(P06-005-25) A New Ingredient Focused Food Database: Examining Common Artificial Ingredients and Their Combinations in Processed Foods Using WISEdata™

Objectives: Demonstrate the utility of WISEdata™, a new ingredient focused food database, in the assessment of commercial food products and their ingredients. Determine if any of 5 specific ingredient categories (or combinations) predicts level of processing.

Methods: WISEdata™ is a database of 650,837 commercial foods ( ~2.9 million unique UPCs) and 5,468 ingredients ( ~ 32,000 ingredient name variations).  Each ingredient has hundreds of unique attributes (nutritional - “fat content”, functional - “flavor”, characteristics - “vegan”, etc), with endless combinations.  The following five ingredient categories (with combinations) were used to filter foods, and determine the proportion deemed highly processed:  Artificial Colors, Artificial Flavors, Artificial Preservatives, Emulsifiers, Seed Oils.  For each ingredient category, and for 11 different combinations, the proportion of foods using the ingredient / ingredient combination and the processed level were obtained.

Results: Of the 650,837 foods examined, Seed Oils were the most common ingredient appearing in 34% of foods, of which 52.3% were highly processed.  In descending order, the usage for Emulsifiers, Artificial Colors, Preservatives, and Artificial Flavors was 29.0%, 17.1%, 16.0% and 13.1% of foods, respectively, of which 61.0%, 79.4%, 76.8% and 78.7% were highly processed.  While very few foods (1.5%) used at least one ingredient from all 5 categories, 100% of those foods were highly processed.  With one exception, foods that used ingredients from at least 2 of the 5 categories had proportions of highly processed foods ranging from 83% (Seed Oils + Preservatives) to 94.5% (Artificial Flavors + Preservatives).  The exception was Seed Oil + Emulsifier, where 70.4% were highly processed.  These combinations do not represent an exhaustive analysis but rather serve to demonstrate the depth, breadth and analytical possibilities provided by the WISEdata™ food database.

Conclusions: Food data at the ingredient level is essential to understand our diet, and the ability to probe those data is critical.  When combined with ingredient attributes and functions, the WISEdata™ food database achieves this at a depth and breadth previously unavailable, informing future research, re-evaluation of past research and developing policy.

Funding Sources: WISEcode, LLC funded all work.