Date of Award

January 2016

Document Type


Degree Name

Master of Public Health (MPH)


School of Public Health

First Advisor

Nicole C. Deziel


In the past two decades, thyroid cancer incidence has increased at a rate faster than any other malignancy. Environmental contamination has been a suspected risk factor in this trend. One potential source of environmental exposures is pollution from industrial manufacturing facilities. We investigated whether proximity to industrial facilities was associated with an increased risk of thyroid cancer in an exploratory analysis within an existing population-based case-control study in Connecticut using a novel data resource.

Complete residential histories of 408 thyroid cancer cases and 470 controls were collected and geocoded. Manufacturing facility addresses were gathered from a novel source of publicly available data, the Connecticut Point Source Inventory, and categorized by 2-digit Standard Industrial Classification (SIC) code from the years 1990-2009. Binary proximity exposure metrics were created and were defined as ever having lived within 5 km or 2 km of a facility in any manufacturing sector. Additional facility-specific binary markers were created that defined if an individual had ever lived within 5 km or 2 km of each specific SIC explored in this study (20-38).

Additionally, for each participant, a cumulative inverse distance-weighted (IDW) metric was calculated for all facilities within 5 km of all residences lived in between 1990 and 2009. SIC-specific IDW metrics were also created. Odds ratios (OR) and confidence intervals (CI) detailing the relationship between each proximity metric and thyroid cancer were calculated using logistic regression, adjusting for potential confounders. Cumulative IDW was evaluated as a continuous variable, a natural-log-transformed variable, and as a categorical variable.

In both unadjusted and adjusted analyses, nonsignificant elevated risk estimates were observed among individuals who lived within either 5 km (adjusted OR = 1.15, 95% CI: 0.86 – 1.54) or 2 km of chemical facilities (SIC 28, adjusted OR = 1.20, 95% CI: 0.84 – 1.73). After adjustment, ever having lived within 5 km of lumber and wood products facilities (SIC 24, OR = 0.64, 95% CI: 0.43 – 0.94) or within 2 km of transportation equipment facilities (SIC 37, OR = 0.54, 95% CI: 0.31 – 0.96) was associated with decreased risk of thyroid cancer. All other associations were null. Analyses of the cumulative IDW exposure data revealed similar results.

The results of this exploratory analysis do not appear to support a link between residential proximity to manufacturing facilities and increased risk of thyroid cancer. However, this study was limited by crude proximity metrics and imprecise recall of residential histories, which potentially led to exposure misclassification. Future studies could refine exposure metrics by incorporating emissions and meteorological data or by refining the accuracy of the residential addresses.


This thesis is restricted to Yale network users only. It will be made publicly available on 06/07/2018