Compared metrics using Spearman's rank correlations.

Main Results:

• Density and proximity metrics showed high correlations (e.g., proximity r(s)=0.865, density r(s)=0.667-0.764).
• Densities per km² and per 1000 population were highly correlated (r(s)=0.831).
• Metrics based on administrative areas and buffered centroids showed weaker correlations (r(s)=0.299-0.658).

Conclusions:

• Density and proximity metrics for food environments are largely comparable, with some exceptions.
• Standardizing environmental metrics can enhance inter-study comparability and reduce methodological variation.
• Acknowledging researchers' influence on constructing 'food environment realities' is important.