Date of Award




Document Type


Degree Name

Doctor of Philosophy (PhD)


Department of Epidemiology and Biostatistics



Content Description

1 online resource (viii, 189 pages) : illustrations (some color), color maps.

Dissertation/Thesis Chair

Allison Appleton

Committee Members

Joyce Barlin, Margaret Gates Kuliszewski, Melissa Tracy


Cervical cancer, Trachelectomy, Cervix uteri, Behavior disorders in children

Subject Categories

Epidemiology | Public Health


BackgroundRadical trachelectomy (RT) is a surgical procedure that offers a substantial benefit to women diagnosed with invasive cervical cancer: fertility preservation. However, as a newer surgical option, there are still many unknowns about the characteristics and outcomes of women who undergo this procedure after a cervical cancer diagnosis. Initial research in this area indicates that some disparities exist in who has the surgery versus the standard radical hysterectomy procedure; however, these are not fully understood and characterized. Furthermore, little is known about how women plan their families and how their high-risk pregnancies are managed in the absence of a standard of care following RT. Furthermore, women who undergo RT tend to deliver preterm, which in turn carries health risks for the infants including behavioral problems in later childhood. Thus, the impacts of RT are likely generational. Methods to mitigate the long-term risks associated with being born late preterm have not been thoroughly explored. Harsh discipline styles and low maternal involvement have been associated with behavior problems in term children and, as such, interventions designed to reduce these factors might mitigate behavioral health risks for children born late preterm. Objectives Three projects aimed to characterize the risk factors associated with not having RT, how pregnancies after RT are managed, and the generational effects of RT on children’s health. Specially, we addressed how behavior problems in childhood associated with late prematurity may be mitigated by maternal discipline style and involvement. Methods Project 1: The social determinants and descriptive epidemiology of access to RT were examined using the Surveillance, Epidemiology, and End Results (SEER) program dataset. A multinomial multilevel logistic regression model was developed to examine the association between individual clinical and demographic characteristics (age, race/ethnicity, marital status, year at diagnosis, and cancer stage, histology, location, and grade) and county-level characteristics (income, income inequality, poverty, education, percent uninsured, percent foreign-born, residential segregation, rurality, and access to primary care physicians), and surgical treatments for invasive cervical cancer including conization, RT, and radical hysterectomy. Project 2: A novel self-report survey was developed to ascertain information on family planning and pregnancy management following RT. The survey was fielded to women who have had RT and participate in the ‘TTL - Trachelectomy Group’ on Facebook. In addition to developing and piloting the survey, the descriptive epidemiology of the sociodemographic and pregnancy management characteristics of these women were examined and compared to SEER to determine generalizability. Project 3: The Avon Longitudinal Study of Parents and Children (ALSPAC), which includes regular follow-up among children born in Avon County, England since 1990, was used to examine trajectories of the maternal discipline styles (including harshness, cuddles, avoidance, and abandonment in response to child tantrums) and maternal involvement throughout early childhood. Associations between the trajectories and externalizing behavior problems were then examined using a logistic regression model to assess the potential for maternal behaviors to jointly mitigate the risk of behavior problems in childhood. Prematurity was considered as a potential effect modifier. Results Project 1: Despite the use of RT increasing between 2004 and 2016, Black and Hispanic women had lower odds of RT than White women and this disparity did not improve over time for Black women. Moreover, women who lived in counties having higher poverty, higher residential segregation, and lower percent foreign-born, or alternatively, higher rurality, had lower odds of having RT compared to radical hysterectomy. Clinical factors and higher county-level poverty were predictive of women having simple hysterectomy rather than radical hysterectomy. Project 2: A new survey to elicit the comprehensive fertility and pregnancy experience after RT was developed and fielded among women participating in a peer support group on Facebook. N=70 women participated from countries worldwide. Most women surveyed self-identified as White, non-Hispanic, were highly educated, did not have difficulty meeting their monthly financial obligations, and lived within an hour of a medical center capable of performing RT. Women who had conceived after RT were frequently prescribed vaginal progesterone and monitored frequently via ultrasound, and received advice from their medical providers to avoid physical activity, vaginal sexual intercourse, swimming, and taking baths. These behavior restrictions were often stricter in Europe than in the US and Canada and c-sections were frequently scheduled early (before or at 37 weeks gestation) in all countries. Project 3: Seven patterns of maternal involvement and discipline styles for children 18 months to 5 years old were identified. Odds of externalizing behavior problems at age 7 were higher among children exposed to the lowest levels of maternal involvement and early exposure to harsh discipline, and among children having experienced early exposure to discipline styles that incorporated harshness and abandonment compared to a neutral profile of maternal involvement and discipline style. On the other hand, never utilizing abandonment in response to tantrums reduced the odds of externalizing behavior problems compared to the neutral parenting style. Gestation at delivery did not modify the associations between trajectories of maternal involvement and discipline style and externalizing behavior problems. Conclusion Demographic rather than solely clinical factors predicted which women receive RT compared to radical hysterectomy. Moreover, county-level characteristics were associated with RT and simple hysterectomy use, which may offer an avenue for more efficient and efficacious intervention. With further research, these results have the potential to reduce disparities in access to RT for fertility preservation. Consistent with this, women who participated in a peer support group on Facebook and completed a survey after having undergone RT had characteristics favorable to having access to a state-of-the-art surgery such as RT including being White, non-Hispanic, not struggling financially, and living within close proximity to a center that can perform RT. However, pregnancies in these survey respondents were frequently managed closely by providers inexperienced in managing pregnancies after RT, and women frequently received inconsistent guidance to restrict several behaviors during pregnancy that are untested in women having undergone RT and may not be efficacious in prolonging gestation. Moreover, the risk of late preterm delivery is higher after RT and c-sections are frequently scheduled early, which increases the risk of neurodevelopmental morbidities. Maximizing maternal involvement and promoting gentle, supportive discipline styles through toddlerhood and early childhood may offer parents an avenue by which to reduce the generational impacts of RT that come as a result of early delivery.

Included in

Epidemiology Commons