Secondary Logo

Journal Logo

Gluten Contamination of Restaurant Food: Analysis of Crowd-Sourced Data Lawlor Resident Award


Lerner, Benjamin A. MD1; Vo, Lynn Phan BA2; Yates, Shireen MBA3; Rundle, Andrew G. DrPH2; Green, Peter H. R. MD1; Lebwohl, Benjamin MD, MS1

American Journal of Gastroenterology: October 2018 - Volume 113 - Issue - p S658
doi: 10.1038/ajg.2018.302

1. Columbia University Medical Center, New York, NY;

2. Columbia University Mailman School of Public Health, New York, NY;

3. Nima Labs, San Francisco, CA

Introduction: Adherence to a gluten-free (GF) diet is the mainstay of therapy for celiac disease (CD). Potential gluten exposure when eating in restaurants can be a hazard and source of anxiety for CD patients. Advice about avoiding cross-contaminated GF restaurant food has relied on clinical expertise with a relative dearth of data. The advent of portable gluten detection devices allows for crowd-sourced evaluation of restaurant food for GF safety. We aimed to identify factors associated with presence of gluten in restaurant food.

Methods: We analyzed data from a portable gluten detection device (Nima), collected across the USA during an 18-month period by users (n = 804) who opted to share results of their point-of-care tests. Each test included a date/time, food item, restaurant name and address, presence/absence of GF label, and presence/absence of gluten. Data was sorted by region, time of day, median household income in the restaurant's vicinity (extrapolated from zip code and 2016 American Community Survey), restaurant genre, and food item. We used chi-square test for bivariate analysis and multiple logistic regression for multivariate analysis to identify predictors of the presence of gluten in restaurant food.

Results: There were 5624 tests in the examined period, 3327 (59%) during dinner hours, 2576 (46%) in Western states, and 3449 (63%) in the highest income quartile of zip codes. 4732 (84%) of tested foods were labeled GF; 32% of GF labeled foods contained gluten. Presence of gluten in GF labeled foods differed by meal, with 27.2% at breakfast and 34.0% at dinner detecting gluten (p = 0.001). On multivariate analysis adjusting for region, income, and time of day, GF labeled food was less likely to contain gluten in the West than in the Northeast (OR 0.80, 95% CI 0.67 to 0.95, p = 0.01). Compared to other foods, GF labeled pizza and pasta were most likely to be contaminated, with 53.2% of pizza (OR 2.5, 95% CI 2.0 to 3.2, p = < 0.0001) and 50.8% of pasta (OR 2.1, 95% CI 1.5 to 3.1, p = < 0.0001) detecting gluten.

Conclusion: One-third of restaurant foods labeled GF contained at least 20 ppm of gluten. Higher rates of gluten contamination during dinner may reflect cumulative contamination of GF ingredients/equipment over the course of the day. Lower rates in the West may relate to greater interest in GF diets in that region. Our findings of higher rates of cross-contamination in pizza and pasta provide practical data when providing dining strategies for patients with CD.

© The American College of Gastroenterology 2018. All Rights Reserved.