Date of Award


Document Type

Open Access Thesis

Degree Name

Master of Public Health (MPH)


School of Public Health

First Advisor

Linda Niccolai


Introduction: Incidence of high-grade cervical lesions (HGCL) caused by human papillomavirus (HPV) has declined in the U.S following the introduction of the HPV vaccine in 2006. However, disparities in HPV cervical infection and subsequent sequelae by race, ethnicity and income continue to persist. The purpose of this analysis was to identify spatial areas with significantly elevated HGCL burden in Connecticut, and determine socioeconomic characteristics associated with high incidence clusters.

Methods: Data from statewide surveillance in Connecticut for cervical intraepithelial neoplasia grades 2, 2/3, 3 and adenocarcinoma in situ (CIN 2+) from 2008-2016 were used for this analysis. Spatial analyses were performed using SaTScan v9.6 to identify significant clusters of HGCLs by census tract among women age 21-39 years across aggregated year groups from 2008–2010, 2011–2013, and 2014–2016. Four separate mixed effects models with varying sociodemographic covariates were constructed to assess the fit of the model to the number of HGCL cases per census tract. The likelihood of the predicted incidence for each model was used to calculate the Deviance Information Criterion (DIC), which balances goodness of fit and model complexity.

Results: From 2008–2016, incidence of HGCLs declined, particularly in women aged 20–29. Spatial analyses identified four significant clusters of HGCLs over time. These clusters varied in time, number of census tracts, as well as racial, ethnic and economic composition. The most recent cluster of HGCL, located in south central Connecticut among women aged 30–39, displayed significantly higher proportions Black, Hispanic and below poverty populations compared to the rest of the state. The proportion of Hispanic individuals per census tract was the most significant predictor of the number of HGCL cases.

Conclusions: The decline in HGCL incidence overall and among young women may suggest HPV vaccine impact in CT. The differences in location and socioeconomic composition of clusters suggest there are differences in incidence of HGCLs between neighborhoods of varying socioeconomic statuses. The relevant nature of ethnicity in predicting HGCL incidence may be indicative of a cultural disparity in the health coverage of Hispanic women in CT. The disparities in racial, ethnic and income characteristics in census tracts with high incidence of HGCLS indicate a need for continued surveillance and targeted interventions.


This is an Open Access Thesis.

Open Access

This Article is Open Access