Date of Award

January 2015

Document Type

Open Access Thesis

Degree Name

Master of Public Health (MPH)


School of Public Health

First Advisor

James L. Hadler


Introduction: Shiga toxin-producing Escherichia coli (STEC) are important foodborne pathogens, causing approximately 30 deaths a year in the US. Case-control studies to monitor risk factors are challenging; finding representative controls can be labor-intensive. The FoodNet Population Survey (FNPS) is conducted periodically and asks about foodborne disease risk factors.

Objectives: To determine leading risk factors for STEC in Connecticut using the 2006 FNPS participant responses as controls.

Methods: Cases were reported STEC cases in Connecticut 2000-2009 who were interviewed for risk factors following onset. Controls were respondents to the Connecticut portion of the 2006 FNPS. Comparable questions for exposure to dietary, travel and recreational risk factors in the 7 days before illness onset (cases) or interview (controls) were included. FNPS questions that were non-identical to case questions were examined when similar composite variables could be created.

Results: Data from 559 cases and 1,801 controls on 14 variables were included. Statistically significant risk factors, stratified by age, included `ground beef' for the 6-17 year and ≥18 age groups (OR 2.24, 95% C.I 1.47-3.43; OR 1.8, 95% C.I. 1.35-2.43, respectively), `pink burger patties' in the 6-17 year group (OR 2.07, 95% C.I. 1.01-4.25), and for all age groups, `traveling outside of the U.S.' (age-adjusted OR 6.66, 95%C.I. 3.36-13.19) and `visiting a petting zoo' (age-adjusted OR 4.06; 95% C.I. 3.28-7.92). Among similar but not identical variables, poultry was a risk factor for adults while consumption of lettuce, spinach and sprouts were risk-reducing factors in various age categories.

Conclusions: Several known STEC risk factors (ground beef, pink burger, travel, petting zoo) were confirmed, but several other findings are suspect. FNPS participants' responses show potential for use as controls but FNPS questions need to be matched to those asked cases.


This is an Open Access Thesis.